AD-A13 431 •FLUSAFSAM-TR-85-31 41U FILE Cp. Y

SPATIAL ORIENTATION IN FLIGHT

~. -. ~,, -

Kent K. Gillingham, M.D., Ph.D.James W. Wolfe, Ph.D.

DTiC

E* L -!• AUG 1 2• W7"

Decem ber 1986 .- . -.

Final Report for Period January 1982 - January 1985

* 0

Approved for public release; distribution Is unlimited.

USAF SCHOOL OF AEROSPACE MEDICINEAerospace Medical Division (AFSC)Brooks Air Force Base, IX 78235-b301

UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAGE

REPORT DOCUMENTATION PAGE'aF REPORT SECURITY CLASSIFICA7 ION lb. HESTRICTIVE MARKINGSUncla ssified

HTRCt M'NS

2F. SECURITY CLASSIFICATION AUTHORITY 3. OISTRIBUTIONAVAILABILITY OF REPORT

Approved for public release; distribution is2b. OECLASSIFICATIONOOWNGRADING SCHEDULE unlimited,

4. PERFORMING ORGANIZATION REPORT NUMErEISI 5. MONITORING ORGANIZATION REPORT NUMBERIS)

USAFSAM-TR-85-316&. NAME OF PERFORMING ORGANIZATION b. OFFICE SYMBOL 7a 14AME OF MONITORING ORGANIZATION

USAF School of Aerospace f applicable,

Medicine USAFSAM/VNB6c. ADDRESS (City. State and ZIP Code) 7b ADDRESS (City, Stote and ZI:1 CodetAerospace Medical Division (AFSC)Brooks Air Force Rase,TX 78235-5301

Bi. NAME OF FUNDING/SPONSORING 8b. OFFICE SYMBOL 9 PROCUREMENT INSTRUMENT IDENTIFICATION NUMBERORGANIZATION USAF School of (tf applicable,

Aerospace Medicine USAFSAM/VNB

Re AnnRESS ,Citv. State and /IP Code) 10 SOURCE OF F UNDINJG NOS.

PROGRAM PROJECT TASK WORK UNITAercsi~aL-` Mediczl Divjs.*on (AFSC) IELEMENTNO4 NO0 NO. NO.Brooks ikir Force Base, TX 78235-5301 62202F 7930 14 57

1 1. TI TLE ,iciude Securiy Claosifca(ica. .

SPATIAL ORIENTATION IN FLIGHT

12. PERSONAL AIJTHOR(SI

Gillingham, Kent K.; Wolfe, James W.13s. TYPE OF REPORT I 13b. TIME COVERED 14. DATE OF REPORT 1Yr, M.o.. Day) 15. PAGE COUNT

Final report . ROMjan 1982 TO Jan1985l 1986, December 13416. SUPPLEMENTARY NOTATION

17. COSATI CODES 18. SUBJECT TERMS (Continue On reverse if necessary and waentify by block number)

FIELD GROUP SUB. GRA Spatial orientation Visual function06 05 spatial Disorientation vestibular function06 10 Motion sickness Pilot vertigo

,19. ABSTRACT XConlnue on ,e•'erse if necessary and identify by block numbetr,

Man's orientational mechanisms, and how those mechanisms fail in flight, are discussed indetail in this comprehensive review. Specific topics include: mechanics and associatedphysiologic nomenclature: visual orientation; vestibular function and information process-ing; other senses of motion and position; spatial disorientation, including causes, types,examples, statistics, and methods of preventing spatial disorientation mishaps; and thesignificance, etiology, and therapy of motion sickness. Forty-three figures are included,many illustrating vestibular anatomy and physiology, and others depicting the more commonvisual and vestibular illusions in flight. Sixty-five classic references and a recommended

reading list are also provided.

20, OISTRIBUTIONAVAILABILITY OF ABSTRACT 21. ABSTRACT SECURITY CLASSIFICATION

UNCLASSIFIED/UNLIMITED FO SAME AS RPT - OTIC USERS C Unclassified

22s. NAME OF RESPONSIBLE INDIVIDUAL 22b TELEPHO'c NI.I)MRER 22c OFFiCF SYMBOL,include A•rea Cadet

Kent K. GillinghamKen K.Gilingam(512) 536-3521 USAFSAM/VNB

DD FORM 1473, 83 APR EDITION OF I JAN 73 IS OBSOLETE. UNCLASSIFIEDi SECURITY CLASSIFICATION OF THIS PAGE

rIACKNOWLEDGMENTS

The authors are extremely grateful to the artists who

produced the figures for this report: David Schall, Major, aUSAF, MC, who gave us Figures 14-20, 23, 28, 29, 33, 36, 37, and

40; Mel Jordan, who provided Figures 10, 12, 27, 30, 32, 34, 35,and 43; and Sharon Tice, who drew Figures 6, 21, and 38. We arealso very appreciative of Dorothy Baskin's assistance in prepar-ing the manuscript. We owe the greatest debt of gratitude toRoy DeHart, Colonel, USAF, MC (Retired), Commander of the USAFSchool of Aerospace Medicine from 1980 to 1983, whose desire to

promulgate the educational material contained herein for thebenefit of all aeromedical professionals ultimately led to thepublication of this report.

I"

z .C ' s>o el Lo I"%:

- Ir),T I C

I.--,. -t - 1 -.

' e .. .. . .. ..... < '

ii , " . '.II".

CONTENTS

M MCoAiCS....................................................... 1

Linear Motion ... ................................ 2 'Angular Motion .... .. .. .. ....................... 4 .

Force, Inertia, and Momentuu...........................6.".6Force and Torque ...... .. ....................... 6 "

Mass and Rotational inertia .................... 6 :

Directions of Action and Reaction...................0Vehicular Motions ... ............ .............. 10 -.Physiologic Acceleration and Reaction Nomenclature. 11e'

%..-

VISUAL ORIENTATION......... .............................. 5 '

Anatomy of the Visual System ............................ 15 •General .5.. ...............................................Visual-vestibular Convergence....................... 16,

Visual Information Processing ............................ 17Focal Vision ..................................... 17Ambient Visionr..................................... 19Eye Movements .......................

VESTIBULAR FUNCTION.......... .. ............... 22 -

Vesibular Anatomy .................................... 22

End--organs...... . . . . . . .22

Neural Pathways............... . . .27

Vestibular Information Processing .......................... 29Vestibular Reflexes .................................... 29Voluntary Movement............................. 36Conscious Percepts..................................... 37Thresholds of Vestibular Perception .............. . 7..3Vestibular Suppression and Enhancement.............. 1739

OTHER SENSES OF viOTION AND POSITION........................... 1..

Nonvestibui V Proprioceptors ............................. 41

Muscle and Tendon Senses............................ 24Joint Sensation ....................................... 23

Cutaneous Exteroceptors .................................... 244echanoreceptors .................................... 2442

AuditorylOrientation ............................Con ci us Pe ce ts.. .. .. .. .. .. .. .. .. .. ... 3Threhold ofVestbulr Pecepton ............ ýVetiuarSupesin ndEhaceet ........ 3

SPATIAL DISORIENTATION .......................... ............. 46

Illusions in Flight ................... ......... 46Visual Illusions ..................................... 46Vestibular Illusions ............................... 59

Disorientation in Flight ................................ 81Definitions .. . . . . . . . . . . . . . . . .o . . 81 '

Types........................................81Taptes .ti .......................................... 81Statistics ................................. ........ 83 '

Dynamics of Spatial Orientation and Disorientation. 84Conditions Conducive to Disorientation .............. 88

Prevention of Disorientation Mishaps .................... 91Education and T'raining ....... ..... .. ...... . .... 93Inflight Procedures ........................ 96Cockpit Layout and Flight Instruments.............. 99

Other Sensory Phenomena ........................ 104-Flicker Vertigo .............................. 105Fascination ............................ ....... 105Breakoff ........................................... 106

MOTION SICKNESS ................ .............. 107

Definition, Description, and Significance ofMotion Sickness .................................... 107

Military Experience ......................... 107Civil Experience ............................. 109Space Sickness ..................................... 109

Etiology .. . . . . . . . .. . . . . . . . .. . . . . . . 1 0

Correlating Factors ................................ 110Unifying Theory ............. ............. 112 ".Teleology .......................................... 114

Prevention and Treatment .............................. 115

Physiologic Prevention ............... 115Physiologic Treatment .............................. 116Pharmacologic Prevention .......................... 116Pharmacologic Treatment ............................ 117Aeromedical Use of Antimotion-Sickness 6

Preparations ................................... 117

REFERENCES ................................................. 119 '

RECO ENDED ......................................... 125

V I

-°-. ,J

I•'% • • %." - • • .. . , • - ,.. •,- -.. . . . . . . .

List of Fiqures

FigureNo. Paagce

1. Axes of linear and argular aircraft motions ........................ 11

2. System for descrioing accelerations and inertial reactions in man.. 13

3. Gross anatomy of the inner ear .................................... 23 . .

4. The vestibular end-organs ......................................... 24 -•

5. Function of a vestibular hair cell ................................. 25

6. Morphologic polarization in vestibular neuroepithelia .............. 26

7. Major connections and projections of the vestibular system ......... 28

8. The cardinal principle of vestibular mechanics ..................... 30

9. Mechanism of action of a horizontal semicircular duct, andresulting reflex eye movement ..................................... 32

10. Ocular nystagmus--repeating compensatory and anticompensatory ..-

eye movements--resulting from vestibular stimulation ............... 33

11. Mechanism of action of an otolith organ ............................ 34 .

12. Ocular countertorsion, a vestibulo-ocular reflex of otolith-organorigin ............................................................. 35

13. Some of the nonvestibular proprioceptive and cutancousexteroceptive receptors subserving spatial orientation ............. 43 .

14. Effect of runway slope on pilot's image of runway during final .approach, and potential effect on ?-'proach slope anqle flown ....... 47

15. Effect of runvwav width on pilot's image of runway and potentialeffect on approach flown .......................................... 48

16. Potential effect o, slope of terrain under' the approach on approachslope flown ...................................................... 49

17. Potential effect of unfamiliar co;;mposition 0' dpproach terrainon approach slope flown ............................................ 50

18. Effect of loss of peripheral visual orientation tues on Perceptionof runway orientation during a black-hole approarch...............

19. A colitmon and particularly danaerous type of black-hole approach .... 53

20 . Visual autok inesis ................................................. b,-,-

vii

FigureNo. Page

21. Vection illusions .................................................. 57 %

22. A sloping cloud deck ........................................... 58

2. 1ispcrccption of the horizontal at night ......................... 60

24. Transfer characteristics of the semicircular duct system as afunction of sinusoidal stimulus frequency .......................... 62

25. Effcct of stimulus pattern on perception of angular velocity ....... 63

26. Pictorial representation of mechanical events occurring in asemicircular duct and resulting action potentials in associatedarnpullary nerve during somatogyral illusions ...................... 64

27. The graveyard spin ................................................. 65

Z . ne graveyard spiral ............................................... 67

29. Mechanism of the Coriolis illusion ................................. 70

30. A somatogravic illusion occurring on takeoff ....................... 71

31. Flight recorder data from a wide-body jetliner that crashed less 4

than 2 min after taking off over the ocean on a dark night......... 74

32. The inversion illusion ............................................. 75

33. Mechanism of the G-excess illusion ................................. 76

34. [levator illusion resulting from an updraft ....................... 78

35. The leans .......................................................... 80

36. Flow of orientation information in flight ........................ 90

37. The chain of events leading to a spatial disorientation mishap ..... 92

38. Iwo types of antivertiqo trainer currently in use ................. 95

39. A well-designed instrument panel ................................... 100 "*.

40. A typical head-jo display (HID) ................................. 102

41 . he Crane Al weather El ite, .................................. 103

4?. The peripheral visual horizon display (PVHD), or Malcolm horizon... 104

43. Conditioned .lot4 on sickness ........................................ 113

viii

I• -..A

SPATIAL %.RIENTATION IN FLIGHT

MECHANICS

Operators of today's and tomorrow's air and space vehiclesmust understand clearly the terminology and physical principlesrelating to the motions of their craft so they can fly withprecision and effectiveness. These crewmembers also have aworking knowledge of the structure and function of the variousmechanical and electrical systems of which their craft arecomprised, to help them understand the performance limits oftheir machines, and to facilitate trouble-shooting and promotesafe recovery when the machines fail in flight. So, too, muststudents of aerospace medicine understand certain basic defini-tions and laws of mechanics so they can analyze and describe themotional environment to which the flyer is exposed. In addition,thP A Pr orne Peic n.r.f..cai ^- Ir mozt bc fm.. Aar withz Lie jdybio-logic bases and operational limitations of the flyer's orienta-tional mechanisms. This understanding is necessary to enableaeromedical professionals to speak intelligently and crediblywith aircrew about spatial disorientation and to enable them tocontribute significantly to investigations of aircraft mishaps inwhich spatial disorientation may be implicated.

There are only two types of physical motion: linear moion

or motion Q1 translation, and dagular motion or motiQln odggtatio. Linear motion can be further categorized asr, meaning motion in a straight line, or cilinea,meaning motion in a curved path. Both linear motion and angularmotion are composed of an infinite variety of subtypes, ormotion parameters, based on successive derivatives of linear orangular position with respect to time. The most basic of thesemotion parameters, and the most useful, are displacement,v ity, acceleration, and jCL&. Table 1 classifies linear andangular motion parameters and their symbols and units, andserves as an outline for the following discussions of linear andangular motion.

S °% i .... f q .. ..- -"- -.. ... . . . . . ."

TABLE 1. LINEAR AND ANGULAR MOTION--SYMBOLS AND UNITS

Linear AngularMotion parameter

Symbols Units Symbols units

Displacement x meter (m) 6 degree(deg)nautical mile radian (cad)

(-1852 m) (-360/2n deg)

Velocity v, meter/second (m/sl c, e deg/sknot (10.514 rrs) rad/s

Acceleration a, v, x mis2 a, •, deg/s

2

g ('9.81 nxVs 2 ) rad/s 2

Jerk j, a, R/8;,x m/ 3 y, , D, " deq/s

3

g/s rad/s3

jiljj displacement. The other parameters--velocity, accelera-tion, jerk, etc.--are derived from the concept of displacement.Linear displacement, x, is the distance and direction of theobject under consideration from some reference point; as such,it is a vector quantity, having both magnitude and direction.The position of an aircraft located at 25 nautical miles on the150-degree radial of the Sari Antonio vortac, for exampLe,describes the linear displacement of the aircraft from thenavigational facility serving as the reference point. The meter(m), however, is the unit of linear displacement in the SI, orInternational System of Units, and will eventually replace otherunits of linear displacement such as feet, nautical miles, andetatute miles. Often miles or feet must be converted intometers to facilitate calculations involving linear motionparameters. Thus, the aircraft in the above example is 25 nauti-cal miles times 1.852 x 103 m/nautical mile, or 4.63 x 104 m,from San Antonio.

When linear displacement is changed during a period oftime, another vector quantity, linear vlo , occurs. Theformula for calculating the mean linear velocity, v, during timeinterval At is

x2 - x1 (I)

At

where x, = initial linear displacement

and x 2 = final linear displacement

An aircraft that travels from San Antonio to New Orleans in 1 hr,for example, moves with a mean linear velocity of 434 knots

2 .........

(nautical miles per hour) on a true bearing of 086 degrees.Statute miles per hour and feet per second are other commonlyused units of linear spe, the magnitude of linear velocity;but meters per second (m/s) is the SI unit, and is preferred.Frequently it is important to describe linear velocity at a %Nparticular instant in time, i. e. , as At approaches zero. Inthis situation one speaks of inas.tjat 2 1n=" vo y, ,("x-dot"), which is the first derivative of linear displacementwith respect to time, dA.

dt

When the linear velocity of an object changes over time,the difference in velocity, divided by the time required for themoving object to make the change, gives its mean linear acceler-al~ion, a.

V 2 - V 1

The formula a = (2) "-__ (2At

where V1 = initial velc ,t-

V2 = final velocity

and At = elapsed time

is used to calculate the mean linear acceleration, which, like .displacement and velocity, is a vector quantity with magnitudeand direction. Acceleration is thus the rate of change ofvelocity, just as velocity is the rate of change of displacement.The SI unit for the magnitude of linear acceleration is meterspec second squared (m/s 2 ). Consider, for example, an aircraftwhich accelerates from a dead stop to a velocity of 100 m/s in5 s; the mean linear acceleration is (100 m/s - 0 mn/s) + 5 s, or20 m/s 2 . The instantaneous linear acceleration, 3c ("x-double-dot*) or ;i, is the second derivative of displacement or thefirst derivative of velocity, • or dy, respectively.

dt dt

A very useful unit of acceleration is g, which for ourpurposes is equal to the constant g , the amount of accelerationexhibited by a freely falling body near the su.rface of theearth--9.81 m/s 2 . To convert values of linear accelerationgiven in m/s 2 into g units, simply divide by 9.81. In the aboveexample in which an aircraft accelerates at a mean rate of20 n's 2, 20 mVs 2 is divided by 9.81 m/s 2 per g to obtain anacceleration of 2.04 g.

A special type of linear acceleration, radial or c t-p'.acceleration, results in curvilinear, usually circular, motion.The acceleration acts along the line represented by the radiusof the curve and is directed toward the center of curvature.Its effect is a continuous redirection of the linear velocity,in this case called tanQenta• v, of the object subjected

3 6

-- -. *' -. %. *- * - , - -. - herONN... . . . . . . . .

to the acceleration. Examples of this type of linear acceler-ation occur when an aircraft pulls out of a dive after firing ona ground target, or engages an enemy in a circular path inaerial combat. The value of the centripetal acceleration, ac,can be calculated if one knows the tangential velocity, vt, andthe radius, r, of the curved path followed:

vt2

ac = -- (3)r

As an example, we can calculate the centripetal acceleration ofan aircraft traveling at 300 m/s (approximately 600 knots) andhaving a radius of turn of 1500 m. Dividing (300 m/s) 2

1.500 m gives a value of 60 m/s 2 which when divided by 9.81 m/sper g comes out to 6.12 g.

One can go anothe: step in the derivation of linear motionparameters by obtaining the rate of change of acceleration.This quantity, j, is known as lna ijjk. Mean linear jerk iscalculated in the following way:

a 2 - a1

At

where al = initial acceleration

a 2 - final acceleration

and At = elapsed time

Instantaneous linear j , *" or A, is the third derivative oflinear displacement or the first derivative of linear accelera-tion with respect to time; i.e., or d, respectively.

dt dt

Although the SI unit for jerk is m/s 3 , it is generally moreuseful to speak in terms of S-onset rA", measured in g's persecond (g/s).

Angular Motion. The derivation of the parameters ofangulat motion follows in a parallel fashion the scheme used toderive the parameters of linear motion. The I-asic parameter ofangular motion is angular displacement. For an object to beable to undergo angular displacement it mu:3t be polarized; i.e.,it must have a front and back, so that it can face, or bepointed in, a particular direction. A simple example of angulardisplacement is seen in a person facing east, in which case hisangular displacement is 90 degrees clockwise from the referencedirection, north. Angular displacement, symbolized by 0, isgenerally measured in degrees, revolutions (1 rev = 360 deg), orradians (1 rad = 1 rev 4 2v, approximatCely 57.3 deg) . Theradian is a particularly convenient unit to use when dealing

4

with circular motion (e.g., motion of a centrifuge) because oneneeds only to multiply the angular displacement of the system,in radians, by the length of the radius to find the value of thelinear displacement along the circular path. This relationholds because the radian is the angle subtended by a circulararc the same length as the radius Of the circle.

Angula velow, is the rate of change of angulardisplacement. The mean angular velocity occurring in timeinterval At is calculated thus:

2 01S= 02 - Qi(5)

At ,..•

where 01 = initial angular displacement""

and 02 = final angular displacement2 '4-

Instantaneous a yglove is 6, or dj. As an example ofdt

angular velocity we can consider the standard-rate turn ofinutrument ilying, in which a heading change of 180 degrees .is made in I min. Then 4) = (180 deg - 0 deg) + 60 s, or3 degrees per second (deg/s). This angular velocity can also bedescribed as 0.5 revolutions per minute (rev/min, or rpm) or as0.052 radians per second (rad/s) (3 deg/s divided by 57.3 degper rad). The fact that an objec' may be undergoing curvilinearmotion during a turn in no way affects the calculation of its .angular velocity: an aircraft being rotated on the ground on aturntable at a rate of half a turn per minute has the sameangular velocity as one flying a standard-rate instrument turn(3 deg/s) in the air at 300 knots.

Because radial or centripetal linear acceleration results ;when rotation is associated with a radius from the axis of .rotation, a formula for calculating the centripetal acceleration,ac, from the angular velocity, w, and the radius, r, is oftenuseful:

ac 2 r (6)

where • is the angular velocity in radians per second. One canconvert readily to the formula for centripetal acceleration interms of tangent.'al velocity (Eq. 3) if one remembers that

vt = ' r (7)

To calculate the centripetal acceleration generated by a cen-trifuge having a 10-m arm and turning at 30 rpm, one uses Equa-tion 6, after first converting 30 rpm to Ti radians per second.Squaring the angular velocity and multiplying by the 10-m radius,one obtains a centripetal acceleration of 10n2 m/s 2 , or 10.1 g.

a.-"

5o

IN

The rate of change of angular velocity is angular accelera-tion, a. The mean angular acceleration is

"•2 - 'al.-CL (8)

At

where W1 = initial angular velocity

w2 = final angular velocity

and At = time interval over which angularvelocity changes

-- ,

" 2V and can all be used to symbolize instantaneousdt- dtanua acceleration1, the second derivative of angular displace- V

ment or the first derivative of angular velocity with respect totime. If a figure skater is spinning at 6 revolutions per second

(2160 deg/s, or 37.7 rad/s) and then comes to a complete stop in2 s, the rate of change of angular velocity, or angular accelera- vtion, is (0 rad/s - 37.7 rad/s) + 2 s, or -18.9 radians persecond squared. Angular acceleration cannot be expressed in gunits, which measure magnitude of linear acceleration only.

Although not commonly used in aerospace medicine, anotherparameter derived from angular displacement is ngulax gj,, therate of change of angular acceleration. Its description iscompletely analogcus to that for linear jerk, but angular rather Nthan linear symbols and units are used. :1-

Force. Inertia, and Momentum

Linear and angular motions by themselves are of littlephysiologic importance. It is the forces and torques whichresult in, or appear to result frcm, linear and angular velocitychanges that stimulate or compromise the crewmember's physiologicmechanri smis.

Folce and Torque. FQ/g is an influence which produces ortends to produce linear motion or changes in linear motion; itis a pushing or pulling3 action. T produces or tends toproduce angular motion or changes in angular motion; it is atwisting or turning action. The SI unit of force is thenewton (N) . Torque ha. dimenrions of force and length, becausetorque is applied a,-:? Lcrce at a certain aistance from thecenter of rotation. "'ne ziewton meter (N m) is the SI unit oftorque.

Mass and Rotational Inertia. According to Newton's Law ofAcceleration:

F =m a (9)

6

where F = unbalanced force applied to anobject

m = mass of the object

and a = linear acceleration

To describe the analogous situation pertaining to angularmotion, we state that

M jct (10)

where M = unbalanced torque (or moment)applied to the rotating object

j = rotational inertia (moment

of inertia) of the object

and a angular acceleration

The m of an object is thus the ratio of the force acting onthe object to the acceleration resulting from that force. Massis therefore a measure of the ijnkgti of an object--its resist-ance to being accelerated. Similarly, rott•ioali2 i is theratio of the torque acting on an object to the angular accelera-tion resulting from that torque--again, a measure of resistanceto acceleration. The kilogram (kg) is the SI unit of mass, andis equivalent to 1 N/(m/s 2 ). The SI unit of rotational inertiais merely the N mn/(Lad/s 2 ).

Since F = m a, one can calculate the centripetal force, FC,

needed to produce a centripetal acceleration, ac, of a mass, m:

Fc = m ac (11)

iM Vt 2

So from Equation 3 Fc = (12)c

r

or from Equation 6 Fc n2 r (13)

where vt = tangential velocity

and ' = angular velocity

Newton's Law of Action and Reaction, that for every forceapplied to an object there is an equal and opposite reactiveforce exerted by that object, provides the basis for the conceptof in.al force.. Inertial force is an apparent force oppositein direction to an accelerating force and equal to the mass ofthe object times the acceleration. An aircraft exerting anaccelerating forward thrust on its pilot causes an inertialforce, the product of the pilot's mass and the acceleration, tohe exerted on the back of the seat by the pilot's body. Similar-ly, an aircraft undergoing positive cent ipetal acceleration as a

7

'S

result of lift generated in a turn causes the pilot's body to .5

exert inertial force on the bottom of the seat. More importantfrom our standpoint, however, are the inertial forces exerted onthe pilot's blood and organs of equilibrium, as physiologiceffects result directly from such forces.

At this point it is appropriate to introduce Q, which isused to measure the strength of the gravitoinertial force envi-rornment. (Do not confuse G with 0, the symbol for the universalgravitational constant, equal to 6.70 x 10-11 N m2 /kg2.)Strictly speaking, G is a measure of relative we:igh:

W .5-

G = -( 14)Wo

where W = weight observed in theenvironment under consideration

and Wo - normal weight on the surface

of the earth .

In the physical definition of weight,

W =m a (15)

and Wo = m go (16)

where m = mass

a - acceleratory field (vector sum 5,

of actual linear acceleration plusan imaginary acceleration ,ogit_the force of gravity)

and go = the standard value of the accelera-tion of gravity (9.81 m/ss2 )

Thus, a man having a mass of 100 kg would weigh 100 kg times9.81 m/s 2 or 981 N on earth (although conventional spring scaleswould read "100 kg"). At some other location or under someother acceleratory condition the same man could weigh twice asmuch--l,962 N--and cause a scale to read "200 kg," He wouldthen be in a 2-G environmenrt; or if he were in an aircraft, hewould be "pulling" 2 G. Consider also that,

Wsince G - ".

"W'_

m a

m go

8

-* -A. P

a

then G (17)go

So we see that the ratio between the ambient acceleratory field(a) and the standard acceleration (go) can also be representedin terms of G.

Thus, we use g as a unit of acceleration (e.g., ac = 8 g),and reserve the dimensionless ratio of weights, G, for describingthe resulting gravitoinertial force environment (e.g., a forceof 8 G, or an 8-G load). When in the vicinity of the surface ofthe earth, one feels a G force equal to 1 G in magnitude directedtoward the center of the earth. If one also sustains a G forceresulting from linear acceleration, the magnitude and directionof the resultant gravitoinertial G force can be calculated byadding vectorially the 1-G gravitational force and the inertialG force. An aircraft pulling out of a dive with a centripetalacceleration of 3 g, for example, would exert 3 G of centrifugalforce. At the bottom of the dive the pilot would experience the3-G centrifugal force in line with the I-G gravitational force,for a total of 4 G directed toward the floor of the aircraft.If the pilot could continue his circular flight path at aconstant airspeed, the G force experienced at the top of theloop would be 2 G, since the l-G gravitational force wouldsubtract from the 3-G inertial force. Another common example ofthe addition of gravitational G force and inertial G forceoccurs during application of power on takeoff or on a missedapproach. if the forward acceleration is 1 g, then the inertialforce is 1 G directed toward the tail of the aircraft. Theinertial force adds vectorially to the 1-G force of gravity,directed downward, to provide a resultant gravitoinertial forceof 1.414 G pointing 45 degrees down from the aft direction.

Just as inertial forces oppose acceleratory forces, so doine joj oppose acceleratory torques. No convenientderived unit exists, however, for measuring inertial torque;specifically, there is no such thing as angular G.

M•nL•uim. To complete this discussion of linear andangular motion, we introduce the concepts of Momnetum andimil. Linear igmentu is the product of mass and linearvelocity, m v. AnqulaL mgmgtum is the product of rotationalinertia and angular velocity, j w. Momentum is a quantity whicha translating or rotating body conserves; i.e., an object cannotgain or lose momentum unless it is acted upon by a force ortorque. A translational i is the product of force, F, andthe time over which the force acts on an object, At, and isequal to the change in linear momentum imparted to the object.Thus

F At = m v 2 - m v 1 (18)

where v, = initial linear velocity

_j- W V V W , UV'j W, 1P . 1%_%M TAX AA l~ft mS 'ýA K&M "R 6. It U W1 b'U .ý- 6I .~N -k~ .. U r-X._ V

and v 2 = final linear velocity 'p

When dealing with angular motion, one defines a r Iinimpulu to be the product of torque, M, and the time over whichit acts, At. A rotational impulse ic equal to the change inangular momentum. Thus ;

M At =j W2 - j W, (19)

where = initial angular velocity

and w2 = final angular velocity

The above re]ations can be seen to have been derived fromthe Law of Acceleration: "4

m aC-

Mj "

V 2 - W 1

since aAt

and a• - •.J

Directions of Action and Reaction

A number of conventions have been used in aerospace medicineto describe the directions of linear and angular displacement,velocity, and acceleration and of linear and angular reactiveforces and torques. The more commonly used of those conventionnwill be described.

Vehicular Motions. Because space is three-dimensional, wedescribe linear motions in space by reference to three linearaxes and angular motions by three angular axes. In aviation itis customary to speak of the longitudinal (fore-aft), lateral(right-left), and vertical (up-down) linear axes, and the roll,pitch, and yaw angular axes, as shown in Figure 1.

Most linear accelerations in aircraft occur in the verticalplane defined by the longitudinal and vertical axes, sincethrust is usually developed along the former and lift is usuallydeveloped along the l.atter axis. Aircraft capable of vectoredthrust are now operational, however, and vectored-lift aircraftare currently being flight tested. This means that in therelatively near future aircraft will operate in a complete,three-dimensional, linear acceleration environment. We canexpect that spacecraft will do the same in the more distantfuture. Most angular accelerations in aircraft occur in theroll plane (perpendicular to the roll axis), and to a lesserextent in the pitch plane. Angular motion in the yaw plane is

10

MW ~ ~ ~ ~ ~ ~ I "U- UJM U~WFU&UUW * WW IX V W VM V A V WYf' AF ILMX M U USMPM~NK M K i N --- A. 'h~U.MTZT ~TZ~

very limited in normal flying, although it does occur duringspins and several other aerobatic maneuvers. Again, aircraftand space vehicles of the future can be expected to operate withconsiderably more freedom of angular motion than those of thepresent.

AXIS OFSYAW

OF PITCH

-//9

oOF ROLL . ,

AXIS

Figure 1. Axes of linear and angular aircraft motions. Linearmotions are longitudinal, lateral, and vertical, andangular motions are roll, pitch, and yaw.

Physiolooic Acceleration and Reaction Nomenclature. Fig-ure 2 depicts a practical. system for describing linear andangular accelerations acting on man. This system is usedextensively in aeromedical scientific writing. In this system,a linear acceleration of the type associated with a conventionaltakeoff roll is in the +ax direction; i.e., it is a +a accelera-tion. Braking to a stop during a landing roll resuylts in -axacceleration. Radial acceleration of the type usually developedduring air combat maneuvering is +az acceleration--foot-to-head.The right-hand rule for describing the relations between tnreeorthogonal axes aids recall of th6 positive directions of a., ay,

]I

and a. accelerations in this particular system: if the forward-pointing index finger of the right hand represents the positive 4A

x-axis, and the left-pointing middle finger of the right hand .represents the positive y-axis, then the positive z-axis isrepresented by the upward-pointing thumb of the right hand. Beaware, however, of a different right-hand rule used in anotherconvention, one for describing vehicular coordinates. In thatsystem, +ax is noseward acceleration, +a is to the right, and+a is floorward; an inverted right hand Yllustrates that set of

9axes,

The angular accelerations, x, % ,, and .Z, are roll, pitch,and yaw accelerations, respectiveli, in the system shown inFigure 2. Note that the relation between the positive x-, y-,and z-axes is identical to that for linear accelerations. Thedirection of positive angular velocity or acceleration isdescribed by another right-hand rule, wherein the flexed fingersof the right hand indicate the direction of angular motioncorresponding to the vector represented by the extended, abductedright thumb. Thus, a right roll results from +aX acceleration,a pitch down results from +a acceleration, and a left yawresults from + °l acceleration Yn this system. Again, be awareof the inverted right-hand coordinate system commonly used todescribe angular motions of vehicles. In that convention apositive roll acceleration is to the right, positive pitch isupward, and positive yaw is to the right.

The nomenclature for the directions of gravitoinertial (G)forces acting on humans is also illustrated in Figure 2. Notethat the relation of these axes to each other follows a backward,inverted, right-hand rule. In the illustrated convention, +aXacceleration results in +Gx inertial force, and +a, results in+Gz&. This correspondence of polarity is not achieved on they-axis, however, as +a acceleration results in -G force. Ifthe +G direction wer[ reversed, full polarity cgrrespondencecould Ye achieved between all linear accelerations and allreactive forces; and that convention has been used by someauthors. An example of the usage of the symbolic reactionterminology would be: "An F-16 pilot must be able to sustain+9.0 Gz without losing vision or consciousness."

Fig¶re 2. System for describing accelerations and inertialreactions in man. (Adapted from Hixson et al. (1).)

12

+Z

ANATOMICAL AXES x,y,z ANATOMICAL AXES xyz'GxV

+ +Gy;..

+Gz -"-

LINEAR ACCELERATION axE a, z LIEAR REACTION GxE Gy, Gz

+a•z +R x"-"-<

t- +ay ýR y,.,

+Rz

ANGULAR ACCELERATION •x, •y, (z ANGULAR REACTION Rx, R y, Rz

13

• . .- ,

The "eyeballs" nomenclature is another very useful set ofterms for describing gravitoinertial forces. In this system, .the direction of the inertial reaction of the eyeballs when thehead is subjected to an acceleration is used to describe thedirection of the inertial force. The equivalent expressions,,"eyeballs-in acceleration" and "eyeballs-in G force," leavelittle room for confusion about either the direction of the K[.applied acceleratory field or the resulting gravitoinertialforce environment.

Inertial torques can be described conveniently by means ofthe system shown in Figure 2, in which the angular reaction axesare the same as the linear reaction axes. The inertial reactivetorque resulting from +a, (right roll) angular acceleration is+Rx, and +az (left yaw) results in +Rz; but +a (downward pitch)results in -R This incomplete correspondence between accelera- - -

tion and rea tion coordinate polarities again results from themathematical tradition of using right-handed coordinate systems.

It should be apparent that the potential for confusing theaudience when speaking or writing about accelerations andinertial reactions is such as to make it a virtual necessity todescribe the coordinate system being used. For most applicationsthe "eyeballs" convention is perfectly adequate.

14A

S.%

;.. A

'A .. •-,

-h

VISUAL ORIENTATION

Vision is by far the most important sensory modalitysubserving spatial orientation, especially so in moving vehiclessuch as aircraft. Without it, flight as we know it would beimpossible, whereas this would not necessarily be the case inthe absence of the vestibular or other sensory systems thatprovide orientation information. For the most part, the function 4of vision in spatial orientation is obvious--one sees where heis and how he is moving--so we shall not devote to it a dis-cussion proportional in size to the importance of that functionin orientation. Certain recent developments in our understandingof visual orientation deserve mention, however. First, the fact 4that there are actually two separate visual systems with twodistinct functions--one being object recognition and the otherspatial orientation--is of extreme importance, both for under-standing visual illusions in flight and for appreciating thedifficulties inherent in using flight instruments for spatialorientation. Second, the realization that visual and vestibularorientation information are integrated at very basic neurallevels helps us understand why spatial disorientation so fre-quently is not amenable to correction by higher-level neuralprocessing. b* .

Anatomy of the Visual System

G . The retina, an evaginated portion of the embryonicbrain, consists of an outer (more peripheral) layer of pigmentedepithelium and an inner layer of neural tissue. Containedwithin the latter layer are the sensory rod and cone cells, thebipolar and horizontal cells that comprise the intraretinalafferent pathway from the rods and cones, and the multipolarganglion cells, the axons of which are the fibers of the opticnerve. The cones, which number approximately seven million in , J*the human eye, have a relatively high threshold to light energy.They are responsible for sharp visual discrimination and colorvision. The rods, of which there are over a hundred million,are much more sensitive to light than are the cones; theyprovide the ability to see in twilight and at night. In theretinal macula, near the posterior pole of the eye, the conepopulation achieves its greatest density; within the macula, thefo g j--a small pit totally comprised of tightly packedslender cones--provides the sharpest visual acuity, and is the .anatomic basis for Lnpi or central vision. The remainder ofthe eye is capable of far less visual acuity, and subservesR ntLU and xI.eral vision.

Having dendritic connections with the rods and cones, thebipolar cells provide axons that synapse with the dendrites orcell bodies of the multipolar ganglion cells, whose axons in

15

turn course parallel to the retinal surface and converge at theoptic disc. Emerging from the eye as the ot a=.t.U, they meettheir counterparts from the opposite eye in the o t ia gkJA&,and then continue in one of the optic L , most likely toterminate in a lateral gj l .q-•X, but possibly in a

u g ir .ulul or the r/ t &LCA. Second-order neuronsfrom the lateral geniculate body comprise the geniculocalcarinetract, which becomes the ot r and terminates in the

Svisual cortex, the striate area of the occipital cerebralcortex (Area 17) . In the visual cortex the retinal image isrepresented as a more or less point-to-point projection from thelateral geniculate body, which receives a similarly topograph-ically structured projection from both retinas. The lateralgeniculate body and the primary visual cortex are thus struc-turally and functionally suited for analysis of and recognitionof visual images. The superior colliculi, on the other hand,project eventually to the motor nuclei of the extraocularmuscles and muscles of the neck, and thus appear to provide apathway for certain gross ocular reflexes of visual origin.Fibers entering the pretectal area are involved in pupillaryreflexes. In addition, most anatomic and physiologic evidenceindicates that information from the Pajj cr a cortethe f .y& m &L" (Area 8), and the oassociation aXeaj (Areas 18 and 19) is relayed through thep igIn _ntine r±icniar formation to the nuclei of thecranial nerves (III, IV, and VI) innervating the extraocularmuscles. Via this pathway and perhaps others involving the I%.superior colliculi, saccadic (fast) and pursuit (slow) eyemovements are initiated and controlled. Interestingly, saccadiceye movements appear to originate in the contralateral hemi-sphere, while pursuit movements are under ipsilateral corticalcontrol (2).

Visual-vestibular Converqence. We now know that visuallyperceived motion information and probably other visual orienta-tional data reach the vesibl i in the brain stem (3,4).The neural pathways by which this convergence is accomplishedare not known, however. The hypothesis that the superiorcclliculi project through the reticular formation to the vestibu-lar nuclei is attractive, because of the demonstrated responsive-ness of most superior colliculus neurons to moving rather thanstationary visual stimuli; but this hypothesis is not yet sup-ported by anatomic evidence. Most likely the cerebellum isinvolved in visual-vestibular convergence via the recentlydiscovered accessory optic system. Visual information fromthe retina reaches the i !jliy by way of the access-r

.i trac and gentral j tract. Fibers from theinferior olive provide inputs to the Purkinje cells of thecerebellar flocclus, which have inhibitory connections inipsilateral vestibular nuclei. Because this area of the cerebel-lum receives mossy fibers directly from vestibular end-organs,visual-vestibular interactions can also occur within thecerebellum; and as the Purkinje cells project onto the vestibu-lar end-organs themselves (at least in some species), such

36

bib""""

interactions can take place peripherally as well as centrally. %There are also possible visual-vestibular interactions thatoccur as a result of cerebral cortical projections to theRonting nuclei, thence into the cerebellar vi and cerebellar 4hemix.&. Of course, the confluence of visual and vestibularpathways in the paramedian pontine reticular formation makespossible the integration of visual and vestibular inputs to thefinal common pathway to the nuclei of cranial nerves III, IV,and VI. As we might expect, there are also afferent vestib-ular influences on visual system nuclei; and these have been 4demonstrated in the lateral geniculate body and superiorcolliculus (4).

Visual Information Processinu

Primary control of man's ability to move and orient himselfin three-dimensional space is mediated by the visual system.This is exemplified by the fact that individuals without func-tioning vestibular systems (*labyrinthine defectives") have vir-tually no problems with spatial orientation unless they aredeprived of vision. The underlying mechanisms of visualorientation-information processing are revealed by receptive 4field studies, which have been accomplished for the peripheralretina, nuclear relays, and primary visual cortex. Basically,these studies show that there are several types of movement-detecting neurons, and that these neurons respond differentlyto the direction of movement, velocity of movement, size of thestimulus, its orientation in space, and the level of illumina-tion. (For an excellent review of this fascinating topic, seeGriisser et al., 1973 (5)•) Functionally, however, vision mustbe considered as two separate systems, one involved with objectrecognition and the other with spatial orientation. These twosystems, the f and a visual systems, respectively,will be described below, as will certain functions of yet anothervisual system, that responsible for generating eye movements tofacilitate acquisition of orientation information.

Focal Vision. Liebowitz and Dichgans (6) have provided avery useful summary of the characteristics of focal vision:

"[The focal visual mode] is concerned withobject recognition and identification andin general answers the question of 'what.'Focal vision involves relatively fine detail(high spatial frequencies) and is corre-spondingly best represented in the central "visual fields. Information processed byfocal vision is ordinarily well representedin consciousness and is critically relatedto physical parameters such as stimulusenergy and refractive error. The vast major-ity of studies in 'vision' as well as most

17

existing tests for evaluating performance orindividual differences are concerned withfocal function."

Focal vision is thus not primarily involved with orientingoneself in the environment, but is used in some instances toacquire visual information about orientation. Certainly, focalvision is necessary for reading flight instruments; however, acomplex cognitive process--i.e., instrument flying skill--isrequired to convert such focally acquired orientation cues intousable orientation information.

In the visual (as opposed to instrument) flight environment,focal visual cues provide the primary means by which judgmentsof distance and depth are made. Tredici (7) categorizes thesecues as being either monogiiii or b a. The monocular cuesare: (1) Lj2,g coas .y, the size of the retinal image inrelation to known and comparative sizes of objects; (2) scnta, the shape of the retinal image in relation to theknown shape of the object (e.g., the foreshortening of the imageof a known circle into an ellipsoid shape means one part of thecircle is farther away than another); (3) m pgla, therelative speed of movement of images across the retina--when oneis moving linearly in his environment, the retinal images ofnearer objects move faster than those of objects farther away;(4) interposition, the partial obstruction from view of moredistant objects by nearer ones; (5) textUre or g, theapparent loss of detail with greater distance; (6) linearperspective, the convergence of parallel lines at a distance;(7) illumination R.s•c _v, which results from the tendency toperceive the light source to be above an object, and from theassociation of more deeply shaded parts of an object with beingfurther from the light source; and (8) a T/iers•ec'ive, theperception of objects to be more distant when the image isrelatively bluish or hazy. The binocular cues to depth anddistance are: (1) s.ereR.", the visual appreciation ofthree-dimensional space that results from fusion of slightlydissimilar retinal images of an object; (2) y, the medialrotation of the eyes and the resulting direction of their gazealong more or less converging lines depending on whether theviewed object is closer or farther, respectively; and (3) accom-modation, or focusing of the image by changing the curvatureof the lens of the eye. Of all the cues listed, size and shapeconstancy and motion parallax appear to be most important forderiving distance information in flying, as they are availableat and well beyond the distances at which binocular cues areuseful. Stereopsis can provide orientation information atdistances up to only about two hundred meters; it is, however,more important in orientation than are vergence and accommoda-tion, which are of no use beyond about six meters.

U,. b ~ - ]8

S S * S * * . ** S * - - *. - - -

04

Ambient Vision. Again, we refer to Liebowitz and Dichgans 0,for a summary:

"[The ambient visual mode] subserves spatiallocalization and orientation and is ingeneral concerned with the question of'where.' Ambient vision is mediated by Crelatively large stimulus patterns so thatit typically involves stimulation of theperipheral visual field and relativelycoarse detail (low spatial frequencies).Unlike focal vision, ambient vision is notsystematically related to either stimulusenergy or optical image quality. Rather,provided the stimulus is visible, orientationresponses appear to be elicited on an 'allor none' basis. ... The conscious concomitantof ambient stimulation is low or frequentlycompletely absent. Interest in ambientvisual function has a long history, butanalyses of psychophysical properties,[etc.], ... have been initiated primarilywithin the past ten to fifteen years.'

Ambient vision, therefore, is primarily involved with orientingoneself in the environment. Furthermore, this function iscompletely independent of the function of focal vision. This isevidenced by the fact that one can fully occupy central visionwith the task of reading while simultaneously obtaining suffi-cient orientation cues with peripheral vi6ion to walk or ride abicycle. It is also evidenced by the ability of certain patientswith cerebral cortical lesions to maintain visual orientationresponses even though their ability to discriminate objects islost ( blindsight"--presumably mediated through subcorticalstructures).

The function of ambient vision in orientation can bethought of as two processes, one providing motion cues and theother providing position cues. Large, coherently moving con-trasts detected mainly with peripheral vision result in apercept of self-motion, or vection. If the moving contrastsrevolve relative to the subject, he perceives rotational self-motion, or a y.• , which can be in the pitch, roll,yaw, or any intermediate plane. If the moving contrasts enlargeand diverge from a distant point, become smaller and converge inthe distance, or otherwise indicate linear motion, the perceptof self-motion that results is linear ei, which also can bein any direction. Vection can, of course, be veridical orillusory, depending on whether actual or merely apparent motionof the subject is occurring. One can appreciate the importanceof ambient vision in orientation by recalling the powerfulsensations of self-motion generated by certain scenes in wide-screen motion pictures (e.g., flying through the Grand Canyon).

19

p - ~ -. . . . . ~ -..................... ' 4 *.* ~~ .* *.. . . . . . . .

Position cues provided by ambient vision are readilyevidenced in the stabilization of posture that vision affordspatients with defective vestibular or spinal proprioceptivesystems. The essential visual parameter contributing to posturalstability appears to be motion of the retinal image that resultsfrom minor deviations from desired postural position. Visualeffects on posture can also be seen in the phenomenon of heightvertiao. As the distance from (height above) a stable visualenvironment increases, the amount of body sway necessary for theretinal image movement to be above threshold increases. Above acertain height the ability of this visual mechanism to contribute 4to postural stability is exceeded, and vision indicates postureto be stable despite large body sways. The conflict betweenvisual orientation information, indicating relative stability,and the vestibular and somatosensory data, indicating large bodysways, results in the unsettling experience of vertigo.

One more distinction between focal and ambient visualfunction should be emphasized. In general, focal vision servesto orient the perceived object relative to oneself, whereasambient vision serves to orient oneself relative to the perceivedenvironment. When both focal and ambient vision are present,orienting a focally perceived object relative to the ambientvisual environment is easy, whether the mechanism employedinvolves first orienting the object to oneself and then orientingoneself and the object to the environment, or whether the objectis oriented directly to the environment. When only focal visionis available, however, it can be difficult to orient oneselfcorrectly to a focally perceived orientation cue, because thenatural tendency is to perceive oneself as stable an( upright,and to perceive the focally viewed object as oriented otherwisewith respect to the stable and upright egocentric referenceframe. This phenomenon can cause a pilot to misjudge hisapproach to a night landing, for example, when only the runwaylights and a few other focal visual cues are available forspatial orientation.

Eve Movemen . The maintenance of visual orientation in adynamic environment is greatly enhanced by the ability to movethe eyes, primarily because the retinal image of a moving objector environment can be stabilized by appropriate eye movements.Very basic and powerful mechanisms involved in reflexive vestibu-lar stabilization of the retinal image will be discussed in thesection on vestibular function. Visually controlled eye move-ments that provide image stabilization are the slow nrs-movements and the rapid saccadic movements. Pursuit movementsadequately track targets moving at less than 60 degrees/s;targets moving at higher velocities necessitate either saccadiceye movements or voluntary head movements for adequate tracking.Saccadic eye movements are used voluntarily or reflexively toacquire a target or to catch up to a target which cannot bemaintained on the fovea by pursuit movements. Under somecircumstances, pursuit and saccadic eye movements alternate in a

20

pattern of reflexive slow tracking and fast back-tracking--optokinetic nltagmu&. This type of eye-movement response istypically elicited in the laboratory by surrounding the subjectwith a rotating striped drum; but one can exhibit and experience 4

optokinetic nystagmus quite readily in a more natural setting bywatching railroad cars go by while waiting at a railroad cross-ing. Movement of the visual environment sufficient to elicitoptokinetic nystagmus provides a stimulus that can eitherenhance or compete with vestibular elicitation of eye movements,depending on whether the visually perceived motion is compatibleor incompatible, respectively, with that sensed by the vestibularsystem. V movements, which aid binocular distance andmotion perception at very close range, are also visually con-trolled, but are of relatively minot importance in spatialorientation when compared with the image-stabilizing pursuit andsaccadic eye movements.

Even though gross stabilization of the retinal image aidsobject recognition and spatial orientation by enhancing visualacuity, absolute stability of an image is associated with amarked decrease in visual acuity and form perception, as shownexperimentally by Ditchburn and Ginsborg (8) . This stability-induced decrement is avoided by continual voluntary and involun-tary movements of the eyes, even during fixation of an object.We are unaware of these small eye movements, however, and thevisual world appears stable.

We tend to take for granted that voluntary scanning and 4tracking movements of the eyes are associated with the appearanceof a stable visual environment; but why this is so is notreadily apparent. Early investigators postulated that proprio-ceptive information from the extraocular muscles provided notonly feedback signals for control of eye movements but also theafferent information needed to correlate eye movement withretinal image movement and arrive at a subjective determinationof a stable visual environment. In 1960 Brindley and Merton (9)demonstrated that in man there is no subjective eye-positionsense--i.e., no conscious extraocular-muscle proprioception.An alternative mechanism for oculomotor control and subjectiveappreciation of visual stability then became plausible: that wasthe "corollary discharge" or feed-forward mechanism first pro-posed by Sperry (10). From his studies of the motor effects ofsurgical rotation of the eye in fish, Sperry concluded: "Thus,an excitation pattern that normally results in a movement thatwill cause a displacement of the visual image on the retina mayhave a corollary discharge into the visual centers to compensatefor the retinal displacement. This implies an anticipatoryadjustment in the visual centers specific for each movement withregard to its direction and speed." The theoretical aspects ofvisual perception of movement and stability have been expandedover the years into various models based on "inflow" (afference),"outflow" (efference) , and even hybrid sensory mechanisms. Theinterested reader will enjoy Cohen's (11) concise discussion ofthese models as they relate to spatial orientation.

21

VESTIBULAR FUNCTION

The role of vestibular function in spatial orientation isnot so overt as that of vision, but is extremely important forthree major reasons. First, the vestibular system provides thestructural and functional substrate for reflexes that serve tostabilize vision when motion of the head and body would otherwiseresult in blurring of the retinal image. Second, the vestibularsystem provides orientational information with reference towhich both skilled arid reflexive motor activities are automat-ically executed. Third, the vestibular system provides, in theabsence of vision, a reasonably accurate percept of motion andposition, as long as the pattern of stimulation remains withincertain naturally occurring bounds. Because the details ofvestibular anatomy and physiology are not usually well known bymedical professionals, and because a working knowledge of themis essential to the understanding of spatial disorientation inflight, we present those details here.

Vestibular Anatomy

E-orgaa. The vestibular end-organs are smaller thanmost people realize: measuring just 1.5 cm across, the wholevestibular sensory apparatus would fit inside the bowl of atypical pipe. They reside well-protected within some of thedensest bone in the body, the petrous portion of the temporalbone. Each temporal bone contains a tortuous excavation knownas the Q laby.inh, which is filled with piy , a fluidmuch like cerebrospinal fluid in composition. The bony labyrinthconsists of three main parts: the cochlea, the vtu, andthe semicircular canals (Fig. 3). Within each part of the bonylabyrinth is a part of the delicate, tubular, m ny- ,.rinth, which contains e lvMI•, a fluid characterized by itsrelatively high concentration of potassium. In the cochlea themembranous labyrinth 's called the QQc.Ii: dU or scala mia;this organ converts acoustic energy into neural information. Inthe vestibule lie the two otith organs the utricle and theg . They translate gravitational and inertial forces intospatial otientation information--specifically, information aboutposition and linear i:iotion of the head. The semicircular ýucts, %%contained in the semicircular canals, convert inertial torquesinto information about angular motion of the head. The threesemicircular canals and their included semicircular ducts areoriented in three mutually perpendicular planes, thus inspiringthe names of the canals: antegior vtl1 (or L), ['•Rs vical (or R r ), and horizoali (or latera)"

22

IVOOL V UP" IPACI

"PlI %YMPH IPACI

ANIESIOl VIIrCA4 SIMICAICUILA CANAL ,qA *•

POI1mO.0 V6ETfCAt SIMICICULAL: CANAL AND DUCT ANT#1101 vIIII(At 641AICiULAI DUCT

OU.Nv ANPUUA

1AIMIA| ANOUS A NlP~tI A"

* L INDO00MONAIIC Dv".

U1ICKUtA1 AND $ACCLJLAI NINYII

00*120N1AL %1MKCISCLLIAA CANAL AIED DUCT IBV

(A" AlA * k Vi ?

PPP •AI :...fjOTVAL WliDow

v1§116 to 10V"0 VVINDOCTI" -

Figure 3. Gross anatomy of the inner ear. The bony semicircularcanals and vestibule contain the membranous semi-circular ducts and otolith organs, respectively.

The semicircular ducts communicate at both ends with the . #

utricle, and one end of each duct is dilated to form an amRuJ . 'h"Inside each ampulla lies a crest of neuroepithelium, the gria..iimp)aris; atop the crista, occluding the duct, is a gelatinousstructure called the ciuaj (Fig. 4a). The hair cells of whichthe crista ampullaris is composed project their cilia into thebase of the cupula, so that whenever the cupula moves, the ciliaare bent.

Lining the bottom of the ut-icle in a more or less hori-zontal plane is another patch ot neuroepithelium, the maculaU.X,..,•,1 and on the medial wall of the saccule in a vertical -plane is still another, the m gc (Fig. 4b). The ciliaof the hair cells comprising these structures pruject intooverlying g.lithic mebrg one above each macula. Theotolithic membranes are gelatinous structures containing manytiny calcium carbonate crystals, called o, which are heldtogether by a network of connective tissue. Having almost threetimes the density of the surriuunding endolymph, the otolithicmembranes displace endolymph and shift position relative to

23

I%

their respective maculae when subjected to changing gravito-inertial forces. This shifting of otolithic membrane positionresults in bending of the cilia of the macular hair cells.

-'I.

OTOCO 0TOLA '

(or s~ccLMEMBRA~NEMEROaOPANOUS AMPULLA ENOOLYW.,•-__ _%

CUPriA \_. -_ ]-

NOLYMPH--

HNAP CEI LS aHAP CELLS / "IL,"CRISTA AMPULARIS AMPUUAR UTCIULA,

(o r (or SACCUL) RiE

Figure 4. The vestibular end-organs. a. The ampulla of a semi-circular duct, containing the crista ampullaris andcupula. b. A representative otolith organ, with its ."-macula and otolithic membrane.

The hair cell is the functional unit of the vestibularsensory system: it converts spatial and temporal patterns ofmechanical energy applied to the head into neural information.Each hair cell possesses one relatively large hingcilium on oneside of the top of the cell and up to a hundred smaller scitj~l on the same surface. Hair cells thus exhibit L.gJhQ-.giapolarization; i.e., they are oriented in a particular direction.The functional correlate of this polarization is that when the 'cilia of a hair cell are bent in the direction of its kinocilium,the cell undergoes an electrical depolarization, and the frequen-cy of action potentials generated in the vestibular neuronattached to the hair cell increases above a certain restingfrequency--the greater the deviation of the cilia, the higherthe frequency. Similarly, when its cilia are bent away from the a-

side with the kinocilium, the hair cell undergoes an electricalhyperpolarization, and the frequency of action potentials in thecorresponding neuron in the vestibular nerve decreases (Fig. 5).

24

I .AF

TOWARD AWAY FROM .POSITION OF CILIA NEUTRAL KINOCILIUM KINOCILIUM I

KINOCILIUM _ _

(1)II

STEREOC;LIA I

(60.100)

HAIR CELL~

VESTIBULAR ofAFFERENT NERVE ENDING

ACTION POTENTIALS / y IVESTIBULAR7 ,EFFERENT NERVE ENDING i .

POLARIZATION [ NORMAL DEPOLARIZED HYPEROF HAIR CELL N RPOLARIZEO

FREQUENCY OF RESTING HIGHER LOWERACTION POTENTIALS R GI LOWER

Figure 5. Function of a vestibular hair cell. When mechanical• • ~*, " M'+

forces deviate the cilia toward the side of the cellwith the kinocilium, the hair cell depolarizes and thefrequency of action potentials in the associated 0% %afferent vestibular neuron increases. When the ciliaare deviated in the opposite direction, the hair cellhyperpolarizes and the frequency of action potentialsdecreases.

The same basic process described above occurs in all thehair cells in the three cristae and both maculae; the importantdifferences lie in the physical events that cause the deviationof cilia and in the directions the various groups of hair cellsare oriented. The hair cells of a crista ampullaris respond tothe inertial torque of the ring of endolymph contained in theattached semicircular duct, as the reacting endclymph exerts --pressure on the cupula and deviates it. The hair cells of amacula, on the other hand, respond to the gravitoinertial forceacting to displace the overlying otolithic membrane. As indi- r.N1-4cated in Figure 6a, all of the hair cells in the crista of thehorizontal semicircular duct are oriented so that their kinociliaare on the utricular side of the ampulla. Thus, utriculopetalendolymphatic pressure on the cupula deviates the cilia of thesehair cells toward the kinocilia, and all the hair cells in the

257°..

ACC0ULAR MACULAJ

SPRORjUDIRETIAON"-a

,* .- *-IN -FEREIO

a b-

-eicrulI-dctFEeorete oFta ter io

LATERAL~ih ANTERIOR

a.VET A POSTERIOR EDIAL

ciliTIALa arintedecinothutil hsen

C RIST7A .

Figure 6. Morphologic polarizaticn in vestibular neuroepithelia.a. All the hair cells in the cristae of the horizontalsemicircular ducts are oriented so that their kino-cilia are in the direction of the utriclei those inthe cristae of the vertical ducts have their kinociliadirected away from the utricle. b. The maculae of thesaccule (above) and utricle (below) also exhibitpolarization--the arrows indicate the direction of thekinocilia of the hair cells in the various regions ofthe maculae. (Adapted from Spoendlin (12) .)

crista depolarize. *rhe hair cells in the cristae of the verticalsemicircular ducts are oriented in the opposite fashion; i.e.,their kinocilia are all on the side away from the utricle. Inthe ampullae of the vertical semicircular ducts, therefore,utriculopetal endolymphatic pressure deviates the cilia awayfrom the kinocilia, causing all the hair cells in these cristaeto hyperpolarize. In contrast, the hair cells of the maculaeare not oriented unidirectionally across the neuroepithelium:the direction of their morphologic polarization depends on wherethey lie on the macula (Fig. 6b) . In both maculae there is acentral line of reflection, on opposing sides of which the haircells as3ume an opposite orientation. In the utricular maculathe kinocilia of the hdir cells are all oriented toward thisline of reflection, whereas in the saccular macula they areoriented away from it. Because the line of reflection on eachmacula curves at least ninety degrees, the hair cells, having

26

morphologic polarization roughly perpendicular to this line,exhibit virtually all possible orientations on the plane of themacula. Thus, the orthogonality of the planes of the threesemicircular ducts enables them to detect angular motion in anyplane; and the perpendicularity of the planes of the maculae,plus the omnidirectionality of the orientation of the hair cellsin the maculae, allow the detection of gravitoinertial forcesacting in any direction.

Neural Pathways. To help the student organize better inhis mind, the potentially confusing vestibular neuroanatomy, asomewhat simplified overview of the major neural connections ofthe vestibular system is presented in Figure 7. The utricularnerve, two saccular nerves, and the three ampullary nervesconverge to form the vestibular nerve, a portion of the VIIIthcranial or statoacoustic nerve. Within the vestibular nerve .lies the vestibular (or Scarpa's) ganglion, which is comprisedof the cell bodies of the vestibular neurons. The dendrites ofthese bipolar neurons invest the hair cells of the cristae andmaculaei most of their axons terminate in the four vestibularnuclei in the brain stem--the superior, medial, lateral, and CV

inferior--but some enter the phylogenetically ancient parts ofthe cerebellum to terminate in the fastigial nuclei and in thecortex of the flocculonodular lobe and other parts of the 1.posterior vermis.

The vesLibular nuclei project via secondary vestibulartracts to the motor nuclei of the cranial and spinal nerves andto the cerebellum. As vestibulo-ocular reflexes are a majorfunction of the vestibular system, it is not surprising to findample projections from the vestibular nuclei to the nuclei ofthe oculomotor, trochlear, and abducens nerves (cranial nervesIII, IV, and VI, respectively) The major pathway of theseprojections is the ascending medial longitudinal fasciculus(MLF). The basic vestibulo-ocular reflex is thus served bysensor and effector cells and an intercalated three-neuronreflex arc from the vestibular ganglion to the vestibular nucleito the nuclei innervating the extraocular muscles. In addition,there are indirect multisynaptic pathways that course from thevestibular nuclei through the paramedian pontine reticularformation to the oculomotor and other nuclei. The principle ofipsilateral facilitation and contralateral inhibition via aninterneuron very clearly operates in vestibulo-ocular reflexes,and numerous crossed internuclear connections provide evidenceof this. The vestibulo-ocular reflexes that the various ascend-ing and crossed pathways support serve to stabilize the retinalimage by moving the eyes in the direction opposite that of themotion of the head. Via the descending MLF and medial vestibulo-spinal tract, crossed and uncrossed projections from the vestibu-lar nuclei reach the nuclei of the spinal accessory nerve(cranial nerve XI) and motor nuclei in the cervical cord. Theseprojections form the anatomic substrate for vestibulocollicreflexes, which serve to stabilize the head by appropriateaction of the sternocleidomastoid and other neck muscles. A

27

VESTIBULAR PROKECTIONARAOF CEREBRAL CORTEX

OCULOMOTOR, TROCHLEAR,AND AB-UCENS NUCL.EI MEDIAL LONGITUDINAL FASCICULUS

MEDULLARY RETICULARFORMATIONCREELROTX

VESTIBULAR EFFERENT FIBERS

VESTIBULAR NUCLEI

VESTIBULOSPINAL TRACTS -.

/ VESTIBULAR NERVE a,

VESTIBULAR GANGLION

Figure 7. Major connections and projections of the vestibularsystem.

third projection is that from primarily the lateral vestibularnucleus into the ventral gray throughout the length of thespinal cord. This important pathway is the uncrossed lateralveestibulosDinal tract, which enables the yrestibulospinal (RgAi-rural) K to help stabilize the body with respect to aninertial frame of reference by means of sustained and transientvestibular influences on basic spinal reflexes. Secondary a.

,estibulocerebellar fibers course from the vestibular nucleiinto the ipsilateral and contralateral fastigial nuclei and tothe cerebellar cortex of the flocculonodular lobe and elsewhere.Returning from the fastigial and other cerebellar nuclei,crossed and uncrossed fibers of the cerebellobulbar tractterminate in the vestibular nuclei and in the associated reticu-lar formation. There are also efferent fibers from the cerebel-lum, probably arising in the cerebellar cortex, which terminatenot in nuclear structures, but on dendritic endings of primaryvestibular afferent neurons in the vestibular neuroepithelia.Such fibers are those of the y. u e g re, whichappears to modulate or control the information arising from thevestibular end-organs. The primary and secondary vestibulo-cerebellar fibers and those returning from the cerebellum to thevestibular area of the brainstem comprise the juxtarestiform '4 of the inferior cerebellar peduncle. This structure, alongwith the vestibular end-organs, nuclei, and projection areas in

22%

28

the cerebellum, collectively constitute the so-called gstiblg- '

erebellar axis, the neural complex responsible for processingprimary spatial orientation information and initiating adaptiveand protective behavior based on that information.

Several additional projections, more obvious functionallythan anatomically, are those to certain autonomic nuclei of thebrain stem and to the cerebral cortex. The dosal motor nuIcleuQf cranial a&L& & (vagus) and other autonomic cell groups inthe medulla and pons receive secondary vestibular fibers,largely from the medial vestibular nucleus; these fibers mediateviestibulovegetative reflexes manifesting as the pallor, perspira-tion, nausea, and vomiting--motion sickness--that can resultfrom excessive or otherwise abnormal vestibular stimulation.Via vestibulothalamic and thalamocortical pathways, vestibularinformation eventually reaches the R/a v-- r e.-area of the cerebral cortex, currently thought to be located inthe parietal or parietoinsular cortex (not in the temporal lobe,as previously suspected). This projection ared is provided withvestibular, visual, and somatosensory proprioceptive representa-tion, and is evidently associated with conscious spatial orienta-tion and integration of sensory correlates of higher-order motoractivity. In addition, vestibular information can be transmittedvia long polysynaptic pathways through the brainstem reticularformation and medial thalamus to wide areas of the cerebralcortex; the nonspecific cortical responses to vestibular stimulithat are evoked via this pathway appear to be associated with anarousal or alerting mechanism.

Vestibular Information Processin"

As the reader should have deduced from the discussion ofthe anatomy of the vestibular end-organs, angular accelerationsare the adeguate. i.e.. physiologic, stimuli for the semiQircularducts. and linear -accelerations and gravity are the adequatestimuli for the otolith organs. This statement, pictorializedin Figure 8, is the cardinal principle of vestibular mechanics.How the reactive torques and gravitoinertial forces stimulatethe hair cells of the cristae and maculae, respectively, andproduce changes in frequency of action potentials in the asso-ciated vestibular neurons, has already been described. Theresulting frequency-coded messages are transmitted into theseveral central vestibular projection areas as raw orientationaldata, to be further processed as necessary for the variousfunctions served by such data. These functions are: thevestibular reflexes, voluntary movement, and perception oforientation.

Vestibular Reflexes. As stated so adequately by MelvillJones (13), "...for control of eye movement relative to spacethe motor outflow can operate on three fairly discrete anatomical 4 %

platforms, namely: (1) the eye-in-skull platform, driven by theexternal eye muscles rotating the eyeball relative to the skull;(2) the skull-on-body platform driven by the neck muscles; and

29

ANGULAR ACCELERATION

<1 LINEAR ACCELERATION AND GRAVITY

Figure 8. The cardinal principle of vestibular mechanics:angular accelerations stimulate the semicircularducts; linear accelerations and gravity stimulate theotolith organs.

(3) the body platform, operated by the complex neuromuscularmechanisms responsible for postural control.'

In humans the retinal image is stabilized mainly byvestibulo-ocular reflexes, primarily those of semicircular-ductorigin. A simple demonstration can help one appreciate thecontribution of vestibulo-ocular reflexes to retinal-imagestabilization. Holding the extended fingers half a meter or soin front of the face, one can move them slowly from side to sideand still see them clearly because of visual (optokinetic)tracking reflexes. But as the rate, or correspondingly, thefrequency, of movement becomes greater, one eventually reaches apoint where he is unable to see his fingers clearly--they areblurred by the movement. This point is at about 60 degrees persecond or one to two cycles per second for most people. Now ifone holds his fingers still, and rotates his head back and forthat the frequency at which the fingers became blurred when theywere moved, the fingers remain perfectly clear. Even at con-siderably higher frequencies of head movement, the vestibulo-ocular reflexes initiated by the resulting stimulation of the

30

%aS. . . . - • N . . . . . . . . . . . . o . -..

V

N.,

semicircular ducts function to keep the image of the fingersclear. Thus, at lower frequencies of movement of the externalworld relative to the body or vice versa, the visual systemstabilizes the retinal image by means of optokinetic reflexes.But as the frequencies of such relative movement become greater,the vestibular system, by means of vestibulo-ocular reflexes,assumes progressively more of this function; and at the higherfrequencies of relative motion characteristically generated onlyby motions of the head and body, the vestibular system is theone stabilizing the retinal image.

The mechanism by which stimulation of the semicircularducts results in retinal image stabilization is quite simple, atleast conceptually (Fig. 9). When the head is turned to theright in the horizontal (yaw) plane, the angular acceleration ofthe head creates a reactive torque in the ring of endolymph in(mainly) the horizontal semicircular duct. The reacting endo-lymph thereupon exerts pressure on the cupula, deviating thecupula in the right ear in a utriculopetal direction, depolar-izing the hair cells of the associated crista ampullaris, andincreasing the frequency of the action potentials in the corre-sponding ampullary nerve. In the left ear the endolymph deviatesthe cupula in a utriculofugal direction, thereby hyperpolarizingthe hair cells and decreasing the frequency of the actionpotentials generated. As excitatory neural signals are relayedto the contralateral lateral rectus and ipsilateral medialrectus muscles, and inhibitory signals are simultaneouslyrelayed to the antagonists, a conjugate deviation of the eyesresults from the described changes in ampullary neural activity.The direction of this conjugate eye deviation is thus the sameas that of the angular reaction of the endolymph, and theangular velocity of the deviation is proportional to the pressureexerted by the endolymph on the cupula. The resulting eye move-ment is therefore compensatory; i.e., it adjusts the angularposition of the eye to compensate for changes in angular posi-tion of the head, and thereby prevents slippage of the retinalimage over the retina. Because the amount of angular deviationof the eye is physically limited, very rapid movements of theeye in the direction opposite the compensatory motion areemployed to return the eye to its initial position or to advanceit to a position from which it can sustain a compePsatory sweepfor a suitable length of time. These very rapid eye movementsare anticompnsatorv, and because of their very high angularvelocity, we do not perceive motion during this phase of thevestibulo-ocular reflex.

With the usual rapid, high-frequency rotations of the head,the rotational inertia of the endolymph acts to deviate thecupula as the angular velocity of the head builds, and theangular momentum gained by the endolymph during the briefacceleration acts to drive the cupula back to its restingposition when the head decelerates to a stop. The cupula-endolymph system thus functions as an integrating angularaccelerometer; i.e. it converts angular acceleration data into

31

STPUATIOAR ANUA-CEERTO NUARACLRTO

HUAA

AM~O POTOMIAS - JlY

STATIONARY ANGULAR ACCELERATION ANGULAR ACCELERATIONTO RIGHT TO LEFT

Figure 9. Mechanism of action of a horizontal semicircular duct,and resulting reflex eye movement. Angular accelera-tion to the right increases the frequency of actionpotentials originating in the right ampullary nerveand decreases those in the left. This pattern ofneural signals causes extraocular muscles to rotatethe eyes in the direction opposite that of headrotation, thus stabilizing the retinal image with acompensatory eye movement. Angular acceleration to theleft has the opposite effect.

a neural signal proportional to the angular velocity of thehead. This is true for angular accelerations occurring atfrequencies normally encountered in terrestrial activities; whenangular accelerations outside the dynamic response range of thecupula-endolymph system are experienced, the system no longerprovides accurate angular velocity information. Wh1 en angularaccelerations are relatively sustained, or a cupula is kept in adeviated position by other means, such as caloric testing, thecompensatory and anticompensatory phases of the vestibulo-ocularreflex are repeated, resulting in beats of ocular n(Fig. 10). The compensatory phase of the vestibulo-ocularreflex is then called the slow pha of nystagmus, and theanticompensatory phase is the fast or g•ujk Rbe. The directionof the quick phase is used to label the direction of the nystag-mus, as the direction of the rapid motion of the eye is theeasier to determine clinically. The vertical semicircular ductsoperate in an analogous manner, with the vestibulo-ocularreflexes elicited by their stimulation being appropriate to theplane of the angular acceleration resulting in that stimulation.Thus, a vestibulo-ocular reflex with downward compensatory andupward anticompensatory phases results from stimulation of thevertical semicircular ducts by pitch-up (-a ) angular accelera-tion, and with sufficient stimulation in this plane, up-beating

32

verticil nystagmus results. Angular accelerations in the rollplane quite logically result in vestibulo-ocular reflexes withclockwise and counterclockwise compensatory and anticompensatoryphases, and in rotary nystagmus. Other planes of stimulationare associated with other directions of eye movement, such asoblique or horizonto-rotary.

.114%

RIGHT _..")

z0

I-1-

0~0

IL.ga! ON OFF n,

COUNTERCLOCKWISE

ANGULAR ACCELERATION

LEFT

TIME

Figure 10. Ocular nystagmus--repeating compensatory and anti-compensatory eye movements--resulting from vestibularstimulation. In this case the stimulation is a yawingangular acceleration to the left, and the anti-compensatory, or quick-phase, nystagmic response isalso to the left.

As should be expected, there are also vestibulo-ocularreflexes of otolith-organ origin. Initiating these reflexes arethe shearing actions that bend the cilia of macular hair cellsas inertial forces or gravity cause the otolithic membranes toslide to various positions over their maculae (Fig. 11) . Eachposition that can be assumed by an otolithic membrane relative % %to its macula evokes a particular spatia] pattern if frequencies %%of action potentials in the corresponding utricular or saccularnerve; and that pattern is associated with a particular set ofcompatible stimulus conditions, such as backward tilt of the

33

\ N ~. . . . . . - -. -. ,*.

UPRIGHT PSITION TILT BACKWARD TILT FORWAROo~ootm~a ... 17- "'..,-.

ENDOLYNM .-OTOlITHIC MEMBRANE -,.

NAIR CELLS Of MACULA

ACTION POTENTIALS

GRAVITY GRAVITY GRAVITY%-

%AFORIWARD ACCELFRAIION BACKWARD' ACCELERATION •

LINEAR INERTLIAL INERTIAL LINEARACCELIRATION FORCE fORCE ACCELERATION

p,..

Figure 11. Mechanism of action of an otolith organ. A change indirection of the force of gravity (above) or a linearacceleration (below) causes the otolithic membrane toshift its position with respect to its macula,thereby generating a new pattern of action potentialsin the utricular or saccular nerve. Shifting of theotolithic membranes can elicit compensatory vestibulo-ocular reflexes as well as perceptual effects.

head or forward linear acceleration. These patterns of actionpotentials from the various otolith organs are correlated andintegrated in the vestibular nuclei and cerebellum with orienta-tional information from the semicircular ducts and other sensorymodalities; appropriate orientational percepts and motor activi-ties eventually result. Lateral (a ) linear accelerations can Ielicit horizontal reflexive eye moveaents, including nystagmus,presumably as a result of utricular stimula'cion. Similarly,vertical (az) linear accelerations can elicit vertical eyemovements, most likely as a result of stimulation of the saccule;the term C reflex is sometimes used to describe thisresponse, as it is readily provoked by the vertical linearaccelerations associated with riding in an elevator. Theutility of these horizontal and vertical vestibulo-ocularreflexes of otolith-organ origin is readily apparent: like thereflexes of semicircular-duct origin, they help stabilize the Eretina) image. Less obvious is the usefulness of the ocu'lar'

~34

.b.. -J.

countertorsion reflex (Fig. 12), which repositions the eyesabout their visual (anteroposterior) axes in response to theotolith-organ stimulation resulting from tilting the headlaterally in the opposite direction. Presumably this refiexcontributes to retinal image stabilization by providing aresponse to changing directions of the force of gravity. .

°.4?OIIFi r .

Figure 12. Ocular countertorsion, a vestibulo-ocular reflex ofotolith-organ origin. When the head is tilted to the.left, the eyes rotate to the right to assume a newangular position about the visual axes, as shown.

Our understanding of the vestibulocollic reflexes has not

developed to the same degree as that of the vestibulo-ocularreflexes, although some clinical use has been made of measure-ments of rotation of the head on the neck in response to vestibu-lar stimulation. Perhaps this situation reflects the fact that,in humans, vestibulocollic reflexes are not as effective asvestibulo-ocular reflexes in stabilizing the retinal image.Such is not the case in other species, however; birds exhibitextremely effective reflex control of head position underconditions of bodily motion--even nystagmic head movements arequite easy to elicit. The high level of development of the

35

,•-"i'''f'•'] ... I .......................................................................... ".."...i.."i'

A

vestibulocollic reflexes in birds is certainly either a cause ora consequence of the relative immobility of birds' eyes in theirheads. Nonetheless, the ability of a human (or any othervertebrate with a mobile head) to keep his head upright withrespect to the direction of applicd gravitoinertial force ismaintained through tonic vestibular influences on the muscles ofthe neck,

Vestibulospinal reflexes operate to assure stability of thebody. Transient linear and angular accelerations, such as thoseexperienced in tripping and falling, provoke rapid activation ofvarious groups of extensor and flexor muscles, so as to returnthe body to the stable position or at least to minimize theultimate effect of the instability. We have all experienced thereflex arm movements that serve to break a fall, and haveobserved the more highly developed righting reflexes that catsexhibit when dropped from an upside-down position; these areexamples of vestibulospinal reflexes. Less spectacular, butnevertheless extremely important, are the sustained vestibularinfluences on posture, exerted through tonic activation ofso-called "antigravity" muscles such as hip and knee extensors.These vestibular reflexes, of course, help keep the body uprightwith respect to the direction of the force of gravity.

Volurptarv Movement. We have seen how the various reflexesof vestibular origin serve to stabilize the body in general andthe retinal image in particular. But the vestibular system isalso important in that it provides data for the proper executionof voluntary movement. To realize just how important suchvestibular data are in this context, we must first recognize thefact that skilled voluntary movements are balisti .-- i.e., onceinitiated, they are executed according to a predeterminedpattern and sequence, without the benefit of simultaneoussensory feedback to the higher neural levels from which theyoriginate. The simple act of writing one's signature, forexample, involves such rapid changes in speed and directi.on ofmovement that conscious sensory feedback and adjustment of motoractivity are virtually precluded, at least until the act is ."-

nearly completed. Learning an element of a skill thus involvesdeveloping a computer-program-like schedule of neural activationsthat can be called up, so to speak, to effect a particulardesired end product of motor activity. Of course, the rawprogram for a particular voluntary action is not sufficient topermit the execution of that action: information regarding such "1parameters as intended magnitude and direction of movement mustbe furnished from the conscious sphere, and data indicating theposition and motion of the body platform relative to the surfaceof the earth--i.e., spatial orientation information--must befurnished from the preconscious. The necessity for the addi-tional information is seen when we consider the signature-writingexample cited above, and realize that we can write large orsmall, quickly or slowly, and on a horizontal or vertical A .0surface. Obviously, different patterns of neuromuscular activa-tion, even grossly different muscle groups, are needed to

36I

accomplish a basic act under varying spatial and temporalconditions; but the necessary adjustments are made automatically,without conscious intervention. Vestibular and other sensorydata providing spatial orientation information for use in eitherskilled voluntary or retlexive motor activity are processed intoa Dreconscious orientational p that provides the informa-tional basis upon which such automatic adjustments are made.Thus, without having consciously to discern the direction -of theforce of gravity, analyze its potential effects on planned motoractivity, select appropriate muscle groups and -uodes of activa-tion to compensate for gravity, and then activatc and deactivate Keach muscle in proper sequence and with proper timing r*o accom-plish the desired motor activity, we simply decide what we wantthe outcome of our action to be and initiate the command to doit. Our body takes care of the details, using stored programsfor elements of skilled motor activity, and the current precon-scious orientational percept. This whole process is the majorfunction and responsibility of the vestibulo-cerebellar axis.

Conscious Percepts. Usually as a result of the sameinformation processing that provides the preconscious orienta-tional percept, one is also provided a o&ioug orientationalperp. This percept can be false, i.e., illusory, in whichcase the subject is said to experience an orientational illusion.(The term Rjlot vjrjgo refers to an illusory conscious orienta-tional percept experienced by a pilot. Although clinical use of "the word v usually, but by no means always, denotes apronounced sensation of turning, in aeromedical usage no suchcensation is implied--only a false conscious orientationalpercept.) One can be aware, moreover, that what his body istelling him about his spatial orientation is not what he hasconcluded from other information, such as flight instrumentdata. Conscious orientational percepts thus can be eitherAJL.UX.A1 or d, depending on the source of the orientation

inforination and the perceptual process involved; and one canexperience both natural and derived conscious orientationalpercepts at the same time. Because of this, pilots who havebecome disoriented in flight commonly exhibit vacillatingcontrol inputs, as they alternate indecisively between respondingfirst to one percept and then to the other.

TheshJolds of Vestibular PerceDtion. Often an orientationalillusion occurs because the physical event re.iulting in or froma change in bodily orientation is below the threshold of percep-tion. For that reason, the student of disorientation should beaware of the approximate perceptual thresholds associated withthe various modes of vestibular stimulation.

The lowest reported threshold for perception of rota-tion is 0.035 degrees/s 2, but this degree of sensitivity isobtained only with virtually continuous angular accelerationand long response latencies (20 to 40 seconds) (14). Other aobservations put the perceptual threshold between roughly0.1 and 2.0 degrees/s 2 ; reasonable values are 0.14, 0.5, and

37

," "* ., . ............................... . .'. . ..-. ..'" " , "" . - ," '. - -

0.5 degrees/s 2 for yaw, roll, and pitch motions, respec-tively (15). It is common practice, however, to describe thethresholds of the semicircular ducts in terms of the angularacceleration-time product, or angular velocity, which results injust perceptible rotat.i.un. This product remains fairly constantfor stimulus times of about five seconds or less, and is known asMulder's constant--even though the observed values of thisconstant range between 0.2 and 8.0 degrees/s, depending on thesubjects and methods used to determine it. Using the reasonablevalue of 2 degrees/s for Mulder's constant, we find that anangular acceleration of 5 degrees/s 2 applied for half a secondwould be perceived because the acceleration-time product is abovethe 2-degrees/s angular velocity threshold. But a 10-degrees/s 2

acceleration applied for a centh of a second would not beperceived, because it would be below the angular velocitythreshold; nor would a 0.2-degrees/s acceleration applied for5 s.

The perceptual threshold related to otolith-organ functionnecessarily involves both an angle and a magnitude, as theotolith organs respond to linear accelerations and gravito-inertial forces, both of which have direction and intensity. A1.5-degree change in direction of applied G force is perceptibleunder ideal (experimental) conditions. The minimum perceptibleintensity of linear acceleration has been reported by variousauthors to be between 0.001 and 0.03 g, depending on directionof acceleration and experimental method used. Values of 0.01 gfor az and 0.006 g for ax accelerations are appropriate represen-tative thresholds, and a similar value for a acceleration isprobably reasonable (15). Again, these absolute thresholdsapply when the acceleration is either sustained or applied atrelatively low frequencies. The threshold for linear accelera-tions applied for less than about 5 s is a more or less con-stant acceleration-time product, or linear velocity, of about0.3-0.4 m/s.

Unfortunately for those who would like to calculate exactlywhat orientational percept results from a particular set oflinear and angular accelerations, such as might have occurredprior to an aircraft mishap, the actual vestibular perceptualthresholds are, as expressed by one philosopher, "constantexcept when they vary." Probably the most common reason for anorientational perceptual threshold to be raised is inattentionto orientational cues because of attention to something else.Other reasons might be low state of mental arousal, fatigue,drug effects, or innate individual variation. Whatever thereason, it appears that a given individual can monitor hisorientation with considerable sensitivity under some circum-stances and relative insensitivity under others, which inconsis-tency can itself lead to perceptual errors that result inorientational illusions.

38

Of paramount importance in the generation of orientational '•-

illusions, however, is not the fact that absolute vestibularthresholds exist, or that vestibular thresholds are time-varying.Ratherp it is the fact that the components of the vestibularsystem, like those of any complex mechanical or electricalsystem, have characteristic frequency responses; and stimulationby patterns of acceleration outside the optimum, or "design," 0

frequency-response ranges of the semicircular ducts and otolithorgans causes the vestibular system to make errors. In flight,much of the stimulation resulting from the acceleratory environ-ment is indeed outside the design frequency-response ranges ofthe vestibular end-organs; consequently, orientational illusionsoccur in flight. Elucidation of this important point is providedin the section on spatial disorientation.

Vestibular Suppression and Enhancement. Like all sensorysystems, the vestibular system exhibits a decreased response tostimuli that are persistent ( •adataio) or repetitious (hbki -a--tigjl). Even more important to the aviator is the fact that,with time and practice, one can develop the ability to suppressnatural vestibular responses, both perceptual and motor. Thisability is termed y ±t jja uR~rgenion. Closely related tothe concept of vestibular suppression is that of visual di-nance, the ability to obtain and use spatial orientation cuesfrom the visual environment despite the presence of potentiallyvery strong vestibular cues. Vestibular suppression seems to beexerted, in fact, through visual dominance, as it disappears inthe absence of vision. The opposite effect, that of an increasein perceptual and motor resporaiveness to vestibular stimulation,is termed yti aJ e eg. Such enhancement can occurwhen the stimulation is novel, as in an amusement park ride, orthreatening, as in an aircraft spinning out of control, orwhenever spatial orientation is perceived to be especiallyimportant. It is tempting to attribute to the efferent vestibu-lar neurons the function of controlling the gain of the vestibu-lar system so as to effect suppression and enhancement, and someevidence exists to support that notion (16) . The actual mecha-nisms involved appear to be much more complex than would be Jnecessary to merely provide gross changes in gain of the vestibu-lar end-organs. Vtry precise control of veutibular responses toanticipated stimulation, based on sensory efferent copies ofvoluntary commands for movement, is probably exercised by thecerebellum via a feed-forward loop involving the vestibularefferent system. Thus, when discrepancies between anticipatedand actual stimulation generate a neural error signal, a responseis evoked, and vestibular reflexes and heightened perceptionoccur (17). Vestibular suppression, then, involves the develop-ment of accurate estimates of vestibular responses to orienta-tional stimuli repeatedly experienced, and the active counteringof anticipated responses by spatially and temporally patternedsensory efferent activity. Vestibular enhancement, on the otherhand, results from the lack of available estimates of vestibular r,.responses because of the novelty of the stimulation, or perhapsfromt a revision in neural processing strategy obligated by the

39

failure of normal negative feed-forward mechanisms to provideadequate orientation information. Such marvelous complexity ofvestibular function assures adaptability to a wide variety ofmotional environments, and thereby promotes survival in them.

4I0

I4

=_a.°rab

a,

p

-e

r

!a5

40 .3

OTHER SENSES OF MOTION AND POSITION

Although the visual and vestibular systems play dominantroles in spatial orientation, the contributions of other sensorysystems to orientation cannot be overlooked. Especially impor-tant are the nonvestibular proprioceptors--the muscle, tendon,and joint receptors--and the cutaneous exteroceptors, as theorientational percepts derived from their functioning in flightgenerally support those derived from vestibular informationprocessing, whether accurate or inaccurate. The utility ofthese other sensory modalities is appreciated when we realizethat, in the absence of vision, our vestibular, muscle, tendon,joint, and skin receptors allow us to maintain spatial orienta-tion and postural equilibrium, at least on the earth's surface.SimilaLly, in the absence of vestibular function, vision and theremaining proprioceptors and cutaneous mechanoreceptors aresufficient for orientation and balance. But when two componentsof this triad of orientational senses are absent or substantiallycompromised, we are unable to maintain sufficient spatialorientation to permit postural shability and effective locomo-tion. The following limited discussion of the nonvisual,nonvestibular, orientational sensory modalities is not meant toimply that they are either unstudied or uninteresting. On thecontrary, a large, fascinating body of knowledge has accumulatedin this area, and the reader is referred elsewhere for comprehen-sive reviews of this subject matter (18,19).

Nonvestibular Proprioceptor s

Sherrington's 'proprioceptive" or "self-sensing" sensorycategory includes the vestibular (or labyrinthine) , muscle,tendon, and joint senses. We commonly speak of proprioceptionas though it means only the nonvestibular components, however.

Muscle and Tendon Senses. Skeletal muscle contains witninit complex sensory end-organs, muscle spja•j (Fig. 13a). Theseend-organs are comprised mainly of small int u muscle fibersthat lie in parallel with the larger, ordinary, e Lusal musclefibers, and are enclosed over part of their length by a fluid-filled bag. The sensory innervation of these structures consistsmainly of large, rapidly conducting afferent neurons thatoriginate as primary (annulospiral) or secondary (flower-spray)endings on the intrafusal fibers and terminate in the spinal cordon anterior horn cells and interneurons. Stretching of theassociated extrafusal muscle results in an increase in frequencyof action potentials in the afferent nerve from the intrafusalfibers; contraction of the muscle results in a decrease orabsence of action potentials. The more interestinq aspect ofmuscle spindle function, however, is that the intrafusal musclefibers are innervated by motor neurons (gamma efferents and

41

r %.%.-

others) and can be stimulated to contract, thereby altering the -

afferent information arising from the spindle. Thus, the sensoryinput from the muscle spindles can be biased by descendinginfluences from higher neural centers, such as the vestibulo-cerebellar axis.

While the muscle spindles are structurally and functionallyin parallel with associated muscle groups and respond to changesin their length, the Gojgj. tjl.gn Qorgans (Fig. 13b) are function-ally in series with the muscles and respond to changes intension. A tendon organ consists of a fusiform bundle of small xtendon fascicles with intertwining neural elements, and islocated at the musculotendinous junction or wholly withintendon. Unlike that of the muscle spindle, its innervation isentirely afferent.

The major function of muscle spindles and tendon organs isto provide the sensory basis for uoaic (or mia strtg)-r e. These elementary spinal reflexes operate to stabilizea joint by providing, in response to an increase in length of amuscle and concomitant stimulation of its included spindles,monosynaptic excitation and contraction of the stretched agonist(e.g., extensor) muscle, and disynaptic inhibition and relaxationof its antagonist (e.g., flexor) through the action of aninhibitory interneuron. In addition, tension developed onassociated tendon organs results in disynaptic inhibition of theagonist muscle, thus recgulating the amount of contractiongenerated. The myotatic reflex mechanism is, in fact, at the dvery foundation of posture and locomotion. Modification of thisand other basic spinal reflexes by organized facilitatory orinhibitory intervention originating at higher neural levels,either through direct action on skeletomotor (alpha) neurons orthrough stimulation of fusimotor (primarily gamma) neurons tomuscle spindles, results in sustained postural equilibrium andother purposive motor behavicr. Some have speculated, moreover,that in certain types of spatial disorientation in flight, thisorganized modification of spinal reflexes is interrupted, ascerebral cortical control of motor activity is replaced by lowerbrainstem and spinal control. Perhaps the "frozen-on-the-controls" type of disorientation-induced deterioration of flyingability is a reflection of primitive reflexes made manifest bydisorganization of higher neural functions.

Despite the obvious importance of the muscle spindles andtendon organs in control of motor activity, there is littleevidence to indicate that their responding to orientationalstimuli (such as occur when one stands vertically in a 1-Genvironment) results in any corresponding conscious propriocep-tive percept (20). Nevertheless, it is known that the dorsalcolumn and other ascending spinal tracts carry muscle afferentinformation to "ll] r and thalami a y nul, and thenceto the cerebral se y gL21. Furthermore, extensive projec-tions into the cerebellum from the afferent terminations of themuscle spindles and tendon organs--via d and ventral

42 4 2 <"0

soinocerebellar tracts--ensure that proprioceptive infor-mationfrom these receptors is integrated with other orientationalinformation and is relayed to the vestibular nuclei, cerebralcortex, and elsewhere as needed.

j ipp

•d

.- ,

5'.

*,

5-

Figure 13. Some of the nonvestibular proprioceptive and cutaneousexteroceptive receptors subserving spatial orienta-tion. a. Muscle spindle, with central afferent(sensory) and more peripheral efferent (fusimotor)innervation shown. b. Golgi tendon organ. c. Lamel-lated, spray-type, and free-nerve-ending jointreceptors. d. Two of the many types of mechano-receptors found in skin: lamellated Pacinian cor-puscles and spray-type Ruffini corpuscles.

.5

Joint Sensation. In contrast to the situation with theso-called "muscle sense of position' discussed above, it is wellestablished that sensory information from the joints does reachconsciousness. In fact, the threshold for perception of jointmotion and position can be quite low: as low as 0.5 degrees forthe knee joint when moved at greater than 1.0 degrees/s. Thereceptors in the joints are of three types, as shown in Fig. 13c:

43

(1) lamellated or encapsulated taginian gorDuflcle-like end-organs; (2) spray-type structures, known as Ruffin-.-l.-."endings when found in joint capsules and •GJlj tendon organwhen found in ligaments; and (3) free nerve endings. ThePacinian corpuscle-like terminals are rapidly adapting and aresensitive to quick movement of the joint, whereas both of thespray-type endings are slowly adapting and serve to signal slowjoint movement and joint position. There is evidence thatpolysynaptic spinal reflexes can be elicited by stimulation ofjoint receptors, but their nature and extent are not well under-stood. Proprioceptive information from the joint receptors pro-jects via the dorsal funiculi eventually to the cerebral sensorycortex, and via the spinocerebellar tracts to the anterior lobeof the cerebellum.

One must not infer from this discussion that only muscles,tendons, and joints have proprioceptive sensory receptors. Bothlamellated and spray-type receptors, as well as free nerve end-ings, are found in fascia, aponeuroses, and other connectivetissues of the musculoskeletal system; and presumably theyprovide proprioceptive information to the central nervous system.

Cutaneous Exte roceDtors

The exteroceptors of the skin include the mechanoreceptors,which respond to touch and pressure; thermoreceptors, whichrespond to heat and cold; and nociceptors, which respond tonoxious mechanical and/or thermal events and give rise to sensa-tions of pain. Of the cutaneous exteroceptors, only the mechano-receptors contribute significantly to spatial orientation.

Mechanoreceptors. Quite a variety of receptors are involvedin cutaneous mechanoreception: spray-type &uffini ; -ls'1;lamellated Eaia and M Corupuscle; branched andstraight l t, Merkel celp, and free nerve end-ings (Fig. 13d) . The response patterns of the mechanoreceptorsare also numerous: eleven different types of response, varyingfrom high-frequency transient detection, through several modes ofvelocity detection, to more or less static displacement detec-tion, are recognized. Pacinian corpuscles and certain receptorsassociated with hair follicles are very rapidly adapting and have -the highest mechanical frequency responses, responding to sinu- .soidal skin displacements in the range of 50 to 400 Hz. They arethus well suited to monitor vibration and very transient touchstimuli. Ruffini corpuscles are slowly adapting and thereforerespond primarily to low-frequency, i.e., sustained, touch andpressure stimuli. Merkel cells appear to have a moderatelyslowly adapting response, making them suitable for monitoringstatic skin displacemeiit and velocity. Meissner corpuscles seemto detect primarily velocity of skin deformation. Other recep-tors provide other types of response, so as to complete the spec-trum of mechanical stimuli that we sense through the skin. Themechanical threshold for the touch receptors is quite low--lessthan 0.03 dynes/cm2 on the thumb, (In comparison with the

44

iS ~S~ArA '.rat ~sa * l l X l ~ -a X hlnl~l f lI~ lI- l 5 f -- -~o W --i - -- - -- r~

labyrinthine receptors subserving audition, however, thisthreshold is not so impressive; a 0-dB sound pressure levelrepresents 0.0002 dynes/cm2 -- more than a hundred times lower.)Afferent information from the described mechanoreceptors is con-veyed to the cerebral cortex mainly by way of the dorsal funiculiand medullary relay nuclei into the medial lemnisci and thalamo-cortical projections. The dorsal spinocerebellar and othertracts to the cerebellum provide the pathways by which cutaneousexteroceptive information reaches the cerebellum and is inte-grated with proprioceptive information from muscles, tendons,joints, and vestibular end-organs.

Auditory Orientation

On the surface of the earth, the ability to determine thedirection of a sound source can play an important role inspatial orientation. The major auditory cues providing thisability are differences between the ears in arrival times ofcongruent acoustic stimuli, and interaural differences inmagnitude and phase of arriving stimuli. A revolving soundsource can create a sense of self-rotation and even elicitreflex compensatory and anticompensatory eye movements calledaludiokinetic nt u. In aircraft, binaural sound localizationis of little use in spatial orientation because of high ambientnoise levels and absence of audible external sound sources.Pilots do extract some orientation information, however, fromthe auditory cues provided by the rush of air past the airframe:the sound frequencies and intensities characteristic of variousairspeels and angles of attack are recognized by the experiencedpilot, and he uses them in conjunction with other orientationinformation to create a percept of velocity and pitch attitudeof his aircraft. As aircraft have become more capable, however,and the pilot has become more insulated from such acousticstimuli, the importance of auditory orientation cues in flyinghas diminished.

41

"I.

'p,%

45

SPATIAL DISORIENTATION

The evolution of man saw him develop over millions of years .

as an aquatic, terrestrial, and even arboreal creature, but neveran aerial one. In this development he subjected himself to, andwas subjected to, many different varieties of transient motions,but not to the relatively sustained linear and angular acceler-ations commonly experienced in aviation. As a result, heacquired sensory systems well suited for maneuvering under hisown power on the surface of the earth but poorly suited forflying. Even the birds, whose primary mode of locomotion isflying, are unable to maintain spatial orientation and flysafely when deprived of vision by fog or cloud. Only bats seemto have developed the ability to fly without vision, and thenonly by replacing vision with auditory echolocation. Consideringman's phylogenetic heritage, we should not be surprised that hissudden entry into the aerial environment results in a mismatchbetween the orientational demands of the new environment and hisinnate ability to orient. The manifestation of this mismatch isspatial lisorientation.

Illusions in Flight

An illusion is a false percept. An orientational J 1-j=is a false percept of one's position, attitude, or motion,relative to the plane of the earth's surface. Thus, misper-ceptions of displacement, velocity, or acceleration--eitherlinear or angular--result in orientational illusions. A greatvariety of orientational illusions occur during flight: somenamed, some unnamed; some understood, some not understood.Those that are sufficiently impressive to cause pilots to reportthem, whether because of their repeatability or because of theiremotional impact, have been described in the aeromedical litera-ture, and will be discussed here. We categorize the illusionsin flight into those resulting primarily from visual misper-ceptions and those involving primarily vestibular errors.

Visual Illusions. There are two different modes of visualprocessing, the focal (foveal) mode and the ambient (peripheral)mode. Generally speaking, focal vision is concerned with objectrecognition and ambient vision is concerned with spatial orien-tation. But some visual orientation tasks in flying--e.g.,approach to landing--require a great deal of focal vision, andsuch tasks can be made extremely difficult by illusions resultingfrom misinterpretation of focal visual cues. The ru il.1u-1.s and a illusions are illustrative of this (21).

Figure 14a shows the pilot's view of the runway during an Napproach to landing, and demonstrates the linear perspective andforeshortening of the runwdy that the pilot associates with a

46

3-degree approach slope. If the runway slopes upward 1 degree(a rise of only 35 m in a 2-km runway), the foreshortening ofthe runway for a pilot on a 3-degree approach slope is sub-stantially less (the height of the retinal image of the runwayis greater) than it would be if the runway were level. This cangive the illusion of being too high on the approach. The pilot's % %

natural response to such an illusion is to reshape his image of %the runway by seeking a shallower approach slope (Fig. 14b).This response, of course, could be hazardous. The oppositesituation results when the runway slopes downward. To perceivethe accustomed runway shape under this condition, the pilot mustfly a steeper approach slope than usual (Fig. i4c).

) .

a

-5'

Figure 14. Effect of runway slope on pilot's image of runwayduring final approach (left) and potential effect onapproach slope angle flown (right), a. Flat runway--normal approach. b. -psloping runway creates

illusion of being high on approach--pilot fliesapproach too low. c. Downsloping runway has opposite

A runway that is narrower than that to which a pilot isaccustomed can also create a hazardous illusion on the approachto landing. The pilot tends to perceive the narrow runway to belonger and farther away (i.e., that he is higher) than is

4.7

actually the case, and hu may flare too late and touch downsooner than he expects (Fig. 15b). Likewise, a runway that iswider than what a pilot is used to can lead him to believe heis closer to the runway (i.e., lower) than he really is, and hemay flare too soon and drop in from too high above the runway(Fig. 15c) . Both of these runway-width illusions are especiallytroublesome at night, when peripheral visual orientation cuesare largely absent. The very common tendency for pilots toflare too high at night results at least partly from the factthat the runway lights, being displaced laterally from theactual edge of the runway, make the runway seem wider, andtherefore closer, than it actually is.

a

b4

Figure 15. Effect of runway width on pilot's image of runway(left) and potential effect on approach flown(right). a. Accustomed width--normal approach.b. Narrow runway makes pilot feel he is higner thanhe actually is, so he flies approach too low andflares too late. c. Wide runway appears closer thanit actually is--pilot tends to approach too high andflare too soon.

The slope and composition of the terrain under the approachpath can also influence the pilot's judgment of his height abovethe touchdown point. If the terrain descends to the approach

43

end of the runway, the pilot tends to fly a steeper approach thanhe would if the approach terrain were level (Fig. 16a). If theapproach terrain slopes up to the runway, on the other hand, thepilot tends to fly a less steep approach than he would otherwise(Fig. 16b). Although estimation of height above the approachterrain depends on both focal and ambient vision, the contribu-tion of focal vision is particularly clear: consider the pilotwho looks at a building below him, and seeing it to be closerthan such buildings usually are, he seeks a higher approachslope. By the same token, focal vision and size constancy areresponsible for poor height and distance judgments pilots some-times make when flying over terrain having an unfrniliar composi-tion (Fig. 17). A reported example of this is the tendency tomisjudge approach height when landing in the Aleutians, where theevergreen trees are much smaller than those to which most pilotsare accustomed. Such height-estimation difficulties are by nomeans restricted to the approach and landing phases of flight.One fatal mishap occurred during air combat training over theSouthwest Desert, when the pilot of a high-performance fighterapparently misjudged his height over the desert floor because ofthe small, sparse vegetation, and was unable to arrest in timehis deliberate descent to a ground-hugging altitude.

'•-'

-'i"

Figure 16. Potential effect of slope of terrain under theapproach on approach slope flown, a. Tertain slopesdown to runway; pilot perceives himself to be tooshallow on approach and steepens it. b. Upsiopingterrain makes pilot think he is too high, so hecorrects by making approach too shallow.

49

............................. A .*. . . _ I . J II -_,.

a

U~ '---

Figure 17. Potential effect of unfamiliar composition ofapproach terrain on approach slope flown, a. Normalapproach over trees of familiar size. b. Unusuallysmall trees under approach path make pilot think heis too high, so he makes his approach lower thanusual.

A well-known pair of approach-to-landing situations thatcreate illusions because of the absence of adequate focal visualorientation cues are the smooth-water (or glassy-water) andsnow-covered approaches. A seaplane pilot's perception ofheight is degraded substantially when the water below is still:for that reason he routinely just sets up a safe descent rateand waits for the seaplane to touch down, rather than attemptingto flare to a landing, when the water is smooth. A blanket offresh snow on the ground also deprives the pilot of visual cueswith which to estimate his height, thus making his approach moredifficult--extremely difficult if the runway is also coveredwith snow. Again, approaches are not the only regime in whichsmooth water and fresh snow cause problems. A number of aircraft

have crashed as a result of pilots' maneuvering over smoothwater or snow-covered ground and misjudging their height abovethe water or ground.

Aerial perspective can also play a role in deceiving thepilot. In daytime, fog or haze can make a runway appear fartheraway as a result of the loss of visual dircrimination. At night,runway and approach lights in fog or rain appear less bright thanthey do in clear weather, and can create the illusion that theyare farther away. It has even been reported that a pilot canhave an illusion of banking to the right, for example, if therunway lights are brighter on the right side of the runway than

A A50

they are on the left. Another hazardous illusion of this typecan occur during approach to landing in a shallow fog or haze,especially during a night approach. The vertical visibility yunder such conditions is much better than the horizontal visi-bility, so that descent into the fog causes the more distantapproach or runway lights to diminish in intensity, at the sametiue that peripheral visual cues are suddenly occluded by thefog. The result is an illusion that the aircraft has pitched up,with the concomitant danger of a nose-down corrective action bythe pilot.

Another condition in which a pilot is apt to make a seriousmisjudgment is in closing on another aircraft at high speed.When he has numerous peripheral visual cues by which to establishhis own position and velocity relative to the earth and thetarget's position and velocity relative to the earth, histracking and closing problem is not much different from what itwould be on the ground if he were giving chase to a movingquarry. But when relative position and closure rate cues mustcome from foveal vision alone, as is generally the case ataltitude, at night, or under other conditions of reduced visi-bility, the tracking and closing problem is much more difficult.An overshoot--or worse, a midair collision--can easily resultfrom the perceptual difficulties inherent under such circum-stances, especially when the pilot lacks experience in theenvironment devoid of peripheral visual cues.

Two runway approach conditions that create considerabledifficulty for the pilot, and which, like the closure-rateproblems discussed above, result frozn tasking focal vision toaccomplish by itself what is normally accomplished with bothfocal and ambient vision, are the black-hole and whiteoutapproaches. A black-hole approach is one that is made on a darknight over water or unlighted terrain to a runway beyond whichthe horizon is indiscernible, the worst case being when only therunway lights are visible (Fig. 18). Without peripheral visualcues to help him orient himself relative to the earth, the pilottends to feel he is stable and situated appropriately, but thatthe runway itself moves about or remains malpositioned (isdownsloping, for example). Such illusions make the black-holeapproach difficult and dangerous, and often result in a landingfar short of the runway. A particularly hazardous type ofblack-hole approach is one made under conditions wherein theearth is totally dark except for the runway and the lights of acity on rising terrain beyond the runway. It has been observedthat under these conditions the pilot tries to maintain aconstant vertical visual angle for the distant city lights, thuscausing his aircraft to arc far below the intended approachslope as he gets closer to the runway (Fig. 19) (22). Analternative explanation is thý.t the pilot falsely perceivesthrough ambient vision that the rising terrain is flat, and helowers his approach slope accordingly.

51

,- ". . . . .. .. .. . . . .,..... S B * . . U -I .. ..-. . .IUI... i -" " U • • ' i ' i '

4.:

- . o,

Figure 18. Effect of loss of peripheral visual orientation cueson perception of runway orientation during a black-hole approach. Above: When peripheral visualorientation cues are absent, pilot feels upright and(in this example) perceives runway to be tilted leftand upsloping. Below: With horizon visible, pilotorients himself correctly with peripheral visionand runway appears horizontal in central vision.

. .0

0 .

0 ,[

0!

Figure 19. A common and particularly dangerous type of black-hole approach, in which pilot perceives distant cityto be flat and arcs below desired approach slope.

An approach made under whiteout conditions can be asdifficult as a black-hole approach, and for essentially the samereason--lack of sufficient ambient visual orientation cues.There are actually two types of whiteout, the atmg&I~her.whte~out and the blwiigsn whtot In the atmosphericwhiteout, a snow-covered ground merges with a white overcast,creating a condition in which ground textural cues are absentand the horizon is indistinguishable. Although visibility maybe unrvbtxicted in the atmroEpheric whiteout, there is essentiallynothing to see except the runway or runway markers; an approachmade in this condition must. therefore be accomplished with aclose eye on the altitude and attitude instruments to preventspatial disorientation arid inadvertent ground contact. In theblowing-6now whiteout, visibility is restc:icted drastically bysnowflakes, and often those snowflakes have been driven into theair by the propeller or rutor wash of the affected aircraft.Helicopter landings on bnow-covered ground are particularlylikely to create blowing-snow whiteout. Typically, the heli-COptL'r L ji l ot t r i es t o nia i nt a in v i ,ualI conita ct w ith thle y r ound

r3""°

during the sudden rotor-induced whiteout, gets into an unrecog-nized drift to one side, and shortly thereafter contacts theground with sufficient lateral motion to cause the craft toroll over. Pilots flying where whiteouts can occur must be madeaware of the hazards of whiteout approaches, as the disorien-tation induced usually occurs unexpectedly under visual ratherthan instrument meteorologic conditions.

One puzzling illusion that occurs when visual orientationcues are minimal is vigu.i ainesis . (Fig. 20) (21). A small,dim light seen against a dark background is an ideal stimulusfor producing autokinesis. After 6 to 12 s of visually fixatingthe light, one can observe it to move at anywhere between 0.2and 20 degrees/s, in one particular direction or in severaldirections in succession. Peripheral visual autokinesis isassociated with smooth apparent movements and large apparentdisplacements, while central visual autokinesis is often saccadicor jerky, and results in little apparent displacement of theobject fixated. In general, the larger the object and thebrighter the object the less the autokinetic effect. The shapeof the object seems to have little effect on the magnitude ofthe illusion, however. Nor does providing a larger number ofobjects necessarily reduce the illusory effect, as multipleobjects can appear to move, either as a unit or independently,as vigorously as one alone. The physiologic mechanism of visualautokinesis is not understood; in fact, it is not even estab-lished with certainty whether actual eye movements are associatedwith autokinesis. One suggested explanation for the autokineticphenomenon is that the eyes tend to drift involuntarily, perhapsbecause of inadequate or inappropriate vestibular stabilization,and checking the drift requires efferent oculomotor activityhaving sensory correlates that create the illusion.

Whatever the mechanism, the effect of visual autokinesis onpilots is of great importance. Anecdotes abound of pilots whofixate a star or a stationary ground light at night, and seeingit move because of autokinesis, mistake it for another aircraftand try to intercept or join up with it. Another untowardeffect of the illusion occurs when a pilot flying at nightperceives a relatively stable aircraft--one which he mustintercept or follow--to be moving erratically, when in fact itis not; the unnecessary and undesirable control inputs the pilotmakes to compensate for the illusory movement of the targetaircraft represent increased work and wasted motion at best, andan operational hazaid at worst.

To help avoid or reduce the autokinetic illusion, oneshould try to maintain a well-structured visual environment iiwhich spatial orientation is unambiguous. Since this is rarelypossible in night flying, it has been suggested that: (1) thegaze should be shifted frequently to avoid prolonged fixation ofa target light; (2) the target should be viewed beside orthrough, and in reference to, a relatively stationary structuresuch as a canopy bow; (3) one should make eye, head, and body

54

movements to try to destroy the illusion; and (4) as always, one %should monitor the flight instruments to help prevent or resolveany perceptual conflict. Equipping aircraft with more than onelight or with luminescent strips to enhance recognition at nighthas probably helped reduce problems with autokinesis. It hasnot abolished them, however. -..

".'

4,R'

a-..

Figure 20. Visual autokinesis. A small, solitary light or ,•small group of lights seen in the dark can appear tomove, when in fact they are stationary.:-

So far we have discussed visual illusions created byexcessive orientation-processing demands being placed on focal •vision when adequate orientation cues are not available through--ambient vision, or by strong but false orientation cues being •:received through focal vision. But ambient vision can itself be--respons~ible for creating orientational illusions whenever orien- .•tation cues received in the visual periphery are misleading ormisinterpreted. Probably the most compelling of such illusions %Z%are the vection illusions (4). Vection is the visually inducedperception of motion of the self in the spatial environment-•

55

(self-motion), and can be a sensation of linear motion (linearvection) or angular motion (angular vection).

Nearly everyone who drives an automobile has experiencedone very common linear vection illusion: when a driver iswaiting in his car at a stop light and a presumably stationarybus or truck in the adjacent lane creeps forward, a compellingillusion that his own car is creeping backward can result(prompting a swift but surprisingly ineffectual stomp on thebrake). Another example is that of a passenger sitting in astationary train when the train on the adjacent track begins tomove: he can experience the strong sensation that his own trainis moving in the opposite direction (Fig. 21a). Linear vectionis one of the factors that make close formation flying sodifficult, as the pilot can never be sure whether his ownaircraft or that of his lead or wingman is responsible for therelative motion of his aircraft.

Angular vection occurs when peripheral visual cues conveythe information that one is rotating; and the perceived rotationcan be in pitch, roll, yaw, or any other plane. Althoughangular vection illusions are not common in everyday life, theycan be generated readily in a laboratory by enclosing a station-ary subject in a rotating striped drum. Usually within 10 safter the visual motion begins, the subject perceives that he,rather than the striped drum, is rotating. A pilot can experi-ence angular vectiot. if the rotating anticollision light on hisaircraft is left on during flight through clouds or fog- therevolving reflection provides a strong ambient visual stimulussignalling rotation in the yaw plane. Another condition re-sulting in &nguler vection is that in which a just-activatedlanding light rotates forward under the wing or nose, casting anupward-moving area of illumination in the surrounding fog orcloud, possibly even creating a rising false horizon. As aresult of the illusory pitch vection generated, the pilot couldbe tempted to raise the nose of the aircraft at a most inappro-priate time.

Fortunately, the vection illusions are not all bad. Themost advanced flight simulators depend on linear and angularvection to create the illusion of flight (Fig. 21b) . When thevisual flight environment is dynamically portrayed in wide-field-of-view flight simulators, the illusion of actual flight is socomplete and compelling that additional mechanical motion isrendered superfluous.

Another result of false ambient visual orientational cueingis the lean-Qn-the-suo illusion. On the ground, we are accus-towed to seeing the brighter visual surround above and thedarker below, regard2ess of the position of the sun. Thedirection of this gradient in light iitensity thus helps usorient with respect to the surface of the earth. In cloud,however, such a gradient usually does not exist; and when itdoes, the lighter direction is generally toward the sun and the

56

darker away from it. But the sun is almost never directlyoverhead; as a consequence, a pilot flying in a thin cloud layertends to perceive falsely the direction of the sun as directlyoverhead, and to align his aircraft accordingly. This misper-ception causes him to bank in the direction of the sun--whencethe name of the illusion. A variant of this phenomenon involvesa somewhat different mechanism: sometimes a pilot remembers therelative bearing of the sun when he first penetrated the weather,and he unconsciously tries to keep the sun in the same relativeposition whenever it peeks through the cloud. On a prolongedflight in intermittent weather, the changing position of the sunin the sky can cause the pilot to become mildly confused and flyhis aircraft with less precision than he would in either continu-ously visual or continuously instrument meteorologic conditions.

"Fgr 21. Veto/ilsin. .Lner"eto. Inti. eape

7/.

a b

Figure 21. Vection illusions, a. Linear vection. In this example, 4

adjacent vehicle seen moving aft in subject's periph-eral vision causes him to feel he is moving forward.b. Angular vection. Visually perceived objectsrevolving around subject in flight simulator cause himto sense self-rotation in opposite direction--in thiscase a rolling motion to the right.

Often the horizon perceived through ambient vision is not '-

really horizontal. Quite naturally, this misperception of thehorizontal creates hazards to flight. A sloping cloud deck, forexample, is very difficult to perceive as anything but horizontalif it extends for any great distance in the pilot's peripheralvision (Fig. 22). Uniformly sloping terrain can also create an

57

illusion of horizontality, with disastrouc consequences for thepilot thus deceived. Many aircraft have crashed as a result ofthe pilot's entering a canyon with an apparently level floorthat actually rises faster than the airplane can climb; thepilot arrives at the end of the canyon "out of altitude, air-speed, and ideas," as they say. Sometimes a distant rain showercan obscure the real horizon and create the impression of ahorizon at the proximal edge (base) of the rainfall. If theshower is seen just beyond the runway during an approach tolanding, the pilot can misjudge the pitch attitude of his air-craft and make inappropriate pitch corrections on the approach.

PNJ.

p.,.

.'o

Figure 22. A sloping cloud deck, which the pilot misperceivesas a horizontal surface.

Pilots are especially susceptible to misperception of thehoiizontal when flying at night (Fig. 23). Isolated groundlights can appear to the pilot to be stars, causing him to thinkhe is in a nose-high or one-wing-low attitude. Correcting forsuch a false impression can, of course, be fatal. Frequently nostars arc visible because of an overcast. Unlighted areas ofterrain can then blend with the dark overcast to create theillusion that the unlighted terrain is part of the sky. One

58 0

U'A

extremely hazardous situation is that in which a takeoff is madeover an ocean or other large body of water that cannot bedistinguished visually from the night sky. Many pilots in thissituation have perceived the shoreline receding beneath them tobe the horizon and some have reacted to this false percept withdisastrous consequences.

Pilots flying at very high altitudes can sometimes experi-

ence difficulties with control of aircraft attitude, because athigh altitudes the horizon is lower with respect to the plane oflevel flight than it is at the lower altitudes where most pilotsusually fly. As a reasonable approximation, the angle of depres-sion of the horizon in degrees equals the square root of thealtitude in kilometers. A pilot flying at an altitude of 15 kmthus sees the horizon almost 4 degrees below the extension ofhis horizontal plane. If he visually orients to the view fromhis left cockpit window, he might be inclined to fly with theleft wing 4 degrees down to level it with the horizon. If hedoes this, and *hen looks out his right window, he would see theright wing 8 degrees above the horizon, with half of thatelevation due to his own erroneous control input. He might alsoexperience problems with pitch control, as the depressed horizoncan cause him to perceive falsely a 4-degree nose-high pitchattitude.

Finally, the disorienting effects of the northern lightsand of aerial flares should be mentioned. Aerial refueling atnight in high northern latitudes is often made quite difficultby the northern lights, which provide false cues of verticalityto the pilot's peripheral vision. In addition, the movement ofthe auroral display may make the pilot susceptible to vectionillusions. Similarly, when aerial flares are dropped, they maydescend vertically--or they may drift with the wind, creatingfalse cues of verticality. Their motion may also create vectionillusions. An important additional factor is that the auroraand aerial flares can be so bright ds to reduce the apparentintensity, and therefore the orientational cueing strength, ofthe aircraft instrument displays.

Vestibular Illusions. The vestibulocerebellar axis pro-cesses orientation information from the vestibular end-organs,the nonvestibular proprioceptors, and the peripheral visualfields. In the absence of adequate ambient visual orientationcues, the inadequacies of the vestibular and other orientingsenses can, and generally do, result in orientational illusions.It is convenient and conventional to discuss the vestibularillusions in relation to the two labyrinthine components thatgenerate them---the semicircular ducts and the otolith organs.

The somatogyral illusion results from the inability of thesemicircular ducts to register accurately a prolonged rotation,i.e., sustained angular velocity. When a person is subjected toan angular acceleration about the yaw axis, for example, theangular motion ,- at first perceived accurately, because thedynamics of the cupula-endolymph system cause it to respond as

5SQ

01060

.... . . . . .

Figure 23. Misperception of the horizontal at night. Above:Ground lights appearing to be stars cause earth andsky to blend and a false horizon to be perceived.Below: Blending of overcast sky with unlightedterrain or water causes horizon to appear lower thanis actually the case.

an integrating angular accelerometer (i.e., as a rotation ratesensor) at stimulus frequencies in the physiologic range(Fig. 24). If the acceleration is followed immediately by adeceleration, as usually happens in the terrestrial environment,the total sensation of turning one way and then stopping the turnis quite accurate (Fig. 25). But if the angular acceleration isnot followed by a deceleration, and a constant angular velocityresults instead, the sensation of rotation becomes less and lessand eventually disappears as the cupula gradually returns to itsresting position in the absence of an angular acceleratorystimulus (Figs. 25, 26). If the rotating subject is subsequentlysubjected to an angular deceleration after a period of prolongedconstant angular velocity---say, after 10 s or more I constant-rate tur: ng--his cupula-endolymph system then signais a turn inthe direction opposite that of the prolonged constant angkilarvelocity, even though he is really only turning less rapidly inthe same direction. This sensory response occurs because theangular momentum of the rotating endolymph causes it to pressagainst the cupula, forcing the cupula to deviate in the direc-tion of endolymph flow, *.hich is the same direction the cupulawould deviate if the subject were to accelerate in the directicnopposite his initial acceleration. Even after rotation actuallyceases, the sensation of rotation in the direction opposite thatof the sustained angular velocity persists for several seconds--half a minute or longer with a large decelerating rotationalimpulse. Technically speaking, a somatogyral illusion is thesensation of turning in the opposite direction that occurswhenever one undergoes angular deceleration from a condition ofconstant angular velocity. It is practical, however, to includein this category of illusions the sensation of turning mcreslowly and eventually ceasing to turn while angular velocitypersists, because the two illusions have a conj,,on underlyingmechanism and they inevitably occur in pairs. An even broaderdefinition of somatogyral illusion i.e "any discrepancy betweenactual and perceived rate of self-rotation that results from anabnormal angular acceleratory stimulus pattern." The terr"abnorrral" in this case implies the application of low-frequencystimul;, outside the useful portion of the transfer character-istics of the semnicircular duct system.

U I

+10 - ____k1

'dB)

+45 -- P

PHASEANGLE 0$

, W1_______(degrees) :1

-45 1 I- Io I v)t

0.01 01 1.0FREQUENCY (Hz)

Figure 24. Transfer characteristics of ''-semicircular ductsystem as a function of sinu" .dal stimulus fre-quency. Gain is the ratio of t, magnitude of peakperceived angular velocity to peak delivered angularvelocity; phase angle is a measure of the amount of Iadvance or delay between peak perceived and peakdelivered angular velocity. Note that in the physio-logic frequency range (roughly 0.05 to 1 Hz), percep-tion is acc',rate; i.e., gain is close to unity (0 dB)and phase shift is minimal. At lower stimulusfrequencies, however, the gain drops off rapidly and .4the phase shift approaches 90 degrees, which meansthat angular velocity becomes difficult to detect and -

tha', angular acceleration is perceived as velocity.(Adapted from Peters (23).)

62 4

- U * ,.,* . 4 * * * .* * * ... .. '. -* U .. ' * .'. 5 * E r C . C iS"-. - .. °,-

., ..'

/a b c dIPerceived +Angular 0Velocity

Cupular +0 -- _ "_'

Deviation -

Angular + -0 __ ,Lw

Velocity

+Angular 0 -"_";

Acceleration -

0 4 0 15 30 456

Time (seconds)

Figure 25. Effect of stimulus pattern on perception of angularvelocity. On the left, the high-frequency character ,of the applied angular acceleration results in acupular deviation that is nearly proportional, and Aperceived angular velocity that is nearly identical,to the angular velocity developed. On the right, thepeak angular velocity developed is the same as that .on the left, but the low-frequency character of theapplied acceleration results in cupular deviation andperceived angular velocity that appear more like theapplied acceleration than the resulting velocity.This causes one to perceive: (a) less than the fullamount of the angular velocity, (b) that he ir not -.turning when he actually is, (c) a turn in the oppo-site direction from that of the actual turn, (d) that ..

turning persists after it has actually stopped.These false perceptions are somatogyral illusionL.

I ... *- "

COKSTANT ANGULAR

STATIONARY ANGULAR ACCELtRAT VELOCITY

ANGULAR OECELERATION STATIONARY

Figure 26. Pictorial representation of mechanical events occur-ring in a semicircular duct and resulting actionpotentials in associated ampullary nerve duringsomatogyral illusions. Angular acceleration patternapplied is shown in right side of Figure 25.

In flight under conditions of reduced visibility (night orinstrument weather), somatogyral illusions can be deadly. The

LL is the classic example of how somatogyral illu-sions can disorient a pilot, with fatal results. This situationbegins with the pilot intentionally or uninter.-ionally enteringa spin, let's say to the left (Fig. 27). At first the pilotperceives the spin correctly, as the angular accelerationassociated with entering the spin deviates the appropriatecupulae the appropriate amount in the appropriate direction.The longer the spin persists, however, the more the sensation ofspinning to the left diminishes as the cupulae return to theirresting positions. If the pilot tries to stop the spin to theleft by applying right rudder, the angLlar deceleration causes -

him to perceive a spin to the right, even though the only realresult of his action is termination of the spin to the left. Apilot who is ignorant of the possibility of such an il.usion is -4then likely to make counterproductive left rudd-r inputs tonegate the unwanted erroneous sensation of spinning to theright. These inputs keep the airplane spinning to the left,which gives the pilot the desired sensation of not spinning butdoes not bring the airplane under control. To extricate himselffrom this very hazardous situation the pilot must read theaircraft flight instruments and apply control inputs to make the

64

*19

• 4.

• q

• oa

Figure 27. The graveyard spin. After several turns of a spin thepilot begins to lose the sensation of spinning. Thenwhen lie tries to stop the spin, the resulting somato- -

gyral illusion of spinning in the opposite directionmakes him reenter the original spin. (Solid Iline 4indicates actual motion; dotted line indicatespeLceived motion.)

," .... ... . ... '-, ' i .. ... ........ , ", ......... , i""i1........ ..,i i1 "'

instruments give the desired readings (push right rudder tocenter the turn needle, in this example). Unfortunately, thismay riot be so easy to do. The angular accelerations created byboth the multiple-turn spin and the pilot's spin-recoveryattempts can elicit strong but inappropriate vestibulo-ocularreflexes, including nystagmus. In the usual terrestrial environ- %ment these reflexes help stabilize the retinal image of thevisual surround; but in this situation they only destabilize it,because the visual surround (cockpit) is already fixed withSrespect to the pilot. Reading the flight instruments thusbecomes difficult or impossible in this situation, and the pilotis left with only his false sensations of rotation to rely onfor spatial orientation and aircraft control (24).

While the lore of early aviation provided the graveyardspin as an illustration of the hazardous nature of somatogyralillusions, a much more common example occurring all too often inpresent-day aviation is the graveyard spiral (Fig. 28). In thissituation the pilot has intentionally or unintentionally gottenhimself into a prolonged turn with a moderate amount of bank.After a number of seconds in the turn the pilot loses thesensation of turning because his cupula-endolymph system cannotrespond to constant angular velocity. The percept of being in abank as a result of the initial roll into the banked attitudealso decays with time, because the net gravitoinertial forcevector points toward the floor of the aircraft during coc-dinatedflight (whether the aircraft is in a banked turn or flyingstraight and level) and the otolith organs and other graviceptorsnormally signal that down is in the direction of the net sus-tained gravitoinertial force. As a result, when the pilot triesto stop the turn by rolling back to a wings-level attitude, henot only feels he is turning in the direction opposite that of

Figure 28. The graveyard spiral. The pilot in a banked turnloses the sensation of being banked and turning.Upon trying to reestablish a wings-level attitudeand stop the turn, he perceives that he is bankedand turning in the opposite direction from hisoriginal banked turn; i.e., he experiences a somato-gyral illusion. Unable to tolerate the sensationthat he is making an inappropriate control input, thepilot banks back into the original turn.

" .

0 '•

§ O.., .4.

\ $4•

0•

" ~0

t,, /-7 0

his original turn, but also feels he is banked in the directionopposite that of his original bank. Unwilling to accept thissensation of making the wrong control input, the hapless pilotrolls back into his original banked turn. Now his sensation iscompatible with his desired mode of flight, but his instruments

say he is losing altitude (because the banked turn is wastinglift) and still turning. So he pulls back on the stick andperhaps adds power to arrest the unwanted descent and regain thelost altitude. This action would be successful if the aircraftwere flying wings-level, but with the aircraft in a bankedattitude it tightens the turn, serving only to make mattersworse. Unless the pilot eventually recognizes his error androlls out of his unperceived banked turn, he will continue todescend in an ever-tightening spiral toward the ground--whencethe name, graveyazd spiral.

Whereas a somatogyral illusion is a false sensation, orlack of sensation, of self-rotation in a subject undergoingunusual anjular motion, an oculogxral illusion is a falsesensation of motion of an object viewed by such a subject (25)For example: if a vehicle with a subject inside is rotatingabout a vertical axis at a constant velocity and suddenly stopsrotating, the subject experiences not only a somatogyral illusionof himself rotating in the opposite direction, but also anoculogyral illusion of an object in front of him moving in theopposite direction. Thus, a somewhat oversimplified definitionof the oculogyral illusion is that it is the visual correlate ofthe somatogyral illusion; but its very low threshold and itslack of total correspondence with presumed cupular deviationsuggest a more complex mechanism. The attempt to maintainvisual fixation during a vestibulo-ocular reflex elicited byangular acceleration is probably at least partially responsiblefor the oculogyral illusion. (A similar mechanism underlies theillusory movement of the moon when one tries to fixate itvisually while the relative movement of surrounding clouds iseliciting an optokinetic tracking reflex.) In an aircraftduring flight at night or in weather an oculogyral illusiongenerally confirms a somatogyral illusion: the pilot whofalsely perceives that he is turning in a particular directionalso observes his instrument panel to be moving in the samedirection.

The vestibular Coriolis effect, Coriolis cross-couplingeffect, vestibular cross-coupling effect, or simply the Coriolisillusion, is another false percept that can result from unusualstimulation of the semicircular duct system. To illustrate thisphenomenon, let us consider a subject who has been rotating inthe plane of his horizontal semicircular ducts (roughly the yaw

r plane) long enough for the endolymph in those ducts to attainthe sane angular velocity as his head: the cupulac in theampullae of his horizontal ducts have returned to their restingpositions, and the sensation of rotation has ceased (Fig. 29a).If the subject then nods his head forward iu the pitch plane,let's say a full 90 degrees for the sake of simplicity, he is

completely removing his horizontal semicircular ducts from theplane of rotation and inserting his two sets of vertical semi-circular ducts into the plane of rotation (Fig. 29b) . While theangular momentum of the subject's rotating head is forciblytransferred at once out of the old plane of rotation relative tohis head, the angular momentum of the endolymph in the horizontalduct is dissipated more gradually. The torqu.e resulting from thecontinuing rotation of the endolymph causes the cupulae in thehorizontal ducts to be deviated, and a sensation of angularmotion occurs in the new plane of the horizontal ducts--now theroll plane relative to the subject's body. Simultaneously, the .oendolymph in the two sets of vertical semicircular ducts mustacquire angular momentum, as these ducts have been brought intothe plane of constant rotation. The torque required to impartthis change in momentum causes deflection of the cupulae in theampullae of these ducts, and a sensation of angular motion inthis plane--the yaw plane relative to the subject's body--is theresult. The combined effect of the cupular deflection in allthree sets of semicircular ducts is that of a suddenly imposedangular velocity in a plane in which no actual angular acceler- yation relative to the subject has occurred. In the examplegiven, if the original constant-velocity yaw is to the right andthe subject pitches his head forward, the resulting Coriolisillusion experienced is that he and his immediate surroundingsare suddenly rolling to the right and yawing to the right.

A particular perceptual pheromenon experienced occasionallyby pilots of relatively high-performance aircraft during instru-ment flight has been attributed to the Coriolis illusion becauseit occurs in conjunction with large movements of the head underconditions of prolonged consta,,L zngular velocity. It consistsof a very convincing sensation of rolling and/or pitching thatappears suddenly after the pilot diverts his attention from theinstruments in front of him and moves his head to view someswitches or displays elsewhere in the cockpit. This illusion isespecially deadly because it is most likely to occur during aninstrument approach, a phase of flight in which altitude isbeing lost rapidly and cockpit chores (e.g., radio frequencychanges) repeatedly require the pilot to break up his instrumentcross-check. Whether the sustained angular velocities associatedwith instrument flying are sufficient to create Coriolis illu-sions of any great magnitude is debatable, however (26); andanother mechanism (the G-excess effect) has been proposed toexplain the illusory rotations experienced with head movementsin flight. Even if not responsible for spatial disorientationin flight, the Coriolis illusion is useful as a tool to demon-strate the fallibility of our nonvisual orientation senses.Nearly every military pilot living today has experienced theCoriolis illusion in the Barany chair or some other rotatingdevice as part of his physiological training; and fo• most ofthem it was thein tnat they first realized their own orientationsenses really cannot be trusted--the most important lesson ofall for instrument flying.

69

S~a

Figure 29. Mechanism of the Coriolis illusion. Subject rotatingin yaw plane long enough for endolymph to stabilizein semicircular duct (a), pitches head forward (b).Angular momentum of endolymph deviates cupula,causing subject to perceive rotation in new plane ofsemicircular duct, even though no actual rotationoccurred in that plane.

The otolith organs are responsible for a set of illusionsknown as aic ilusion. The mechanism of illusions ofthis type involves the displacement of otolithic membranes ontheir maculae by inertial forces in such a way as to signal afalse orienta ion when the resultant gravitoirertial force isperceived as gravitational (and therefore vertical). Theparagon of somatogravic illusions, the illusion of pitching upafter taking off into conditions of reduced visibility, isperhap.sý the best illustration of this mechanism. Consider thepilot of a high-performance aircraft holding his position at theend of the runway waiting to take off. Here the only forceacting on his otolithic membranes is the force of gravity, andthe positions of those membranes on their maculae signal accu-rately that down is toward the floor of the aircraft. Supposethe aircraft now accelerates down the runway, rotates, takesoff, cleans up gear and flaps, and maintains a forward accelera-tion of 1 g until reaching desired climb speed. The 1 G ofinertial force resulting from the forward acceleration displacesthe otolithic membranes toward the back of the pilot's head. In

70

Figure 30. A somatogravic illusion occurring on takeoff. Theinertial force resulting from the forward accelera-tion combines with the force of gravity to create aresultant gravitoinertial force directed down andaft. The pilot, perceiving down to be in thedirection of the resultant gravitoinertial force,feels he is in an excesnively nose-high attitude, andis tempted to push the stick forward to correct the 'i•.'Jillusory nose-high attitude.

fact, the new positions of the otolithic membranes are nearlythe same as they would be if the aircraft and pilot had pitched -up 45 degrees, since the new direction of the resultant gravito-inertial force vector (if we neglect the angle of attack andclimb angle) is 45 degrees aft relative to the gravitational Vvertical (Fig. 30). Naturally, the pilot's percept of pitchattitude based on the information from his otolith organs is oneof having pitched up 45 degrees; and the information from hisnonvestibular proprioceptive and cutaneous mechanoreceptivesenses supports this false percept, as the sense organs subser-ving those modalities also respond to the direction and intensityof the resultant gravitoinertial force. Giveii the vely strongsensation of a nose-high pitch attitude, one that is not chal-lenged effectively by the focal visual orientation cues provided

71 6

* - y ~ .~ s~v.- ~-. '...',,

by the attitude indicator, the pilot is tempted to push the noseof the aircraft down to cancel the unwanted sensation of flyingnose-high. Pilots succumbing to this temptation characteristi-cally crash in a nose-low attitude a few miles beyond the end ofthe runway. Sometimes, however, they are seen to descend out ofthe overcast nose-low and to try belatedly to pull up, as thoughthey suddenly regained the correct orientation upon seeing the .

ground again. Pilots of carrier-launched aircraft need to beespecially wary of the somatogravic illusion. These pilotsexperience pulse accelerations lasting 2 to 4 s and generatingpeak inertial forces of +3 to +5 Gx. Although the major accel-eration is over rather quickly, the resulting illusion ofnose-high pitch can persist for half a minute or more afterwards,resulting in a particularly hazardous situation for the pilotwho is unaware of this phenomenon (27).

Pilots of high-performance aircraft are not the only onesto suffer the somatogravic illusion of pitching up after takeoff.More than a dozen air transport aircraft are believed to havecrashed as a result of pilots' experiencing the somatogravicillusion on takeoff (28). A relatively slow aircraft, acceler-ating from 100 to 130 knots over a 10-s period just aftertakeoff, generates +0.16 Gx on the pilot. Although the resultantgravitoinertial force is only 1.01 G, barely perceptibly morethan the force of gravity, it is directed 9 degrees aft, signify-ing to the unwary pilot a 9-degree nose-up pitch attitude. Asmany slower aircraft climb out at 6 degrees or less, a 9-degreedownward pitch correction would put such an aircraft into adescent of 3 degrees or more--the same as a normal final-approachslope. In the absence of a distinct external visual horizon--even worse, in the presence of a false visual horizon (e.g., ashoreline) receding under the aircraft and reinforcing thevestibular illusion--the pilot's temptation to push the nosedown can be overwhelming. This type of mishap has happened atone particular civil airport so often that a notice has been .*placed on navigational charts cautioning pilots flying from this - lairport to be aware of the potential for loss of attitudereference.

Although the classic graveyard spiral was earlier indicatedto be a consequence of the pilot's suffering a somatogyralillusion, it can also be considered to result from a somatogravicillusion. A pilot who is flying "by the seat of his pants"applies the necessary control inputs to create a resultantG-force vector having the same magnitude and direction as thatwhich his desired flight path would create. Unfortunately, anyparticular G vector is not unique to one particular condition ofaircraft attitude and motion; and the likelihood that the Gvector created by a pilot flying without reference to instrumentsis that of the flight condition desired is remote indeed.Specifically, once an aircraft has departed a desired wings-levelattitude because of an unperceived roll, and the pilot does notcorrect the resulting bank, the only way he can create a Gvector which mutches that of the straight and level condition is

-72

WIN1PULWUV~iVJVIVUW~.~U~?VIJ~ UM W1WVVIX VUWLM VW rWV WWWWW1JWU*Vb.IV JU11-RIX~V h lV 1WJW LMM WFWIZUL WU VIW 1_WVWU~ Lr TýWW V -W IUrW LT7VW V W V

with a descending spiral. In this condition, as is always thecase in a coordinated turn, the centrifugal force resulting fromthe turn provides a Gy force which cancels the G component ofthe force of gravity that exists when the aircraft is banked.In addition, the tangential linear acceleration associated withthe increasing airspeed resulting from the dive provides a +Gxforce which cancels the -G component of the gravity vector thatexists when the nose o? the aircraft is pointed downward.Although the vector analysis of the forces involved ir, thegraveyard spiral is somewhat complicated, a pilot can easily andautomatically manipulate the stick and rudder pedals to cancelall vestibular and other nonvisual sensory indications that hisaircraft is turning and diving. In one mishap involving adark-night takeoff of a commercial airliner, the recorded flightdata indicated that the resultant G force which the pilotcreated by his control inputs allowed him to perceive hisdesired 10- to 12-degree climb angle, and a net G force between0.9 and 1.1 G, for virtually the whole flight--even though heactually levelled off and then descended in an acceleratingspiral until the aircraft crashed nearly inverted (Fig. 31).

The inversioillusi o is a type of somatogravic illusionin which the resultant gravitoinertial force vector rotates sofar backward as to be pointing away from, rather than toward,the earth's surface, thus giving the subject of this phenomenonthe false sensation that he is upside down. Figure 32 shows howthis can happen. Typically, a steeply climbing high-performanceaircraft levels off more or less abruptly at the desired alti-tude. This maneuver subjects the aircraft and pilot to a -Gzcentrifugal force resulting from the arc flown just prior tolevel-off. Simultaneously, as the aircraft changes to a morelevel attitude, airspeed picks up rapidly, adding a +Gx tangen-tial inertial force to the overall force environment. Addingthe -Gz centrifugal force and the +Gx tangential force to the1-G gravitational force results in a net gravitoinertial forcevector that rotates backward and upward relative to the pilot.This stimulates his otolith organs in a manner similar to theway a pitch upward into an inverted position would. Even thoughthe semicircular ducts should respond to the actual pitchdownward, for some reason this conflict is resolved in favor ofthe otolith-organ information--perhaps because the semicircularduct response is transient while the otolith-organ responsepersists, or perhaps because the information from the nonvestibu-lar proprioceptors and other mechanoreceptors reintorces thatfrom the otolith organs. The pilot who responds to the inversionillusion by pushing forward on the stick to counter the perceivedpitching up and over backward only prolongs the illusion bycreating more -Gz and +G forces, thus aggravating his situation.Turbulent weather usualy contributes to the development of theillusion; certainly, downdrafts are a source of -G forces thatcan add to the net gravitoinertial force producing the inversionillusion. Again, one does not have to be flying a jet fighterto experience this illusion. A number of reports of the inver-sion illusion describe the loss of control of air transport

20 I-

Y

iLLUSORY PITCH A"TITUOE

0 ACTUAL PITCH ATTiTUDE

-20 2

NET G FORCE I-30I

-40n_ _ _ L - ------.

TIME (GMT) ,-

Figure 31. Flight recorder data from a wide-body jetliner thatcrashed less than two minutes after taking off overthe ocean on a dark night. Although the net gravito-inertial force was essentially 1.0 G directed 10 to12 degrees aft of the aircraft vertical throughnearly the whole flight, the aircraft actuallyleveled off and eventually entered a spiral dive intothe water. The fact that the desired flight prof' .e(a straight climb) would have yielded the samegravitoinertial force environment as was actuallygenerated is strong ev idence that the pilot wasspatially disoriented.

aircraft that occurred when the pilot lowered the nose inappro-priatcly after experiencing the illusion. J upe is the termfor the sequence of events that includes instrument weather,turbulence, inability of the pilot to read his instruments, theinversion illusion, a pitch-down control input, and difficultyrecovering the aircraft because of resulting aerodynamic or --mechanical forces (30).

/4

- -;. -.. . *'.*o ' -

__. -,

Ficure 32. The I nversi on ilIlIu3i on. Cent rif ugalI a nd t a ngentialinertial forces d-&ring a level-off :oimbine witAh tlheforce of gravity to produce a resuit'.:nt gravitu-inertial force thtat rotates backward a.,d upward withrespect to the pilot, causing hi tou pu, ceive that hui s suddenly upsýide down. Tur bulent weather canproduce add-Itional inertial torce3- that contribute tothe illusaon. (Adapted froms Martin and Jones (29).1)

TheL-xc j.S on Ca ' al1 - be Con!--i d erIed z; ferU of0son-,atoyrav~ic ill usion, tecau-ý:.e it i rvolvvcs an a I riorr-al na:mi tude

udor direction of applied qrav,:to1ner tijl force that resul'LsrInf ,r t ion, anid th -l

can Le d mEarmineU' at I ~L alI iLatL i veyL a 6 ifqlj)u mTe chan i ca Ianalys~is Figj. 33) . 1,et L,: L,~r a ::uýIect I-, Si ttin) tuprig-htin a -tiG 1 C -n,: I 10ofl,:n t , nd 11 ( Lipo , 1 i h (a d f o a id' 310 d e rees.-

z5

P.s: a :esuilt of t' i chlan;1( in h'ead pos-itior Iil- o(oL-0 thic::er~ra~es i ie fowar tac¾ L[1 lte nour uo a 30-degrlee

til1t 1c2ia'tiVC to aetcl d (j ýLanc1(2: Of x 1111. NOW Suu'-Pose;

Upi~~~ --it 1113 -- U C ) rP

eiir t, andaigure t32. I he.nI5s iieaos 3i dcee- fonwjtd. TItif Iga and ItiS oethica .-Mnem L a a for; es I v. ard E.rn ta i: Q-f of e dwUu ,. C-d

1 o1r1e tof g av iy to ci. pirodu ; 1e sIt. C.n ra' - ,1L [

repe t........o,.a..g im to pu ¢e v that h..... . ..

°-P-

KN

1 ~ 1 9

2vGii

I G2 1 ... ' ,

• --2e G a , o

c 33. Mehais F the G-xce illusi. Subject in 2-G .m (lo

enionet (ye iyures,) expertiences, the res-Lult of - ulwe i

tilýit;hL hcad onily 30 dc'gL-ees. The illuisory

- %,

.;7

i01,

environ ie t (.IUI vU J ue ) ex erene the: rest1, a t i ute tof -a:t. 13 U I5- kpu]] on h I uti cu yt l t i hee t

blan :; hen he tlts his head 30 egres of te .•6m

the otolithic membranes now corresponds not to a 30-degreeforward tilt in the normal 1-G environment, but to a muchgreater one, theoretically as much as 90 degrees (2 sin 30 deg =sin 90 deg) . The subject had initiated only a 30-degree headtilt, however, and expects to perceive no more than that. The .unexpected additional perceived tilt is thus referred to theimmediate environment; i.e., he perceives his vehicle, if he isin one, to have tilted by the amount equal to the differencebetween his actual and expected percepts of tilt. In a high-performance aircraft the G-excess illusion can occur as a resultof the moderate amount of G force pulled in a turn--a penetrationturn or procedure turn, for example. If the pilot has to lookdown and to the side to select a new radio frequency or to pickup a dropped pencil while in a turn, he should experience anuncommanded tilt in both the pitch and roll planes due to theG-excess illusion. As noted previously, the G-excess illusionmay be responsible for the false sensation of pitch and/or rollgenerally attributed to the Coriolis illusion under such circum-stances (31) mb( .. '

Another illusion of otolith-organ origin, but not classifiedas a somatogravic illusion because it involves a visual percep-tual effect, is the oc.logajyls jjjgjo. The oculogravic illu-sion can be thought of as a visual correlate of the somatogravicillusion, and occurs under the same stimulus conditions (3ý . Apilot who is subjected to the deceleration resulting m omapplication of speed brakes, for example, experiences a nose-downpitch because of the somatogravic illusion. Simultaneously, heobserves the instrument panel in front of him to move downward,confirming his sensation of tilting forward. The oculogravicillusion is thus the visually apparent movement of an objectwhich is actually in a fixed position relative to the subjectduring changing magnitude and/or direction of the net gravito-inertial force. Like the oculogyral illusion, the oculogravicillusion probably results from the attempt to maintain vis;ualfixation during a vestibulo-ocular reflex, elicited in this caseby the change in magnitude or direction of the applied G vectorrather than by angular acceleration.

The •_JvaJo1 i is a special ".o.'c of oculogravicillusion that results from an increase or decrease in themagnitude of the +G fo acting on a subject. When one isaccelerated upward, as in .• elevator, the increase in +G fo;reel icits a ve_ .ibulo-ocular reflex of otolith-organ origin (theelevator reflex) that drives the eyes downward. Attempting tostabilize visually the objects in a fixed position relative tothe observer causes those objects to appear to stift upward whenthe G force is increased. The opposite effect occurs when oneis accelerated downward: the reduction in the magnitude of thenet gravitoiiiertial force to less than +1 G causes a reflexupward shift of the direction of gaze, and tle irhmeuiate s;ul-roundings appear to shift downward. (The latter effect ha,, alsoneen called the Q J7 tjýCeause of its occurr4.2ncedur]r transient weightjessness...) The iJmtuotLance of th(e elcvatou

0- .

illusion in aviation is not well documented. In one tragicmishap, however, it was probably experienced by the pilot of amilitary transport aircraft who became disoriented shortly afterleveling off abruptly from a prolonged steady descent on a darknight over desert terrain. The transient increase in +Gz forcethat occurred as the pilot leveled off at the landing patternaltitude most likely provoked the elevator illusion, and seeinghis instrument panel rise, he compensated by pitching downwardduring the subsequent fatal turn to final approach. We can alsoassume that updrafts and downdrafts produce elevator illusionsin pilots penetrating turbulent weather (Fig. 34).

.'..

.-p.

.4

Figure 34. Elevat~or illusion resulting f rom an updraft. Inthis type of oculogravic illusion, the increase in+G., Lorca elicits a vest ibulo-ocular reflex ofot6lith-organ origin which, when visual fixation is;'.ttempted, resulLL; in a falsely perceived upwardmotion of the object fixated--the instrument panel inthe example shown.

0°*

By far the most common vestibular illusion in flight is thel.eAnls. Virtually every instrument-rated pilot has had it, or I.Wwill get it, in one form or another at some time in his flyingcareer. It consists of a false percept of angular displacementabout the roll axis, i.e., is an illusion of bank, and isfrequently associated with an attempt by the pilot to compensatefor the illusion by leaning in the direction of the falselyperceived vertical (Fig. 35). The usual explanations of the •.y.leans invoke the known deficiencies of both otolith-organ andsemicircular-duct sensory mechanisms. As indicated previously,the otolith organs are not reliable sources of information aboutthe direction of the true vertical because they respond to theresultant gravitoinertial force, not to gravity alone. Further-more, other sensory inputs can sometimes override otolith-organcues and result in false perception of the vertical, even whenthe gravitoinertial force experienced is truly vertical. Thesemicircular ducts can provide such false inputs in flight byresponding accurately to some roll stimuli but rot responding toothers because they are below threshold. If, for example, apilot is subjected to an angular acceleration in roll so thatthe product of the acceleration and its time of application doesnot reach some threshold value--say, 3 degrees/s--then he doesnot perceive the roll. Let us suppose this pilot, who is tryingto fly straight and level, is subjected to an unrecognized anduncorrected 2-degree/s roll for 10 s: a 20-degree bank results.If the pilot suddenly notices the unwanted bank and corrects it .:-•

by rolling the aircraft back upright with a suprathreshold rollrate--say, 15 degrees/s--he experiences only half of the actualroll motion that took place, the half resulting from the cor-recting roll. As he started perceptually from a wings-levelposition, he is left with the illusion of having rolled into a20-degree bank in the directicin of the correcting roll, eventhough he is again wings-level. At this point he has the leans;and although he may be able to fly the aircraft properly by thevery deliberate and difficult process of forcing the attitude 6indicator to read correctly, his illusion can last for many -""minutes, seriously degrading his flying efficiency during thattime.

Interestingly, pilots frequently get the leans afterprolonged turning maneuvers, and not because of alternating 6subthreshold and suprathreshold angular motion stimuli. In aholding pattern, for example, the pilot rolls into a 3-degree/sstandard-rate turn, holds the turn for 1 min, rolls out andflies straight and level for 1 min, turns again for 1 min, andso ort until traffic conditions permit him to proceed toward hisdestination. During the turning segments the pilot initiallyfeels the roll into the turn and accurately perceives th'e bankedattitude. But as the turn continues, his percept of being in abanked turn dissipates and is replaced by a feeling of flyingstraight and level, both because the sensation of turning i3lost when thc endolymph comes up to speed in the semicir-ularducts (somatogyral illusion) and because the net G force beingdirected toward the floor of the aircraft urovides a false cue

7 ()

- .. -.. .*. . . . . . . . . . . .

r

Figure 35. The leans, the most common of all vestibular illusionsin flight. Falsely perceiving himself to be in a rightbank, but flying the aircraft straight and level bymeans of the flight instruments, this pilot is leaningto the left in an attempt to assume an upright posturecompatible with his illusion of bank.

of verticality (somatogravic illusion). When the pilot thenrolls out of the turn, he feels he has rolled into a banked turnin the opposite direction. With experience, a pilot learns tosuppress this false sensation quickly by paying strict atten-tion to the attitude indicator. Sometimes, however, the pilotfinds he cannot dispel the illusion of banking--usually when heis particularly busy, unfortunately. The leans can also becaused by misleading peripheral visual orientation cues, asmentioned in the discussion of the visual illusions. Rollangular vection is particularly effective in this regard, atleast in the laboratory. One thing about the leans is obvious:there is no single explanation for it. The deficiencies ofseveral orientation-sensing systems in some cases reinforce eachother to create an illusion; in other cases the inaccurateinformation from one sensory modality for some reason is selectedover the accurate information from others to create the illusion.Stories have surfaced of pilots suddenly experiencing the leansfor no apparent reason at all, or even of experiencing itvoluntarily by imagining the earth to be in a different directionfrom the aircraft. The point is that one must not think thatthe leans, or any other illusion for that matter, occurs as atotally predictable response to a physical stimulus: there ismuch more to perception tha. stimulation of the end-organs.

H/.H

Disorientation in Flight

Definitions. We have already defined an orientationalillusion to be a false percept of position, attitude, or motion,relative to the plane of the earth's surface. Spatial disorien-tation and the equivalent term, pilot vertigo, are usually takento mean the experiencing of an orientational illusion in flight.There is a major qualitative difference, however, between simplyexperiencing an orientational illusion and having to control an Vaircraft under conditions of misperceived or conflicting orien-tation cues. Furthermore, this difference becomes very importantin the analysis of mechanisms involved in aircraft mishaps due toorientational illusions, and in the development of training aidsfor educating pilots about the potential for lcss of aircraftcontrol while under the influence of orientational illusions.For those reasons, we find it necessary to restrict the use ofthe germ spatial disorientation to the condition wherein one rotonly has an orientational illusion but also needs to have correctperception of orientation for controlling his position, attitude,or motion. When one has an orientational illusion but has noneed for correct information about his orientation, we say he hasspatial unorientation. This distinction is exemplified by thecont :ast between the experience of a pilot, who must fly hisveh.*.2le on a desired path through space by responding to avail-able orientation cues (and to whom such cues are, therefore,highly relevant) , and the experience of an airborne communica-tions monitor, who can perform his duty without regard tL hisspatial orientation (and to whom orientation cues--whether trueor false--are essentially irrelevant). Obviously, it is spatialdisorientation, not spatial unorientation, that causes aircraftmishaps and warrants investigative and educational efforts tcprevent it.

Types. It is also useful to make the distinction betweenunrecognized (also called Ty I) and recognized (Tp II)spatial disorientation. As the term implies, unrecognizedspatial disorientation refers to the situation in which a pilot,oblivious to the fact that he is disoriented, controls hisvehicle completely in accord with and in response to his falseorientational percept. In recognized spatial disorientation thepilot realizes something is wrong with his ability to fly thevehicle, but he may or may not actually realize that the sourceof his problem is spatial disorientation. Even :urther out inthe spectrum of types of disorientation is that in which thepilot not only recognizes that he cannot control his vehicleeffectively because of spatial disorientation, but he also cannotc:btain correct orientation inforration because tne violence ofthe motion imposed is Dlurring his vision with counterproductivevest ibulo-ocular reflexes (ny.;tayniu:;). Fox want of aij adequat-eflydescriptive simple term We shall call s typ r vust ibulo--ocu Iardisorganization, or TIýe Tl_ spatial dis(.riceiLcttiofl.

Exa ples The la,;t of four "-15 Liagi c jlihtc- ailcralft"took off on a daytime j ;uo t i e ri wt itthe :di ng to follow thle

,

-----------------------

other three in a radar in-trail departure. Because of a naviga-tional error committed by the pilot shortly after takeoff, he wasunable to acquire the other aircraft on his radar. Frustrated,he elected to intercept the other aircraft where he knew theywould be in the arc of the standard instrument departure; so hemade a bee-line for that point, presumably scanning his radardiligently for the blips he knew should be appearing at any time.Meanwhile, after ascending to 1200 m (4000 ft) above groundlevel, he entered a descent of approximately 12 m/s (2400 ft/min)as a result of an unrecognized 3-degrees-nose-low attitude.After receiving requested position information from anothermember of the flight, the mishap pilot suddenly realized he wasin danger of colliding with the others, or he suddenly acquiredthem on radar: he then made a steeply banked turn, either toavoid a perceived threat of collision or to join up with therest of the flight. Unfortunately, he had by this time descendedfar below the other aircraft and was going too fast to avoid theground, which became visible under the overcast just before theaircraft crashed. This mishap resulted from an episode ofunrecognized, or Type I, disorientation. The specific illusionresponsible appears to have been the somatogravic illusion,created by the forward acceleration of this high-petformanceaircraft during takeoff and climb-out. The pilot's preoccupationwith the radar task compromised his instrument scan to the pointwhere the false vestibular cues were able to penetrate hisorientational information processing. Having unknowinglyaccepted an inaccurate orientational percept, he controlled theaircraft accordingly until it was too late to recover.

Examples of recognized, or Type II, spatial disorientationare easier to obtain than are examples of Type I, as mostexperienced pilots have anecdotes to tell about how they "gotvertigo" and fought it off. Some pilots were nct so fortunate,however. One F-15 Eagle pilot, after climbing his aircraft information with another F-15 at night, began to experiencedifficulty maintaining spatial orientation and aircraft controlupon leveling off in clouds at 8200 m (27,000 ft). "Talk aboutpractice bleeding," he commented to the lead pilot. Havingdecided to go to another area because of the weather, the twopilots began a descending right turn. At this point the piloton the wing told the lead pilot, "I'm flying upside down."Shortly afterward the wingman considered separating from theformation, saying, "I'm going lost wingman;" then, "No, I've gotyou;" and finally, "No, I'm going lost wingman." The mishapaircraft then descended in wide spiral, crashing into the desertless than a minute later, even though the lead pilot advised thewingman several times during the descent to level out. In thismishap the pilot probably suffered an inversion illusion uponleveling off in the weather, and entered a graveyard spiralafter leaving the formation. Although he kn2w he was disori-ented, or at least recognized the possibility, he still wasunable to control the aircraft effectively. That a pilot canrealize he is disoriented, see accurate orientation informationdisplayed on the attitude indicator, and still. fly into the

82

ground, always strains the credulity of nonaviators. Pilots whohave had spatial disorientation, who have experienced fightingoneself for control of an aircraft, are less skeptical.

The pilot of an F-15 Eagle, engaged in vigorous air combattactics training with two other F-15s on a clear day, initiateda hard left turn at 5200 m (17,000 ft) above ground level. Forreasons that have not been established with certainty, hisaircraft began to roll to the left at a rate estimated at 150 to180 degrees/s. He transmitted, "Out-of-control autoroll," as hedescended through 4600 m (15,000 ft). The pilot made at leastone successful attempt to stop the roll, as evidenced by themomentary cessation of the roll at 2400 m (8,000 ft); then theaircraft began to toll again to the left. Forty seconds elapsedbetween when the rolling began and when the pilot ejected--buttoo late. Regardless of whether the rollinc was caused by amechanical mnalfunction or was induced by the pilot himself, thecertain result of this extreme motion was vestibulo-oculardisorganization, which not only prevented the pilot from readinghis instruments but also kept him from orienting with thenatural horizon. Thus, Type III disorientation probably pre-vented him from taking appropriate corrective action to stop theroll and keep it stopped; if not that, it certainly compromisedhis ability to assess accurately the level to which his situationhad deteriorated.

t st. Despite continuing efforts to educate pilotsabout spatial disorientation and the real hazard it represents,the fraction of aircraft mishaps caused by or contributed to byspatial disorientation remained fairly constant over the threedecades between 1950 and 1980. A number of statistical studiesof spatial disorientation mishaps bear this out for the UnitedStates Air Force. In 1956 Nuttall and Sanford reported that, inone major air command during the period 1954-1956, spatialdisorientation was responsible for 4% of all major aircraftmishaps and 14% of all fatal. aircraft mishaps (33). Moser in1969 reported a study of aircraft mishaps in another major aircommand during the four-year period, 1964-1967: he found thatspatial disorientation was a significant factor in 9% of majormishaps and 26% of fatal mishaps (34). In 1971 Barnum andBonner reviewed the Air Force mishap data from 1958 through1968, and found that in 281 or 6% of the 4,679 major mishaps,spatial disorientation was a causative factor; fatalitiesoccurred in 211 of those 281, accounting for 15% of the 1,462fatal mishaps (35). A comment by Barnum and Bonner summarizessome interesting data about the "average pilot" involved in aspatial disorientation mishap: "He will be around 30 years ofage, have 10 years in the cockpit, and have 1,500 hours of firstpilot/instructor-pilot time. He will be a fighter pilot andwill have flown approximately 25 times in the three months priorto his accident." Barnum next analyzed the mishap data for thethree-year period, 1969-1971, and concluded that spatial disori-entation mishaps again accounted for 6% of major mishaps, butonly for 10% of fatal mishaps during this period (36). In a

8 30

1973 study, Kellogg found the relative incidence of Air Forcespatial disorientation mishaps in the years 1968 thruugh 1972 torange from 4.8 to 6.2%, and confirmed the high proportion offatalities in mishaps resulting from spatial disorientation (37).In 1980 Gillingham and Page (unpublished data) reviewed theAir Force aircraft mishaps of 1979 and determined that at least 9of the 94 major mishaps (9.6%) and 9 of the 49 fatal mishaps(18.4%) occurring that year would not have occurred had thepilots not been spatially disoriented at some time during themishap sequence. The cost of the Air Force aircraft destroyedeach year in disorientation mishaps has been until recently onthe order of $20 million per year. In 1979 it was $40 million;and the figure continues to rise, mainly as a result of therapidly rising cost of new military aircraft. Statistics on theincidence of disorientation-related aircraft mishaps in theUnited States Army and Navy (7.11% and 6.75% of total mishaps,respectively) (38,39) are remarkably similar to those of the AirForce, even though the flying missions of the several militaryservices are somewhat different.

Although statistics indicating the relative frequency ofspatial disorientation mishaps in air-carrier operations are notreadily available, it would be a serious mistake to concludethat there have been no air-carrier mishaps caused by spatialdisorientation. Fourteen such mishaps occurring between 1950and 1969 were reportedly due to somatogravic and visual illusionsresulting in the so-called "dark-night-takeoff accident" (28).In addition, 26 commercial airliners were involved in jet-upsetincidents or accidents during the same period (30). Spatialdisorientation is also a problem in general (non-military,non-air-carrier) aviation. Kirkham et al. reported in 1978that, although spatial disorientation is a cause or factor inonly 2.5% of all general aviation aircraft accidents in theUnited States, it is the third most common cause of fatalgeneral aviation accidents: 627 of the 4,012 fatal mishaps(15.6%) occurring in the years 1970 through 1975 involvedspatial disorientation as a cause or factor (40). Furthermore,the contribution of spatial disorientation to the second mostcommon cause of fatal general aviation accidents--continued VFRflight into adverse weather--is undoubtedly highly significant.Notably, 90% of general aviation mishaps in which disorientationis a cause or factor are fatal.

Dynamics of Spatial Orientation and DiQorintatio. It isnaive to assume that a certain pattern of physical stimulialways elicits a particular veridical or illusory perceptualresponse. Certainly, when a pilot has a wide, clear view of thehorizon, ambient vision supplies virtually all of his orientationinformation; anH potentially misleading linear or angularacceleratory motion cues do not result in spatial disorientation(unless, of course, they are so violent as to cause vestibulo-ocular disorganization). When a pilot's vision is compromised F

by weather, the same acceleratory motion cues can cause him todevelop spatial disorientation, but he usually avoids it by

0.2."

t.] •

NI-

referring to his aircraft instruments for orientation infor-mation. If the pilot is unskilled at interpreting the instru-ments, or if the instruments fail, those misleading motion cuesinevitably cause disorientation. Such is the character ofvisual din, the phenomenon wherein one incorporates visualorientation information into his percept of spatial orientation,to the exclusion of vestibular and nonvestibular proprioceptive,tactile, and other sensory cues. Visual dominance falls intotwo categories: the congenital type, in which ambient visionprovides dominant orientaLiun cues through natural neuralconnections and functions; and the acquired type, in whichorientation cue- are gleaned through focal vision and areintegrated as a result of training and experience into anorientational percept. The functioning of the proficient instru-ment pilot illustrates acquired visual dominance: he haslearned to decode with foveal vision the information on theattitude indicator and other flight instruments and to recon-struct that information into a concept of where he is, what heis doing, and where he is going; and he refers to that conceptwhen controlling his aircraft. This complex skill must bedeveloped through training and maintained through practice; andit is the fragility of this acquired visual dominance that makesspatial disorientation such a hazard.

The term v sup•ression is often used to denote theactive process of visually overriding undesirable vestibularsensations or vestibulo-ocular reflexes. An example of thisaspect of visual dominance is seen in well-trained figureskaters who, with much practice, learn to abolish the post-rotatory dizziness and nystagmus that normally result from thevery high angular decelerations associated with suddenly stoppingrapid spins on the ice (41) . But even these individuals, whendeprived of vision by eye closure or darkness, have the verydizziness and nystagmus we would expect to result from theacceleratory stimuli generated (42). In flight, the ability to"suppress unwanted vestibular sensations and reflexes is developedwith repeated exposure to the linear and angular accelerationsof flight. As is the case with the figure skaters, however, thepilot's ability to prevent vestibular sensations and vestibulo-ocular reflexes is compromised when he is deprived of visualorientation cues--when he must look away from his attitudeindicator to manipulate a radio frequency-selector knob, forinstance.

At this point, we introduce the ccncept of e_._opportunis. By this is meant the propensity of the vestibular

system to fill an orientation-information void swiftly andsurely with vestibular information. When a pilot flying ininstrument weather looks away from his artificial horizon for amere few seconds, this is usually long enough for erroneousvestibular information to break through the pilot's defenses andbecome incorporated into his orientational percept. In fact,conflicts between focal visual and vestibular sources of orien-tation information tend to resolve themselves very quickly in

favor of the vestibular information, without providing the pilot %an opportunity to evaluate the information. It would seem thatany orientation information reaching the vestibular nuclei--whether vestibular, other proprioceptive, or ambient visual--should have an advantage in competing with focal visual cues forexpression as the pilot's sole orientational percept, becausethe vestibular nuclei are primary terminals in the pathways forreflex orientational responses, and are the initial level of Aintegration for any eventual conscious concomitant of perception..f spatial orientation. In other words, although acquiredvisual dominance can be maintained by diligent attention to :

artificial orientation cues, the challenge to this dominancepresented by the processing of natural orientation cues throughprimitive neural channels is very potent and ever present.

The lack of adequate orientation cues, and conflictsbetween competing sensory modalities, are only a part of thewhole picture of a disorientation mishap. Why so many dis-oriented pilots, even those who know they are disoriented, areunable to recover their aircraft has mystified aircraft accidentinvestigators for decades. There are two possible explanationsfor this phenomenon. The first suggests that the psychologicstress of di.sorientation results in a disintegration of higher-order learned behavior, including flying skills. The seconddescribes a complex psychomotor effect of disorientation thatcauses the pilot to feel the aircraft itself is misbehaving.

The disintegration of flying skill perhaps begins with thepilot's realization that his spatial orientation and controlover the motion of his aircraft have been compromised. Undersuch circumstances, he pays more heed to whatever orientationinformation is naturally avaiiable, monitoring it more and morevigorously. Whether the brain stem reticular activating systemor the vestibular efferent system, or both, are responsible forthe resulting heightened arousal and enhanced vestibular infor-mation flow can only be surmised; but the net effect is thatmore erroneous vestibular information is processed and incor-porated into the pilot's orientational percept. This, ofcourse, only makes matters worse. A positive-feedback situationis thus encountered, and the vicious circle can now be brokenonly with a precisely directed, and very determined, effort bythe pilot. Unfortunately, complex cognitive and motor skillstend to be degraded under conditions of psychologic stress suchas occurs during Type II or Type III spatial disorientation.First, there is a coning of attention. Pilots who have survivedsevere disorientation have reported they were concentrating onone particular flight instrument instead of scanniny and inter- -.preting the whole group of them in the usual manner. Pilotshave also reported they were unaware of radio transmissions tothem while they were trying to recover from disorientation.Second, there is the tendency to revert to more primitivebehavior, even reflex action, under conditions of severe psycho-logic stress. The highly developed, relatively newly acquiredskill of instrument flying can give way to primal protective

86

, .w ,r, .•,r,, r.• r. r..Ir:. -- - - -,- ------ - - - ^. I,... %

responses during disorientation stress, making appropriaterecovery action unlikely. Third, it is often suggested thatdisoriented pilots become totally immobilized--frozen to theaircraft controls by fear or panic--as the disintegrationprocess reaches its final state.

The git han phenomenon, described by Malcolm andMoney (30), undoubtedly explains why many pilots have beenrendered hopelessly confused and ineffectual by spatial disorien-tation, even though they knew they were disoriented and shouldhave been able to avoid losing control of their aircraft. Thepilot suffering from this effect of disorientation 'erceivesfalsely that his aircraft is not responding proper ' to hiscontrol inputs, because every time he tries to bring the aircraftto the desired attitude, it seems actively to resist his effortand fly back to another, more stable, attitude. A pilot experi-encing disorientation about the roll axis (e.g., the leans orgraveyard spiral) may feel a force--like a giant hand--trying topush one wing down and hold it there, while the pilot withpitch-axis disorientation (e.g., the classic somatogravicillusion) may feel the airplane subjected to a similar forcetrying to hold the nose down. Pilots who ate unaware of theexistence of this phenomenon and experience it for the firsttime can be very surprised and confused by it, and may not be .able to discern the exact nature of their problem. A pilot'sradio transmission that the aircraft controls are malfuncti 'ningshould not, therefore, be taken as conclusive evidence tnat acontrol malfunction caused a mishap: spatial disorientationcould have been the real cause.

What mechanism could possibly explain the giant hand? Totry to understand this phenomenon, we must first recognize thatout- perception of orientation results not only in the consciousawareness of our position and motion but also in a preconsciouspercept needed for proper performance of voluntary motor activityand reflex actions. A conscious orientational percept can beconsidered rational, in that we can subject it to intellectualscrutiny, weigh the evidence for its veracity, conclude that itis inaccurate, and to some extent modify the percept to fitfacts obtained from other than the primary orientation senses.In contrast, a preconscious orientational percept must beconsidered irrational, in that it consists only of an integrationof data relayed to the bLain stem and cerebellum by the primaryorientation senses, and is not amenable to modification byreason. So what happens when a pilot knows he has becomedisoriented and tries to control his aircraft by reference to aconscious, rational percept of orientation which is at variancewith his preconscious, irrational one? Because only the datacomprising one's preconscious orientational percept are availablefor the performance of primitive orientational rr~flexes (e.g.,vest ibulo-ocular and postural reflexes), higher-order reflexes(e.g., aversive responses), and skilled voluntary motor activity(e.g., walking, zunning, bicycling, driving, flying), we shouldexpect the actual outcome of these t, pes of actions to deviate

87

from the rationally intended outcome whenever the orientationaldata upon which they depend are different from the rationallyperceived orientation. The disoriented pilot who consciouslycommands a roll to recover aircraft control, while the infor-mational substrate in reference to which his body functionsindicates that such a move is counterproductive or even danger-ous, may experience a great deal of difficulty in executing thecommand. Or he may discover that the roll, once accomplished,must be reaccomplished repeatedly, as his body responds auto-matically to the preconsciously perceived orientational threatresulting from his conscious efforts and actions to regaincontrol. Thus, the preconscious orientational percept influencesSherrington's "final common pathway" for both reflex and volun-tary motor activity; and the manifestation of this influence onthe act of flying during an episode of spatial disorientation isthe giant hand phenomenon. To prevail in this conflict betweenhis will and his skill, the pilot must decouple his voluntaryacts from his previously learned flying behavior, by accomplish-ing those motions which produce directly the desired readings ofthe flight instruments, rather than by flying the airplane to anattitude corresponding to the desired readings of the flightinstruments.

The salient features of the dynamics of spatial orientationand disorientation are diagrammed in Figure 36; the concepts ofvisual dominance, vestibular suppression, vestibular opportunism,disintegration of flying skill, conscious and preconsciousorientational percepts, and the giant hand phenomenon arepresented therein as they relate to the cverall scheme oforientation-information processing.

Coditions Conducive to Disorientation. From a knowledgeof the physical bases of the various illusions of flight one canreadily infer many of the specific environmental factors con-ducive to spatial disorientation. Certain visual phenomenaproduce characteristic visual illusions, such as false horizons,linear and angular vection, and autokinesis. Prolonged turningat a constant rate, as in a holding pattern or procedure turn,can precipitate somatogyral illusions or the leans; and Coriolisillusions can conceivably occur with head movements under theseconditions. Relatively sustained linear accelerations, such asthose that occur on takeoff, can produce somatogravic illusions;and head movements during G-pulling turns can elicit G-excessillusions.

What are the regimes of flight and activities of the pilotthat seem most likely to allow these potential illusions tomanifest themselves? Certainly, instrument weather and nightflying are primary factors. But especially likely to producedisorientation is the practice of switching back and forthbetween the instrument flying mode and the visual or coritac.flying mode; a pilot, is far less likely to become disoriented ifhe gets on the instruments as soon as out-of-cockpit vision is 0compromised and stays on the instruments until continuous

L'.

contact flying is again assured. In fact, any event or practicerequiring the pilot to break his instrumenc cross-check isconducive to disorientation. In this regard, aviouics controlswitches and displays in some aircraft are located where thepilot must interrupt his instrument cross-check for more than afew seconds to interact with them, and are thus known (not soaffectionately) as "vertigo traps." Some of these vertigo trapsrequire substantial movements of the pilot's head during thetime his cross-check is interrupted, thereby providing both areason and an opportunity for spatial disorientation to strike.

Formation flying in weather is probably the most likely ofall situations to produce disorientation; indeed, some experi-enced pilots get disoriented every time they fly wing or trailin weather. The fact that a pilot has little if any opportunityto scan his flight instruments while fly;.ng formation on hislead aircraft in weather means he is essentially isolated fromany source of accurate orientation information, and misleadingvestibular and ambient visual cues arrive unchallenged into hissensorium.

Of utmost importance to a pilot in preventing spatialdisorientation is competency and currency in instrument flying.A non-instrument-rated pilot who penetrates instrument weather isvirtually assured of developing spatial disorientation within amatter of seconds, just as the most competent instrument pilotwould develop it it he found himself flying in weather without-functioning flight instruments. Regarding instrument flyingskill, one must "use it or lose it," as they say. For thatreason pilots whose primary flying activity involves missions or-environments in which instrument wearher is rarely encountered(e.g., air combat training in the United States Southwest) •.ustaggressively seek opportunities to practice instzu:ent flying soas to maintain their proficiency at it. Otherwise, they coulddiscover that their instrument flying skill has deteriorate6 toa dangerously low level during a rare occasion when that skillis really needed.

Finally, conditions affecting the pilot's physical ormental health must be considered capable of rendering the pilotmore susceptible to spatial disorientation. The unhealthyeffect of alcohol ingestion on neural information processing isone obvious example; but the less well-known ability of alcoholto produce vestibular nystagmus (positional alcohol nystagmus---PAN) for many hours after its more overt effects have disappearedis probably of equal significance. Other drugs, such as barbitu-rates, amphetamines, and even the quinine in tonic water, aresuspected of possibly having contributed to aircraft mishapsresulting from spatial disorientation. Likewise, physical andmental fatigue, as well as acute or chronic emotional stress,can rob the pilot of his ability to concentrate on his instrureiitcross-chuck, and can therefore have deleterious effects on hisresistance to spatial disorientation.j 0*C"

CiI 1Et7S

C2-

Ila

=i 4i' j

Ii orw

. ,-

iC ' i ° '"

~'. '--'ND'

S... . .. .

I " " "

c• , • - ( ''a.......

wI

f ý3 1"I . ... . I II ,-

I --

_ -- o II ,-.

.1.

II.1

a>%

I-~I i9

B.K

Figure 36. Flow of orientation information in flight. Theprimary information-flow loop involves: stimulationof the visual, vestibular, and other orientation

senses by visual scenes and linear and angularaccelerations; processing of this primary orienta-tion information by brain stem, cerebellum, andlower cerebral centers; incorporating the solutioninto a data base for reflexive and skilled voluntarymotor activity (preconscious orientational percept);and effecting control inputs, which produce aircraftmotions that result in orientational stimuli. Asecondary path of information flow involves theprocessing of largely numerical data from flightinstruments into derived orientation information byninhpr cerebral centers. SublooD a Drovides forfeedback betwv-en various components of the nervoussystem, and includes efferent system influenies onthe sensory end-organs themselves. The phenomena ofvisual dominance, vestibular suppression, andvestibular opportunism occur in conjunction with thefunctioning of this loop. Subloop b generatesconscious perception of orientation, both from thebody's naturally obtained solution of the orien-tation problem and from orientation informationderived from flight instrument data. Voluntarycontrol commands arise in response to consciousorientational percepts; and the psychic stiessresulting from conflicting orientation informationor from apparently aberrantly responding effectorscan influence the manner in which orientationinformation is processed, leading ultimately todisintegration of flying skill. Subloop c incor-porates feedback from muscles, tendons, and jointsinvolved in making control inputs, and provides abasis for the giant hand phenomenon.

Prevention of Disorientation Mishaps

Spatial disorientation can be attacked in several ways.Theoretically, each link in the physiologic chain of eventsleading to a disorientation mishap can be broken by a specificcountermeasure (Fig. 37). Spatial disorientation can many timesbe prevented by modifying flying procedures so as to avoid thosevisual or vestibular stimuli that tend to creat-e illusions inflight. By improving the capacity of flight instruments tocranslate aircraft position and motion information into readilyassimilable orientation cues, we can help the pilot avoid

9 1

,,it

4LCUPITE SPATIX,ORIENTATIOI U_. NORRNTAION

POS ICNS T Z)SENSE 'EACEPUOL SP ==* RC06h::E0 RECOGINI2,O supwvEoW..• ORGAN PRDe' -Ssi• ILLUSI.IN OISOR*lNlTIO PRO ,L M SPATIAL SPATIA.

Y)0110k SfM LT SREITO O;SOAIFITN IA1 lOI .

UN ,NSe RVNEtI PROBLEM SPA'.AL SPATIAL

OISORIENT( h0N OtSORIETATNiO

PkOC[)U2P[S NVPR30 IJSTIkJMLmh AUTOPILO0 TLIgHT PHYSIOLOGICAL INSTRUCTIONCHANCES INSTIJMTNTG PROWIT WN' TRAIlING TRAINING EXPERIENCE

Figure 37. The chain of events leading to a spatial disorien-tation mishap, and where the chain can be attackedand broken. From left: Flight procedures can bealtered to generate less confusing sensory inputs.Improved instrument presentations can aid assimila-tion of orientation cues. Proficiency in instrument -.-

flying hpr-, smrt accurate orientational percepts.In the event the pilot suffers an orientationaliilusion, having the aircraft under autopilot controlavoids disorientation by substituting unorientation. ."Proper flight training allows the disoriented pilotto recognize he is having a problem controlling hiscraft. Once he knows he is having a problem, hisphysiological training helps him realize his problemis spatial disorientation. With appropriate instruc-tion and/or firsthand experience, the pilot withrecognized spatial disorientation can apply thecorrect control forces to recover the aircraft andsurvive the disorientation incident.

disorientation. Through repeated exposure to the environment ofinstrument flight, the pilot becomes proficient in instrumentflying; this involves developing perceptual processes thatresult in accurate orientational percepts rather than orienta-tional illusions. If a pilot who is experiencing an illusioncan relinquish control of his aircraft to an autopilot, he canconvert his situation from one of hazardous spatial disorienta-tion to one of irrelevant spatial unorientation, and reclaimcontrol once the orientational illusion has subsided. Use ofthe autopilot, not only to help the pilot recover from disorien-tation, out also to help prevent it in the first place, is atechnique that has considerable potential for saving lives,particularly in general aviation. Given that a pilot hasdeveloped spatial disorientation, if he can be made to recognizethat he is disoriented, he is halfway along the road to recoveLy.To recognize disorientation is not necessarily easy, however.First, the pilot must recognize that he is having a problemholding his altitude or heading; this he cannot do if he isconcentrating on something other than the flight instruments--onthe radar scope, for instance. Only through proper flight

92.pS~.(I"

W 4- 4 U'JW.T WV WV W•j W'J• W.' '. W• V W' % V. M. U u~''•' '••-•'' 1 '4.F• " V. F.• U~ V. V• ,C V J V. . , -- *•X

%• "

training can the discipline of continuously performing theinstrument cross-check be instilled. Second, the pilot mustrecognize that his difficulty in controlling the aircraft is aresult of spatial disorientation. This ability is promotedthrough physiological training. We said that the pilot whosuspects he is disoriented is halfway down the road to recovery:why not most of the way? Because a pilot's ability to cope withthe effects of disorientation on his control inputs to theaircraft comes through effective flight instruction, proper V.,physiological training, and experience in controlling hisvehicle in an environment of conflicting orientation cues--hissimply being aware that he is disoriented by no means ensureshis survival.

EducatiQn and TraininQ. Physiological training is the mainweapon against spatial disorientation at the disposal of theflight surgeon and aerospace physiologist. This trainingideally should consist of both didactic material and demon-strations. There is no paucity of didactic material on thesubject of disorientation: at least eight films, five video-cassette tapes, tnree slide sets, two handbooks, dnu, numerouschapters in books and manuals have been prepared for the purposeof informing the pilot about the mechanisms and hazards ofspatial disorientation. Although the efforts to generateinformation on spatial disorientation are commendable, there has . -

been a tendency for the didactic material thus far produced todwell too much on mechanisms and effects of disorientationwithout giving much practical advice on how to deal with it. -

Money arid Malcolm (43) noted that none of the available films onspatial disorientation gives sufficient emphasis to what the -.pilot should do when he suspects disorientation. While sev-eral of the films recommend that the pilot believe his instru-,ments, this message is too subdued and, by itself, is inadequate.Money and Malcolm argue that under some circumstances (e.g.,panic) a pilot may in fact believe the instruments but continu.to fly the aircraft according to his false orientational per-cept. If a pilot is told in addition, "Make the instrumentsread right, regardless of your sensations," he has simple,definite instructions on how to bring the aircraft under con-trol when disorientation strikes. We strongly advise, therefore,that every presentation to pilots on the subject of spatialdisorientation emphasize the need to make the instrurrents readright, as well as to believe them, when responding to disoriern-tation stress.

The traditional demonstration accompanying lectures topilots on spatial disorientation is a ridc on a Barany chairor other smoothly rotating device. The subject, sitting inthe device with his eyes closed, is accelerated to a con-stant angular velocity and asked to signal with his thumbs hisperceived direction of turning. After a number of seconds(usually from 10 to 20, at constant angular velocity, the subjectloses the sensation of rotation and signals this fact to theobuservers. Then the instructor suddenly stops the rotation,

(~?) 0]

whereupon the subject immediately indicates that he feels he isturning in the direction opposite his original direction ofrotation. The subject is usually asked to open his eyes duringthis part of the demonstration, and is amazed to see that he isactually not turning, despite the strong vestibular sensation ofrotation. After the described demonstration of somatogyralillusions, the subject is again rotated at a constant velocitywith his eyes closed, this time with head down (facing thefloor) . When the subject indicates his sensation of turninc, hasceased, he is asked to raise his head abruptly so as to face thewall. The Coriolis illusion resulting from this maneuver is oneof a very definite roll to one side: the startled subject mayexhibit a protective postural reflex, and may open his eyes tohelp him visually orient during his falsely perceived upset. Themessage delivered with these demonstrations is not that suchillusions will be experienced in flight in the same manner, butthat the vestibular sen.e can be fooled--i.e., is unreliable--andthat only the flight instruments provide accurate orientationinformation.

Over the years at least a dozen different devices have beendeveloped to augment or supplant the Barany chair for demon-strating various vestibular and visual illusions and the effectsof disorientation in flight. These devices, collectively knownas antivertigo ta_ e, fall into two basic categories:orientational illsio demonstrators and & disorientationdemonstrators. The great majority are illusion demonstrators,in which the subject rides passively and experiences one or moreof the following: somatogyral, oculogyral, somatogravic,oculogravic, Coriolis, G-excess, vection, and autokineticillusions. In an illusion demonstrator, the subject typicallyis asked to record or remember the magnitude and direction ofthe orientationa. illusion, and then is told or otherwiseallowed to experience his true orientation. A few antivertigotrainers are actually spatial disorientation demonstrators,which allow the subject to experience the difficulty in con-trolling the attitude and motion of the trainer while beingsubjected to somatogravic, somatogyral, and/or Coriolis illu-sions. Figure 38 shows two antivertigo trainers presently inuse in the United States Air Force. One must be aware that thename given to any particular antivertigo trainer does notnecessarily describe its function: The USAFSAM Spatial Disorien-tation Demonstrator, for example, was actually an orientationalillusion demonstrator.

Although the maximum use of antivertigo trainers in physi-ological training of pilots is to be encouraged, it is importantto recognize the great potential for misuse of such devices bypersonnel not thoroughly trained in their theory and function.Several antivertigo trainers have aircraft-instrument trackingtasks for the subject to perform while he is experiencingorientational illusions but is not actually controlling themotion of the trainer. The temptation is very strong forunsophisticated operating personnel to tell the subject he is

94

"fighting disorientation" if he performs well on the trackingtask whi' e subjected to the illusion-generating motions. UBecause the subject's real orientation is irrelevant to thetracking task, any orientational illusion is also irrelevant,and he experiences no conflict between visual and vestibularinformaticn in acquiring cues upon which to base his controlresponses. This situation, of course, does not capture theessence of disorientation in flight; and the trainee who is ledto believe he is fighting disorientation in such a ground-baseddemonstration ray develop a false sense of security about hisability to combat disorientation in flight. The increasiag useof spatial disorientation demonstrators, in which the subject -must control the actual motion of the trainer Dy referring totrue-reading instruments while under the irfluence of orieiitational illusions, will most likely reduce the potential formisuse and will improve the effectiveness of presentations to -

pilots on the subject of spatial disorientation.

•-U.-)

a b

Figure 38. Two types of antivertigo trainer currently in use:a. The Vertigon®, an orientational illusion demon-strator, allows the subject to experience Coriolisand somatogyral illusions while performing trackingand dial-setting tasks. b. The Vertifuget, a spatialdisorientation demonstrator, subjects the trainee tosomatogravic, Coriolis, and somatogyral illusionsthat generate orientational conflicts as he tries tocontrol the attitude and motion of the device byreferring to a true-reading attitude indicator.

95

Flight training provides a good opportunity to instructpilots about the hazards of spatial disorientation. Inflight S

demonstrations of vestibular illusions are included in mostformalized pilot training curricula, although the efficacy ofsuch demonstrations is highly dependent on the motivation andskill of the individual flight instructor. Somatogyral andsomatogravic illusions and illusions of roll attitude canusually be induced in a student pilot by a flight instructor whoeither understands how the vestibular system works or knows fromexperience which maneuvers consistently produce illusions. Thevestibular-illusion demonstrations should not be confused withthe unusual-attitude-recovery demonstrations in the typicalpilot training syllabus: the objective of the former is for thestudent to experience orientational illusions and recognize themas such; that of the latter is for the student to learn toregain control of an aircraft in a safe and expeditious manner.In both types of demonstration, however, control of the aircraftshould be handed over to the student pilot with the instruction,"Make the instruments read right."

Part of flight training is continuing practice to maintainflying proficiency, and the importance of such practice inreducing the likelihood of having a disorientation mishap cannotbe overemphasized. Whether flying on instruments, in formation,or engaged in aerobatic maneuvering, familiarity with theenvironment--based on recent exposure to it--and proficiency atthe flying task--based on recent practice at it--result not onlyin a greater ability to avoid or dispel orientational illusions,but also in a greater ability to cope with disorientation whenit does occur.

Inflight Procedures. If a particular inflight procedurefrequently results in spatial disorientation, it stands toreason that modifying or eliminating that procedure should helpreduce aircraft mishaps due to disorientation. Night formationtakeoffs and rejoins are examples of inflight procedures thatvery frequently are associated with spatial disorientation; andthe United States Air Force has, wisely, officially discouragedthese practices in most of its major commands.

Another area of concern is the "lost wingman' procedure,used when a uilot has lost sight of the aircraft on which he hasbeen flying wing. Usually the loss of visual contact is due topoor visibility, and occurs after a period of vacillationbetween formation flying and instrument flying. Such conditions,of course, invite disorientation. The lost wingman proceduremust, therefore, be made as uncomplicated as possible whilestill allowing safe separation from the other elements of theflight. Maintaining a specified altitude and heading away fromthe flight until further notice is an ideal lost wingman proce-dure, in that it avoids frequent or prolonged disorientation-inducing turns and minimizes cognitive workload. Often a pilotflying wing in bad weather does not lose sight of the leadaircraft, but suffers so much disorientation stress as to make

96

1h V

the option of going lost wingman seem safer than that of continu-ing in the formation. A corwmon practice in this situation is forthe wingman to take the lead position in the formation, at least Vuntil the disorientation disappears. This avoids the necessityof having the disoriented pilot make a turn away from the flight •to go lost wingman, which could be especially difficult and Vdangerous because of his disorientation. Onp should question thewisdom of having a disoriented pilot leading a flight, however;and some experts in the field of spatial disorientation areadamantly opposed to this practice, with good reason.

Verbal communication between pilots can help preventdisorientation. In formation flight in weather, for example, itis good practice for the flight leader to inform his wingmanperiodically of the attitude, altitude, airspeed and heading ofthe flight, as the task of formation flying makes it difficultfor the wingman to obtain accurate orientation information bymonitoring flight instruments. In some cases a copilot or othercrew member is available to monitor aircraft attitude, motion,and position during times when the pilot is fully occupied withothe=± "vdLuc,,Iing tasks (e.g., weapons delivery at night). The Nother crew member's verbal orientational status reports orwarnings of hazardous orientations can serve the pilot well undersuch circumstances.

The manner in which others communicate with the pilot whois disoriented can mean the difference between lite and deathfor that pilot and his passengers. Unfortunately, no clear-cutprocedure exists for ensuring appropriate communications to adisoriented pilot. Should he be hounded mercilessly with verbalorders to get on the instruments, or should he be left relativelyundistracted to solve his orientation problems? The extremes ofharassment and neglect are definitely not appropriate; a fewforceful, specific, action-oriented commands probably representthe best approach. "Level the artificial horizonl" and "Rollright 90 degreesi" are examples of such commands. One mustremember that the pilot suffering from spatial disorientationmay be either so busy or so functionally compromised thatfriendly chit-chat or complex instructions may fall on deafears. Simple, emphatic directions may be the only means ofpenetrating the disoriented pilot's consciousness.

To illustrate how official recommendations regardinginflight procedures are disseminated to pilots in an effort toprevent spatial disorientation mishaps, a message from a majorUnited States Air Force command headquarters to field units isexcerpted here:

".. .Review SD procedures in [various Air Forcemanuals] ... Discuss the potential for SD during flightbriefings prior to flight involving night, weather, orconditions where visibility is significantly reduced...Recognize the [SD] problem early and initiate correc-tive actions before aircraft control is compromised.

9?

A. Single Ship:

(1) Keep the head in the cockpit.Concentrate on flying basic instrumentswith frequent reference to the attitudeindicator. Defer non-essential cockpitchores.

(2) If symptoms persist, bring aircraftto straight and level flight using theattitude indicator. Maintain straight andlevel flight until symptoms abate--usually 30 to 60 seconds. Use autopilotif necessary.

(3) If necessary, declare an emergencyand advise air traffic control. Note: itis possible for SD to proceed to thepoint where the pilot is unable to see,interpret, or process information fromthe flight instruments. Aircraft controlin such a situation is impossible. Apilot must recognize when physiological/psychological limits have been exceededand be prepared to abandon the aircraft.

B. Formation flights:

ýI) Separate aircraft from the formationunder controlled conditions if theweather encountered is either too denseor turbulent to insure safe flight.

(2) A flight lead with SD will advisehis wingmen that he has SD and he willcomply with procedures in Paragraph A. Ifpossible, wingmen should confirm straightand level attitude and provide verbalfeedback to lead. If symptoms do notabate in a reasonable time, terminate themission and recover the flight by thesimplest and safest means possible.

(3) Two-ship formation. Wingman willadvise lead when he experiences signifi-cant SD symptoms. A

(a) Lead will advise wingman ofaircraft attitude, altitude,heading, and airspeed.

,*.

98 .

(b) The wingman will adviselead if problems persist. If so,lead will establish straight andlevel flight for at least 30 to60 seconds.

(C) If the above procedures arenot effective, lead shouldtransfer the flight lead e %position to the wingman while, ins t r a i h a n l e v el f l ig h t. O n c eassuming lead, maintain straightand level flight for 60 seconds.If necessary, terminate the Imission and recov r by the

simplest and saf ieanspossible.

(4) More than two-ship tion. Leadshould separate the flighL ir,&o elementsto more effectively handle a wingman withpersistent SD symptoms. Establishstraight and level flight. The elementwith the SD pilot will remain straightand level while other element separatesfrom the flight."

Cockpit Layout and Flight Instruments. One of the mostnotorious vertigo traps is the communications-transceiverfrequency selector or transponder code selector located in anobscure part of the cockpit: to manipulate this selectorrequires the pilot not only to look away from his flight instru-ments, thus interrupting his instrument scan, but also to tilthis head to view the readout, thus potentially subjecting him toa Coriolis or G-excess illusion. Aircraft designers are nowaware that easy accessibility and viewing of such frequentlyused devices minimize the potential for spatial disorientation;accordingly, most modern aircraft have communications frequencyand transponder code selectors and readouts located in front ofthe pilot near the flight instruments.

The location of the flight instruments themselves is alsovery important: they should be clustered directly in front of -[.the pilot, and the attitude indicator--the primary provider oforientation cueing and the primary instrument by which theaircraft is controlled--should be in the center of the cluster(Fig. 39). When this principle is not respected, the potentialfor spatial disorientation is increased. A certain modernfighter aircraft, for example, was designed to have the pilotsittiig high in th. cockpit to enhance his field of view during .".air-to-air combat in conditions of good visibility. This designrelegated the attitude indicator to a position more or lessbetween the pilot's knees. As a result, at night and duringinstrument weather, the pilot is subjected to potentially

99 9

disorienting peripheral visual motion and position cueing byvirtue of his being surrounded by a vast expanse of canopy,while he tries to glean with central vision the correct orien-tation information from a relatively small, distant attitudeindicator. The net effect is an unusually difficult orientationproblem for the pilot, and a greater risk of developing spatial %

disorientation in this aircraft than in others with a larger andmore advantageously located attitude indicator.

L.

Figure 39. A well-designed instrument panel, with the attitudeindicator located directly in front of the pilot, andthe other flight instruments clustered around it.Radios and other equipment requiring frequentmanipulation and viewing are placed close to theflight instruments to minimize interruption of thepilot's instrument scan and to obviate his having tomake head movements that could precipitate spatialdisorientation. (Photo courtesy of Gen-Aero, Inc., -

San Antonio, TX.)

The verisimilitude of the flight instruments is also amajor facrc'r in their ability to convey readily assimilable"orientation information. The old "needle, ball, and airspeed"

100

indicators (a ieeuae pointer showing the direction and rate ofturn, a ball showing whether the turn is being properly coordi-nated with thý? rudders, and an airspeed indicator showingwhether the airp.1ane is climbing or diving) required a lot ofinterpretation for the pilot to perceive his spatial orientation "..through them; nevertheless, this combination sufficed for nearlya generation of pilots. When the attitude indicator (also knownas the gyro hoLizon, artificial horizon, or attitude gyro) wasintroduced, it greatly reduced the amount of work required tospatially orient during instrument flying because the pilotcould readily imagine the artificial horizon line to be the realhorizon. In addition to becoming more reliable and more versa-tile over the years, it became even easier to interpret: theface was divided into a gray or blue "sky" half and a black orbrown "ground" half, with some models even having lines ofperspective converging to a vanishing point in the lower half.Such a high degree of similarity to the real world has made theattitude indicator the mainstay of instrument flying today.

A relatively new concept in flight instrumentation, theheAd-uj. U ij play (i-iUD, )LVjeLb iiumeric and orher symDo0Licinformation to the pilot from a combining glass near the wind-screen, so he can be looking forward out of the cockpit andsimultaneously monitoring flight and weapons data. When thepilot stlects the appropriate display mode, pitch and rollattitude of the aircraft are observed on the "pitch ladder"(Pig. 40) and heading, altituK:-, airspeed, and other flight 1parameters are numerically displayed elsewhere on the HUD. Itsup-front location and its close-together arrangement of most ofthe required aircraft control and performance data make the HUDan attractive alternative to the conventional cluster of instru-ments, and some pilots use the HUD as the primary instrument forspatial orientation and aircraft control during instrumentflight. Pilots' acceptance and use of the HUD for flying ininstrument weather has not been universal, however: many preferto use the HUD under conditions of good outside visibility anduse the conventional instruments for flying at night and inweather. The reason fcr this preference may be that the horizonon the conventional instrument looks more like the naturalhorizon than does the zero-pitch indicator on the HUD pitchladder. Another reason may be that the HUD presents such anarrow view of the outside world--a 'vernier" view with highresolution--while the conventional attitude indicator aives anexpansive pict•.ial view oui the spatial environment. Further-more, the relative instability of the HUD pitch ladder and thetendency for the zero-pitch line to disappear from view make theHUD somewhat difficult to use during moderately active maneuver-ing, as would be necessary during an unusual-attitude recoveryattempt.

As good as they are, both the attitude indicator and theHUD leave much to be desired as flight instruments for assuringspatial orientation. Both suffer from the basic design defi-cie ' of presenting visual spatial orientation information to

.'. 4

I I . . .. . . . . . . . . . . . . . . . . . . . .. . .IF.. . II

%.

Figure 40. A typical head-up display (HUD) The pitch ladder inthe center of the. display pirovides pitch and rollattitude information.

>-p.

the wrong senisory system--the focal vi-suail syster.. Two untowardef fects result. First, the pij-ot'Ls focal vision not only n-,ustserve to d SC r41;4riniate numer-ical data fromt a number of in~stru-nents, but also .-must take on the task of spatially orienting thep i lot. The p--ilot- thus has; to emtploy h-is focal visual systen- ina soinewiiat inetfficioneL marinci duriima i nstrunertt 1icjht , withabout 70% oil his timr.e- !;pr-nt viewing the atti.,u.- ucic dicator,while his airdicrit v>-m fir.I., unutilized. s-econd, uthe fcactkthat I oc I. vision ii u- naLwq1lyciprpe)Cd to. ip.ov ide primaryspatial orientation cues-L caases pilots djiffdculty in interpreting 4

the artificial horizon J~rect.ly: th~eie is a ten-idency, especiallyamong novice pilots, to soL; aukwaiciL; the di:splayed deviationsin roll and pitch, anud to make mi n i al i o" I and,, 1iitch correcti onsi ri t he h. ~~ o Ug directL i on. eve a- 1 a piuuo IICiVC LG0 aknt

t ry to mip ov( he c f f i i jcy ofth 11juteL ' s Z.c (IiU t i oI o 1oi i jei nt atL( io inr )io iia t-i on if rem tie att- L týLud U id i ccvit (r a nu osso i;-a ted f 1 iglt inst.rumeitLS . One ha:;ei: t.0 H~akC LIOe larti I ,c alher izon. stat icoiy 1but . e rollI ond *jicý tL tIj( W~i!t bi aI c:za it uOhtie I1ntru::iCIt- to r*:diCd~t-e the:;.t o ut L14 It-l ciii"C eiLz C

I ~i 2

- 5,-- -

theceterofthediply p:oide ptc and rol.

so-called "outside-in" presentation, as opposed to the "inside-out" presentation of conventional attitude indicators). Theoret- k %.0,

ically, this configuration relieves the pilot of having tospatially orient himself before trying to fly the aircraft: thepilot merely flies the small aircraft on the attitude instrument -

and the real aircraft follows, so to speak. Another approachinvolves letting the artificial horizon provide pitch inforina-tion, but having the small aircraft on the attitude instrument '01%provide roll information, and collocating heading, verticalvelocity, airspeed, throttle setting, and navigation parameterswith the aircraft attitude information on a 7-cm instrument face.Such an instrument is the Crane Alweather Flitegage* (Fig. 41). -.

Neither of these approaches, however, frees foveal vision fromthe unnatural task of processing spatial orientation information.Another concept, the peripheral visual horizon display (PVHD), -

also known as the Malcolm horizon, attempts to give pitch androll cues to the pilot through his paracentral and peripheralvision, thus sparing foveal vision for tasks requiring a highdegree of visual discrimination (44,45). The several varietiesof PVHD that have been developed project across the instrumentpanel a long, thin line. of light representing the true horizon,which line of light moves directly :.n accordance with therelative movement of the true horizon (Fig. 42) . The potential

Figure 41. The Crane Alweather Fiitegageg. Thle artificial horizonprovides pitch infOrr~at~iOr in tlle usual mranner, butthe small airplane on the -Instrumnent rolls in accor-dance with the rollingy mutio,otu of the real aircraft.Other inipoitarit flight dat. are displayedi on theinstru~ment face, thusý nI'ixiin~iz~ixi the work involved inthe crosschieck,

A -

I (130

* • . ' ,

- .- -

Figure 42. The peripheral visual horizon display (PVHD)}, orMalcolm horizon. An artificial horizon projectedacross the instrument panel moves in accordance withthe real horizon, and the pilot observes the pro-jected horizon and its movement with his peripheralvision. Theoretically, this enables him to processspatial orientation information in the naturalfashion, and spares his foveal vision for tasksrequiring a high degree of visual discrimination.

for further development and eventual pilot acceptance of PVHD-type aircraft attitude displays appears good, as the PVHD isbased on the physiologically sound concept of providing primaryspatial orientation cueing through ambient vision--i.e., in thenatural fashion.

Other Sensory Phenomena

Flicker vertigo, fascination, and target hypnosis are tradi- .tionally described in conjunction with spatial disorientation--evern though, strictly speaking, these entities involve alter-ations of attention rather than aberrations of perception. Nor

]04

. W 1W;

is the breakoff phenomenon related directly to spatial disorien-tation, but the unusual sensory manifestations of this conditionmake our discussion of it here seem appropriate. .

Flicker Vertigo. As most people are aware from personalexperience, viewing a flickering light or scene can be distract-ing, annoying, or both. In aviation, flicker is sometimescreated by helicopter rotors or idling airplane propellersinterrupting direct sunlight, or less frequently by such thingsas several anticollision lights flashing in nonunison. Pilotsreport that such conditions are indeed a source of irritation --and distraction, but there is little evidence that flickerinduces either spatial disorientation or clinical vertigo innormal aircrew. In fact, one authority insists there is no suchthing as flicker vertigo and that the original reference to itwas merely speculation (46). Certainly, helicopter rotors orrotating beacons on aircraft can produce angular vection illu-sions because they create revolving shadows or revolving areasof illumination; but vection does not result from flicker.Symptoms of motion sickness also can conceivably result from thesensory conflict associated with angular vection; but again,these symptoms would be produced by revolving lights and shadows,not by flicker.

Nevertheless, one should be aware that photic stimuli atfrequencies in the 8- to 14-Hz range, that of the electro-encephalographic alpha rhythm, can produce seizures in thoserare individuals who are susceptible to flicker-induced epilepsy.Although the prevalence of this condition is very low (less than0.00005) , and the number of pilots affected are very few, somehelicopter crashes are thought to have been caused by pilot-,; N"%

sutfering from flicker-induced epilepsy.

Fascination. Coning of attention is something everyoiieexperiences every day, but it is especially likely to occurwhen we are stressed by the learning of new skills or by therelearning of old ones. Pilots are apt to concentrate on oneparticular aspect of the flying task, to the relative exclusionof others, when that aspect is novel or unusually demanding. Ifthis concentration is of such a degree as to cause the pilot todisregard important information to which he should respond, itis termed fascination. An extreme example of fascination isthat in which the pilot becomes so intent on delivering weaponsto the target that he ignores the obvious cues of ground prox-imity and flies into the ground. Mishaps of this sort are saidto result from target hypnosis; no actual hypnotic process issuspected or should be inferred, however. Other examples offascination in aviation are: the noniLoring of one flightinstrument rather than cross-checking during particularlystressful instrument flight; paying so much attention to flyingprecise formation that other duties are neglected; and theaviator's most ignominious act of negligence, landing an airplanewith the landing gear up, despite the clearly perceived warningfrom the gear-up warning horn. These examples help us appreciate

J]J5

the meaning of the original definition of fascination by Clarket al. (47): "a condition in which the pilot fails to respondadequately to a clearly defined stimulus situation in spite ofthe fact that all the necessary cues are present for a properresponse and the correct procedure is well known to him." From othe definition and the examples given, we can see that fasci- P.nation can involve either a sensory deficiency or an inability rto act, or perhaps both. We also know that fascination--atleast the type involving sensory deficiency--occurs not onlyunder conditions of relatively high workload, but also can occurwhen workload is greatly reduced and tedium prevails. Finally,the reader should understand that coning of attention, as occurswith fascination, is not the same thing as tunneling of vision,which occurs with G stress: even if all pertinent sensory cuescould be made accessible to foveal vision, the attentionallapses associated with fascination could still prevent thosecues front being perceived ei eliciting a response.

Breakoff. Clark and Graybiel (48) repoxtLed in 1957 acondition perhaps best described by the title of their paper:"The Breakoff Phenomenon--A Feeling of Separation from the EarthExperienced by Pilots at High Altitude." They found that 35% of137 United States Navy and Marine Corps jet pilots interviewedby them had had feelings of being detached, isolated, or phbs-ically separated from the earth when flying at high altitudes.The three conditions most frequently associated with the experi-ence were high altitude (5000 to 15,000 m, with a mediar. of10,000 m), being alone in the aircraft, and not being particu-larly busy with operating the aircraft. The majority of thepilots interviewed found the breakoff experience exhilarating,peaceful, or otherwise pleasant; over a third, however, feltanxious, lonely, or insecure. No operational importance couldbe ascribed to the breakoff phenomenon; specifically, it was notconsidered to have a significant effect on a pilot's ability tooperate his aircraft. The authors nevertheless suggested thatthe breakoff experience might precipitate significant effects on 1.,

a pilot's performance when coupled with preexisting anxiety orfear, and for that reason the phenomenon should be described topilots before they go alone to high altitudes for the firsttime. Breakoff may, on the other hand, have a profound, positiveeffect on motivation to fly. Who could deny the importance ofthis experience to John Gi)]espie Nagee, Jr., who gave us "HighFlight," the most memorable poem in aviation?

"On, I have slipped the surly bonds of earth...Put out my hand, and touched the face of God."

'4-O

MOTION SICKNESS

Motion sickness is a perennial aeromedical problem. Thisimportant. syndrome is discussed here to emphasize the criticalimportance of the spatial orientation senses in its pathogenesis.So closely entwined, in fact, are the mechanisms of spatialorientation and those of motion sickness, that orientationsickness is sometimes (and legitimately) used as the generalterm for the category of related conditions that we commonlyrefer to as motion sickness.

Definition, Description, and Significance of Motion Sickness

Motion sickness is a state of diminished health charac-terized by specific symptoms that occur in conjunction with, andin response to, unaccustomed conditions existing in cne'smotional environment. These symptoms usually progress fromlethargy, apathy, and stomach awareness to nausea, pallor, andcold eccrine perspiration, then to retching and vomiting, andfinally to total prostration if measures are not taken to arrestthe progression. The sequence of these major symptoms issufficiently predictable that vestibular scientists have deviseda commonly used scale, consisting of five steps from Malaise Ithrough Frank Sickness, to quantify the severity of motionsickness according to the level cf symptoms manifested (49)Other symptoms sometimes seen with motion sickness are headache,increased salivation and swallowing, decreased appetite, eructa-tion, flatulence, and feeling warm. Although vomiting sometimesprovides temporary relief from the symptoms of motion sickness,more commonly the motion-sick individual will continue to besick if the offending motion or other condition to which he isnot accustomed continues, and the vomiting will be replaced bynonproductive retching, or "dry heaves." A wide variety ofmotions and orientation conditions qualify as offensive, sothere are many species of the generic term, "motion sickness."Among them are seasickness, airsickness, car sickness, trainsickness, amusement-park-ride sickness, camel sickness, motionpicture sickness, flight simulator sickness, and the most recentaddition to the list: space sickness.

Military Experience. Armstrong (50) provides us with someinteresting statistics on airsickness associated with the WorldWar II military effort:

"It was learned that 10 to 11 percent of allflying students became airsick during their

first 10 flights, and that 1 to 2 percent ofthem were eliminated from flying trainingfor that reason. Other aircrew members intraining had even greater difficulty and the

07

airsickness rate among them ran as high as50 percent in some cases. It was also foundthat fully trained combat crews, other thanpilots, sometimes [developed airsickness]which affected their combat efficiency. Aneven more serious situation was found toexist among airborne troops. Under veryunfavorable conditions, as high as 70 per-cent of these individuals became airsick andupon landing were more or less temporarilydisabled at a time when their services weremost urgently needed."

More recent studies of the incidence of airsickness inUnited States and British military flight training reveal thatapproximately 40% of aircrew trainees become airsick at sometime during their training. In student pilots there is a 15% to18% incidence of motion sickness severe enough to interfere withcontrol of the aircraft. Airsickness in student aviators occursalmost exclusively during the first several training flights,during spin training, and during the first dual aerobaticflights. The adaptation of which most people are capable isevidenced in the fact that only about 1% of military pilottrainees are eliminated from flying training ber'ause of intrac-table airsickness. The percentage of other aircrew traineeseliminated because of airsickness is considerably higher,however.

Alt~iough trained pilots almost never become airsick whileflying the aircraft themselves, they surely can become sickwhile riding as copilot or as a passenger. Other trainedaircrew, such as navigators and weapon systems operators, arelikewise susceptible to airsickness. Particularly provocativefor these aircrew are flights in turbulent weather, low-level"terrain-following" flights, and flights in which high G forcesare repeatedly experienced, as in air combat training andbombing practice. Both the lack of foreknowledge of aircraftnotion, which results from not having primary control tf theaircraft, and the lack of a constant view of the external world,which results from having duties involving the monitoring ofin-cockpit displays, are significant factors in the developmentof airsickness in these aircrew.

Not surprisingly, seasickness is a hazard to aircrew whobail out over water. A life raft bobbing in the ocean providesa notoriously provocative stimulus for seasickness: in one studythree-fourths of the subjects became seasick and over halfvomited during one hour of exposure to artificial wave motionwhile riding in a life raft (51). The debilitating and demoral-izing effect of seasickness on the ability and will of downedaircrew to survive should be considered when planning sea rescueoperations.

103

Flight simulator sickness is getting increased attentionnow as aircrew spend more and more time in flight rimulators .capable of ever greater realism. Interestingly, simulatorsickness is more likely to occur in pilots having considerableexperience in the aircraft which the simulator is simulatingthan in pilots without such experience. The symptoms of simu- .Velk

lator sickness are for the most part those of the other forms ofmotion sickness, with nausea occurring in over 70% of pilotsflying certain simulators (52). Progression of symptoms to thepoint of vomiting is uncommon in simulator sickness, however.Simulators providing wide-field-of-view visual presentations arereported to produce the additional symptoms of kinestheticaftereffects, involuntary visual flashbacks, and disturbances ofbalance and locomotion for up to 10 hr after exposure (52). Forthat reason it is recommended that pilots get a good night'ssleep before flying real aircraft again after training inten-sively in wide-field-of-view flight simulatos,.

Civil Experience. The incidence of airsickness in flighttraining of civilians can only be estimated, but is probablysomewhat less than that for their military counterparts becausethe training of civil pilots usually does not include spins orother aerobatics. Fewer than 1% of passengers in today'scommercial air transport aircraft become airsick, largelybecause the altitudes at which these aircraft generally fly areusually free of turbulence. This cannot be said, however, forpassengers of most of the lighter, less capable, general aviationaircraft, who often must spend considerable portions of theirflights at the lower, "bumpier" altitudes.

Space Sickness. The challenge of space flight includescoping with space sickness, a form of motion sickness experiencedfirst by cosmonaut Titov and subsequently by approximately 29%of spacecrew (46 out of 157, as of this writing). The incidenceof sickness aboard the more recently launched, larger spacecraftis even greater (43%). We must therefore acknowledge thepotential impact of space sickness on manned space operationsand hope that its impact will be minimized by appropriatemission planning. The sheer expense associated with each spacelaunch, the high task loading of spacecrew, and the relativeirrevocability of each space mission once committed, make spacesickness a hazard of primary importance.

If the duration of a space mission is long enough to allow ecrewmembers to adapt to the motional environment of space, theyalso risk disadapting to that of earth, and can consequentlý -Esuffer earth sickness upon returning from the zero-G to the 1-Gcondition. Another type of space sickness will be encounteredin the event that large space stations are rotated to generate Gloading for the purpose of alleviating the fluid shift, cardio-vascular deconditioning, and skeletal demineralization thatoccur in the zero-G environment. Vestibular Coriolis effectscreated in occupants of such rotating systems are very potentproducers of motion sickness, and would be expected to plague

209 -

the occupants for several days after arrival. Again, to theextent they were to adapt to the rotating environment, theywould risk disadapting to the nonrotating one, and would sufferfrom motion sickness upon leaving the rotating space station.

Etiology

Man has speculated about the causes of and reasons formotion sickness for thousands of years. Largely because of thescientific interest in motion sickness that has been generatedby naval and aerospace activities of the present century, we maynow have a satisfactory explanation for this puzzling malady.

Correlating Factors. As already mentioned, motion sicknessoccurs in response to conditions to which one is not accustomedexisting in his motional environment. By "motional environment"we mean all of the linear and angular positions, velocities, andaccelerations that are directly sensed or secondarily perceivedby an individual as determining spatial orientation. Theprimary quantities of relevance here are mechanically (asopposed to visually) perceived linear and angular accelerations--those stimuli that act on the vestibular end-organs. Certainlythe pitching, rolling, heaving, and surging motions of ships inbad weather are clearly correlated with motion sickness, as -'oare the pitching, rolling, yawing, and positive and negativeG-pulling of aircraft during maneuvering. Abnormal stimulationof the semicircular ducts alone, as with a rotating chair, canresult in motion sickness. So also can abnormal stimulation ofthe otolith organs alone, as occurs in an elevator or a four-poleswing. Whether the stimulation provided is complex, as isusually the case on ships and in aircraft, or simple, as gen-erated in the laboratory, the important point is that abnormallabyrinthine stimulation is associated with the production ofmotion sickness. Not only is a modicum of abnormal vestibular estimulation sufficient to cause motion sickness, but some amountof vestibular stimulation is also necessary for motion sicknessto occur: labyrinthectonmized experimental animals, and humanswithout functioning vestibular end-organs (so-called "labyrin-thine defectives") are completely immune to motion sickness.

The visual system can play two very important roles in theproduction of motion sickness. First, motion sensed solelythrough vision can make some people sick. Examples of thisphenomenon are: motion picture sickness, in which wide-screenmovies of rides on airplanes, roller-coasters, and ships inrough seas are provocative; microscope sickness, in whichsusceptible individuals cannot tolerate viewing moving microscopeslides; and flight simulator sickness, in which wide-field-of-view visual motion systems create motion sickness in the absenceof any mechanical , 1otlon. Mrnolkial stimulation of ambient(peripheral) vision rather than focal (foveal) vision appears tobe the salient feature of visuall) induced motion sickness. Thefact that orientation information processed through the ambient 4visual sy-,em "onverges on the vestibular nuclei makes visually

110

"I

induced motion sickness less mysterious a phenomenon, even ifnot totally explicable. The second role of vision in theetiology of motion sickness is illustrated by the well-knownfact that absence of an outside visual reference makes personsundergoing abnormal motion more likely to become sick than theywould be if an outside visual reference were available. Casesin point: the sailor who becomes sick below deck but preventsthe progression of motion sickness by coming topside to view thehorizon, and the aircrewman who becomes sick while attending toduties inside the aircraft (e.g., radarscope monitoring) butalleviates symptoms by looking outside.

Other sensory systems capable of providing spatial orien-tational information are also capable of providing avenues formotion-sickness-producing stimuli. The auditory system, whenstimulated by a rcvolving sound source, is responsible foraudiogenic vertigo, audiokinetic nystagmus, and concomitantsymptoms of motion sickness. Nonvestibular proprioceptors maycontribute to the development of motion sickness when thepattern of stimulation of these senses by linear and angularaccelerations is unfamiliar. But perhaps more important thanthe actual sensory channel employed or the actual pattern ofstimulation delivered is the degree to which the spatial orien-tational information received deviates from that anticipated.The experience with motion sickness in various flight simulatorsbears witness to the importance of unexpected patterns of motionand unfulfilled expectations of motion. Instructor pilots inthe 2-FH-2 helicopter hover trainer, for example, were muchmore likely to become sick in the device than were studentpilots (53) . It is postulated that imperfections in flightsimulation are perceived by pilots who, as a result of theirexperience in the real aircraft, expect certain orientationa]stimuli to occur in response to certain control inputs. Pilotswithout time in the real aircraft, on the other hand, have nosuch expectations, and therefore notice no deviations from themin the simulator. Another example of the role played by expecta-tion of motion in the generation of motion sickness is seen inthe pilot who does not become sick as long as he has control ofthe airplane, but does become sick when another pilot is flyingthe same maneuvers in the same airplane. In this case thepilot's expectation of motion is always fulfilled whenever he iscontrolling the airplane, but is not when someone else is flying.

Several other variables not primarily related to spatialorientation seem to correlate well with motion sickness suscepti-bility. Aqe is one: susceptibility increases with age untilpuberty, then decreases thereafter. Sex is another: women aremore susceptible than men at any given age. The personalitycharacteristics of emotional lability and excessive rigidity arealso positively correlated with motion sickness susceptibility.Whether one is mentally occupied with a significant task duringexposure to motion, or is free to dwell on orietiLation cues andthe state of his stomach, seems to affect his susceptibility--thelatter, more introverted state being the more conducive to

ill

motion sickness. Likewise, anxiety, fear, and insecurity,either about one's orientation relative to the ground or aboutone's likelihood of becoming motion sick, seem to enhancesusceptibility. We must be careful, however, to distinguishbetween sickness caused by anxiety and sickness caused bymotion: a paratrooper who vomits in an aircraft while waitingto jump into battle may be suffering either from extreme anxietyor from motion sickness, or both. Finally, we must recognizethat many things, such as mechanical stimulation of the visceraand malodorous aircraft compartments, even though they arecommonly associated with conditions that result in motionsickness, do not in themselves cause motion sickness.

A mildly interesting but potentially devastating phenomenonis conditioned motion sickness. Just as Pavlov's canine subjectslearned to salivate at the sound of a bell, student pilots andother aircrew, with repeated exposure to the conditioningstimulus of sickness-producing aircraft motion, can eventuallydevelop the autonomic response of motion sickness to the condi-tioned stimulus of being in, or even just seeing, an aircraft(Fig. 43). For this reason it is advisable to initiate aircrewgradually to the abnormal motions of flight, and to providepharmacologic prophylaxis against motion sickness, if necessary,in the early instructional phases of flight.

Unifying TheorL. Current thinking regarding the underlyingmechanism of motion sickness has focused on the "sensoryconflict" or "neural mismatch" hypothesis, proposed originallyby Claremont in 1931 (54). In simple terms, the sensory conflicthypothesis states that motion sickness results when incongruousorientation information is generated by various sensory modali-ties, one of which must be the vestibular system. In virtuallyall examples of motion sickness one can, with sufficient scru-tiny, identify the sensory conflict involved. Usually theconflict is between the vestibular and visual senses or betweenthe different components of the vestibular system, but conflictsbetween vestibular and auditory or vestibular and nonvestibularproprioceptive systems are also possible. A clear example ofsickness resulting from vestibular-visual conflict is that whichoccurs when an experimental subject wears reversing prisms overhis eyes so that his visual perception of self-motion is exactlyopposite in direction to his vestibular perception of it.Another example is motion-picture sickness, the conflict beingbetween visually perceived motion and vestibularly perceivedstationarity. Airsickness and seasickness are most often aresult of vestibular-visual conflict: the vestibular signals oflinear and angular motion are not in agreement with the visualpercept of being stationary inside the vehicle. Vestibular-visual conflict need not even be in relation to motion, but canbe in relation to static orientation: some people become sickin antigravity houses, which are built in such a way that thevisually apparent vertical is quite different from the true

112

%IN..

Figure 43. Conditioned motion sickness. A student aviator whorepeatedly gets airsick during flight can becomeconditioned to develop symptoms in- response to thesight or smell of an aircraft even before flight. Use

Sof antimotion-sickness medication until the studentadapts to the novel motion can prevent conditionedmotion sickness.

gravitational vertical. Intravestibular conflict i8 an espe-cially potent means of producing motion sickness. Whet. vestib-ular Coriolis effects cause the semicircular ducts to signalfalsely that angular velocity about a rionvertical axis isoccurring, and the otolith organs do not confirm a resultingchange in angular position, the likelihood of developing motionsickness Jis great. In a zero-gravity environment, when onemoves his head u.- or down the semicircular ducts sense rotationbut the otolith organs cannot sense any resulting change ofangular position relative to a gravity vectorl the generation ofthis intravestibular conflict is believed to be the underlyingmechanism of space sickness. It can also be argued, however,that the sensory conflict responsible for s~pace sickness isbetween the otolith organs, which detect no gravitational

113

+0

S -. *...*.%..*.* *.~..*. *-. ~ --. 1.- - - - - - - - - -

vertical, and the eyes, to which structures within the spacevehicle take on apparent verticality by virtue of some familiaraspect of their configuration.

But what determines whether orientation information isconflicting or not? It is one's prior experience in the motionalenvironment, and the degree to which orientation informationexpected on the basis of that experience agrees with the actualorientation information received. Thus, according to Reason(55), the important sensory conflict is not so much an absolutediscrepancy between information from the several sensory modali-ties as it is between anticipated and actual orientation informa-tion. We see evidence of this in the gradual adaptation tosustained abnormal motional environments such as the sea, space,slow rotation, and the wearing of reversing prisms, and in the - -

readaptation to the normal environment that must take place uponreturning to it. We also see that being able to anticipateorientation cues confers immunity to motion sickness, as evi-denced by the fact that pilots and automobile drivers almostnever make themselves sick, and by the fact that we activelysubject ourselves to many motions (jumping, dancing, acrobatics)that would surely make us sick if we were to be subjected tothem passively. It appoars, then, that the body refers to aninternal model of orientatiunal dynamics, both sensory andmotor, to effect voluntary and involuntary control over orien- - -

tation (17) . When transient discrepancies between predicted andactual orientation data occur, corrective reflex activity isinitiated, or the internal model is updated, or both. But whensustained discrepancies occur, motion sickness is the result.

Teleology. Even if the mechanism of motion sickness could - -

be described completely in terms of cellular dnd subcellularfunctions, the question would remain; "What purpose, if any, -<

does motion sickness serve?" The idea that a chance mutation -.rendered countless generations of vertebrates potential victims 6of motion sickness, and that the relatively recent arrival oftransportation systems gave expression to that otherwise innoc-uous genetic flaw, strains credulity. Steele has suggested thatthe symptoms of motion sickness are "caused by cardiovascularinadequacy secondary to diversion of circulating blood to themuscles in response to a threatened need for vigorous muscularaction on the basis of inadequately perceived inertial anddynamic environment" (56). A more recent hypothesis, and in ouropinion a more satisfying explanation of motion sickness, isthat of Treisman (57) He proposed that the orientation senses,in particular the vestibular system, serve an important functionin the emetic response to poisons. When an animal ingests a 4toxic substance and experiences its effects on the centralnervous system--riamely, deterioration of the finely tunedspatial orientation senses and consequeýnt degraded predictabilityof sensory responses to motor activity--reflex vomiting occursand the animal is relieved of the poison. The positive survivalvalue of such a mechanism to eliminate ingested poisons isobvious. The essentiality of Llhe vestibular end-organs apn.

. .f

p.%

certain parts of the cerebellum, and the role of sensory conflict.

as manifested through the functioning of those structures, areprovided a rational basis in Treisman's theory. Finally,experimental support for Treisman's theory has been provided by 0Money and Cheung (58) who found that labyrinthectomized animals,in addition to being immune to motion sickness, exhibit markedimpairment of the emetic response to certain naturally occurringpoisons. p.

Prevention and Treatment

The variety of methods at our disposal for preventing andtreating motion sickness is less an indication of how easymotion sickness is to uontrol than it is of how incompletelyeffective each method can be. Nevertheless, logical medicalprinciples are generally applicable; and several specifictreatments have survived the test of time and become tradi-tionalized, while some newer approaches appear to have greatpotential.

Physiologic Prevention. An obvious way to prevent motionsickness is to avoid environments that produce it; but for manyindividuals in today's world, this is neither possible nordesirable. The most common and ultimately most successful wayis to adapt to the novel motional environment through constantor repeated exposure to it. The rapidity with which adaptationoccurs is highly variable, depending mainly on the strength ofthe challenge and on the adaptability of the individual involved.Usually, several days of sustained exposure to mild orientationalchallenges (like sea and space travel) or several sessions ofrepeated exposure to vigorous challenges (like aerobatics orcentrifuge riding) will confer immunity. The use of antimotion-sickness medications to prevent symptoms during the period ofadaptation does not appear to compromise the process of adapta-tion and is reconmmended where practicable.

Selection of individuals resistant to motion sickness, orscreening out those unusually susceptible to it, has beenconsidered as a method for reducing the likelihood of motionsickness in certain operations, such as military aviationtraining. The fact that susceptibility to motion sickness is socomplex a characteristic makes selection less efficacious ameans of prevention than might be supposed. At least threeseparate factors are involved in motion sickness susceptibility,as identified by Reason and Brand (59): receptivity, the degreeto which a given orieritational information conflict is perceivedand the intensity with which it is experienced and responded to;adaptability, the rate at which one adjusts to a given abnormalorientational environment as evidenced by his becoming less andless symptomatic; and retentivity, the ability to remain adaptedto the novel environment after leaving it. These factors appearto be independent. This means a paruicular prospective aviatorwith high receptivity might also adapt very rapidly and aemairiadapted for a long time, so that it would be unwise to eliminate

115

,.ihim from flying training on the basis of a history of motionsickness or even a test of susceptibility. Nevertheless,although the great majority of aircrew trainees do adapt to theaerial environment, use of either vestibular stimulation (60) ora motion sickness questionnaire (61) reveals that sensitivity tomotion sickness is inversely related to success in flyingtraining. Furthermore, sound judgment dictates that an attemptto select against crewmembers with a high probability of beccmingmotion sick is appropriate for some of the more critical andexpensive aerospace operations.

Some very promising results have been obtained recentlywith biofeedback-mediated behavior modification 2n± reducingsusceptibility to motion sickness. Cowings and Toscano (62)have reported that subjects receiving autogenic-feedback training(AFT), a biofeedback-based autonomic conditioning procedure,were able to adjust significantly more rapidly to nauseogenicvestibular Coriolis stimulation than were control subjects notprovided with AFT. But can biofeedback be used to rehabilitateaircrew suffering from refractory airsickness? Apparently so,as Levy et al. (63) found that the use of relaxation trainingand biofeedback in the treatment of highly motivated aircrew whowere disabled by recurrent airsickness has yielded an 84% rateof return to duty; this rate is substantially better than the45-50% rate for a comparable group of aircrew not provided thebiofeedback treatment.

Phsiologic Treatment. Once symptoms of motion sicknesshave developed, the first step to take to bring about recovery isto escape from the environment producing the symptoms. If thisis possible, relief usually follows rapidly; but symptoms canstill progress to vomiting, and nausea and drowsiness cansometimes persist for many hours, even after termination of theoffending motion. If escape is not possible, assuming a supineposition, or just stabilizing the head, seems to offer some ,relief. As mentioned previously, passengers subjected to motionin enclosed vehicles can help alleviate symptoms by obtaining aview of the natural horizon. One of the most effective physio-logic remedies is turning over control of the vehicle to thesymptomatic crewmnember. Generations of flight instructors haveused this technique to avert motion sickness in their students,even though they were probably unable to explain how it works interms of reducing conflict bctwcen anticipatcd and actualorientation cues. Another procedure which has proved useful inpractice is to cool the affected individual with a blast of airfrom the cabin air vent; such thoughtfulness on the part oftheir instructors has undoubtedly saved many student pilots fromhaving to clean up the cockpit.

Pharmacologic Prevention. The most effective singlemedication for prophylaxis against motion sickness is scopolamine(1-hyoscine), 0.3 to 0.6 rrg, taken orally 30 min to 1 hr beforeexposure to notion. Upfortunately, when scopolamine is taken inoral~y efiective doses, the side effects (i.e., drowsiness, dry %

] 6

,/*'

mouth, pupillary dilation, and paralyzed visual accommodation)make routine oral administration of this drug to aircrew highlyinadvisable. When prophylaxis is needed for prolonged exposureto abnormal motion (e.g., an ocean voyage), oral scopolamine canbe administered every 4 to 6 hry but again, the side effects aretroublesome and may preclude repeated oral administration. Anovel approach to the problem of prolonged prophylactic admin-istration of scopolamine is the Transderm-ScopO system, in which0.5 mg of scopolamine is delivered transcutaneously over a 3-dayperiod from a small patch worn on the skin behind the ear.Although the side effects associated with this system of admin.-istration are reportedly less than with oral scopolamine, athorough evaluation of its efficacy has not yet been undertaken.

The antimotion-sickness preparation most commonly used inaircrew is probably the "scope-dex" combination, 0.3-0.6 mg ofscopolamine and 5-10 mg of dextroamphetamine, taken orally 1 hrprior to exposure to motion. Not only is this combination moreeffective than scopolamine alone, but the stimulant effect ofthe dextroamphetamine counteracts the drowsiness side effect ofthe scopolamine. Another very useful oral combination is 25 mgeach of promethazine and ephedrine, taken approximately 1 hrbefore exposure. As the individual response to the severaleffective antizmotion-sickness preparations is highly variable,it may be worthwhile to perform individual assessments ofdifferent drug combinations and dosages to obtain the maximum V.benefit. The reader is referred to Graybiel et al. (64) andJohnson et al. (65) for additional information on efficacy ofantimotion-sickness preparations.

Pharmacologic Treatment. If motion sickness progresses tothe point of nausea, and certainly if to vomiting, oral medica-tion is useless. If the prospect of returning soon to the 4t,

accustomed motional environznent is remote, it is important totreat the condition to prevent the dehydration and electrolyteloss that result from protracted vomiting. Intramuscularinjection of scopolamine, 0.3 mg, or promethazine, 25 mg, isrecommended. Scopolamine administered intravenously or even bynasal spray or drops is also effective. Promethazine rectalsuppositories are used to control vomiting in many clinicalsituations, and their use in treatment of motion sickness shouldalso be successful. If parenteral administration of scopolaminewith intravenous phenobarbital may be necessary to preventSor prorrethazine does not provide relief froz,• vonit~ing, sedation...

p~rogressive deterioration of the patient's condition. Ofcourse, fluid and electrolyte 2osses must be replaced in patientswho have been vomiting for prolonged periods.

Aeromedical Use of Antimrotion-sickness Preparations. Asmentioned previously, routine use of antimotion--sickness drugsin aircrew is not appropriate because of the undesirable sideeffects of these drugs. Prophylactic medication can be veryuseful, howevei, in helping the student aviator cope with the

117,

.. .-

novel motions that can cause sickness during flight training--thus promoting better conditions for learning, and preventingthe development of conditioned motion sickness. Prophylaxis canalso help reduce a student's anxiety over becoming motion sick,which can develop into a self-fulfilling vicious circle. Afterusing medication, if necessary, for two or three dual trainingsorties--usually at the beginning of flight training and againduring the introduction to aerobatics--student pilots should nolonger need antimotion-sickness drugs. Use of drugs for soloflight should absolutely be forbidden. A more liberal approachcan perhaps be taken with other aircrew trainees, such asnavigators, because of their greater propensity to become motionsick and their less critical influence on flight safety. %Trained aircrew, as a rule, should not use antimotion-sicknessdrugs. An exception to this rule is made for spacecrew, whoseexposure to the zero-gravity condition of space flight isusually very infrequent, and whose pre-mission adaptation byother means cannot be assured. Spacecrew should also be expectedto need prophylaxis for reentry into the normal gravitationalenvironment of the earth after a prolonged stay at zero gravity.Airborne troops, who must arrive at the battle zone fullyeffective, are also candidates for antimotion-sickness prophy-laxis under certain circumstances, such as prolonged low-levelflight in choppy weather. In all such cases the flight surgeonmust weigh the risks associated with developing moti.on sicknessagainst the risks associated with the side effects of theantimotion-sickness drugs, and arrive at a judgment of whethcror not to medicate.

Thus we see how man's recent transition into the motionalenvironment of aerospace has introduced him not only to newsensations but also to new sensory demands. If he fails toappreciate the fallibility of his natural orientation senses inthis novel environment, he succumbs to spatial disorientation.If he recognizes his innate limitations, however, he can meetthe demands of the environment and function effectively in it.We see also how man 's phyiogenetic heritage, by means of orien-tationial mechanisn1,, renders him susceptible to motion sickness.That same heritage, however, enables him to adapt to new motionalenvironments. The profound and pervasive influence of ourorientation senses in aerospace operations cannot be denied orignored; but through knowledge and understanding, it can becontrolled.

• .. "

118 "'0

3.

%3

REFERENCES

1. Hixson, W.C., Niven, J.I., and Correia, M.J.: Kinematicsnomenclature for physiological accelerations, with Vspecial reference to vestibular applications. Mono-graph 14, Naval Aerospace Medical Institute, Pensacola,FL, 1966.

2. Daroff, R.B., and Troost, B.T.: Supranuclear disorders ofeye movements. 12 Clinical Ophthalmology, Volume II.Glaser, J. (ed.). New York, Harper and Row, 1976.

3. Henn, V., Young, L.R., and Finley, C.: Vestibular nucleusunits in alert monkeys are also influenced by movingvisual fields. Brain Res 71:144-149, 1974.

4. Dicbgans, J., and Brandt, T.: Visual-vestibular interaction: "".effects on self-motion perception and postural control.In Handbook of Sensory Physiology. Volume VIII: Percep-tion. Held, R., Liebowitz, H.W., and Teuber, H.-L. (eds.).Berlin, Springer-Verlag, 1978.

5. Grusser, O.-J., and Gr{isser-Cornehls, U.: Chapter 6. Neuronalmechanisms of visual movement perception and some psycho-physical and behavioral correlations. 11 Handbook ofSensory Physiology, Volume VII/3: Central Processing ofVisual Information; A: Integrative Functions and Compara-tive Data. Jung, R. (ed.). Berlin, Springer-Verlag, 1973.

6. Liebowitz, H.W., and Dichgans, J.: The ambient visual systemand spatial orientation. L Spatial Disorientation inFlight: Current Problems. AGARD-CP-287. NATO/AGARD,Neuilly-sur-Seine, France, 1980.

7. Tredici, T.J.: Visual illusions as a probable cause ofaircraft accidents. 1n Spatial Disorientation in Flight:Current Problems. AGARD-Ci'-287. NATO/AGARD, Neuilly-sur- N.Seine, France, 1980.

8. Ditchburn, R.W., and Ginsborg, B.L.: Vision with a stabi- i1lized retinal image. Nature 170:36-37, 1952. I.

9. Brindley, G.S., and Merton, P.A.: The absence of positionsense in the human eye. J Physiol 153:127-130, 1960.

10. Sperry, R.W.: Neural basis of the spontaneous optokineticresponse preceded by visual inversion. J Comp PhysiolPsych 43:482-489, 1950.

,- ,.-

139 I

11. Cohen, M.M.: Chapter 6. Visual-proprioceptive interactions.IL IntersensoLy Perception and Sensory Integration.Walk, R.D., and Pick, H.L., Jr. (eds.). New York, PlenumPublishing Corporation, 1981.

12. Spoendlin, H.H.: Ultrastructural studies of the labyrinth insquirrel monkeys. IM The Role of the Vestibular Organs inthe Exploration of Space. NASA SP-77, National Aeronauticsand Space Administration, Washington, DC, 1965.

13. Jones, G.M.: Disturbance of oculomotor control in flight. -

Aerospace Med 36:461-465, 1965.

14. Mann, C.W., and Ray, J.: Absolute thresholds of perception ofdirection of angular acceleration. Joint Proj. ReportNo. 41, Naval School of Aviation Medicine, Pensacola, FL,May 1956. I-

15. Meiry, J.L.: The vestibular system and human dynamic spaceorientation. Doctoral thesis, Massachusetts Institute ofTechnology, Cambridge, MA, June 1965.

16. Gillingham, K.K.: Training the vestibule for aerospace Voperations: central control of vestibular function.Military Med 131:696-704, 1966.

17. Oman, C.M.: Space motion sickness and vestibular experimentsin Spacelab. SAE Technical Paper 820833, presented at theTwelfth Intersociety Conference on Environmental Systems,San Diego, CA, 19-21 July 1982.

18. Handbook of Sensory Physiology, Volume II: SomatosensorySystem. Iggo, A. (ed.). Berlin, Springer-Verlag, 1973.

19. Handbook of Sensory Physiology, Volume 111/2: Muscle Recep-tors. Hunt, C.C. (ed.). Berlin, Springer-Verlag, 1974.

20. McIntyre, A.K..: Chapter 3. Central actions of impulses inmuscle afferent fibers. In Handbook of Sensory Physiology,Volume 111/2: Muscle Receptors. Hunt, C.C. (ed.). Berlin,Springer-Verlag, 1974.

21. Pitts, D.G.: Visual illusions and aircraft accidents. SAM-TR-67-28, USAF School of Aerospace Medicine, Brooks AFB, TX,1967.

22. Kraft, C.L.: A psychophysical contribution to air safety: -simulator studies of visual illusions in night visualapproaches. ID Psychology--From Research to Practice.Pick, H.L., Jr. et al. (eds.). New York, Plenum Press,1978.

120 4-S

23. Peters, R.A.: Dynamics of the Vestibular System and TheirRelation to Motion Perception, Spatial Disorientation, andIllusions. NASA-CR-1309, National Aeronautics and SpaceAdministration, Washington, DC, 1969. _

24. Jones, G.M. : Vestibulo-ocular disorganization in the aero-dynamic spin. Aerospace Med 36: 976-983, 1965.

25. Graybiel, A., and Hupp, D.I.: The oculo-gyral illusion: aform of apparent motion which may be observed followingstimulation of the semicircular canals. Report No. 4,Naval School of Aviation Medicine, Pensacola, FL, 1945.

26. Gillingham, K.K.: Some notes on the threshold of thevestibular Coriolis effect and its significance toaircrew. SAM-TR-65-55, USAF School of Aerospace Medicine,Brooks AFB, TX, 1965. .

27. Cohen, M.M., Crosbie, R.J., and Blackburn, L.H.: Disorientingeffects of aircraft catapult launchings. Aerospace Med V44:37-39, 1973.

28. Buley, L.E., and Spelina, J.: Physiological and psychologicalfactors in "the dark night takeoff accident." AerospaceMed 41:553-556, 1970. "-

29. Martin, J.F., and Jones, G.M.: Theoretical man-machineinteraction which might lead to loss of aircraft control. VAerospace Med 36:713-716, 1965. o

30. Malcolm, R., and Money, K.E.: Two specific kinds of disorien-tation incidents: jet upset and giant hand. IL The Dis-orientation Incident. AGARD-CP-95-Part 1. Benson, A.J.(ed.). NATO/AGARD, Neuilly-sur-Seine, France, 1972.

31. Gilson, R.D., et al.: Observations on perceived changes inaircraft attitude attending head movements made in a 2-Gbank and turn. Aerospace Med 44:90-92, 1973.

32. Graybiel, A.: Oculogravic illusion. Arch Ophthal 48:605-615,1952.

.4"j

33. Nuttall, J.B., and Sanford, W.G.: Spatial disorientation inoperational flying. Publication M-27-56, USAF Directorateof Flight Safety Research, Norton AFB, CA, 12 Sep 1956.

34. Moser, R.: Spatial disorientation as a factor in accidents inan operational command. Aerospace Med 40:174-176, 1969.

35. Barnum, F., and Bonner, R.H.: Epidemiology of USAF spatialdisorientation aircraft accidents, 1 Jan 1958-31 Dec 1968.Aerospace Med 42:896-898, 1971.

12]1

36. Barnum, F.: Spatial disorientation aircraft accidents,1 Jan 1969-31 Dec 1971: a comparison. Presented at theAnnual Scientific Meeting of the Aerospace MedicalAssociation, Las Vegas, NV, 7-10 May 1973.

37. Kellogg, R.S.: Letter report on spatial disorientationincidence statistics. From Aerospace Medical ResearchLaboratory, Wright-Patterson AFB, OH, to AMD/RDL,30 Mar 1973.

38. Hixson, W.C., Niven, J.I., and Spezia, E.: Orientation-erroraccidents in Army aviation aircraft. In The DisorientationIncident. AGARD-CP-95-Part I. Benson, A.J. (ed.). NATO/AGARD, Neuilly-sur-Seine, France, 1972.

39. Tyler, P.E., and Furr, P.A.: Disorientation, "fact andfancy." 1n The Disorientation Incident. AGARD-CP-95-Part I. Benson, A.J. (ed.). NATO/AGARD, Neuilly-sur-Seine,France, 1972.

40. Kirkham, W.R., et al.: Spatial disorientation in generalaviation accidents. Aviat Space Environ Med 49:1080-1086, 1978.

41. McCabe, B.F.: Vestibular suppression in figure skaters. TransAmer Acad Ophthal Otolaryngol 64:264, 1960.

42. Collins, W.E.: Vestibular responses from figure skaters.Aerospace Med 37:1098-1104, 1966.

43. Money, K.E., and Malcolm, R.E.: Assessment of films andslides for orientation training of military pilots. InOrientation/Disorientation Training of Flying Personnel:A Working Group Report. AGARD-R-625. Benson, A.J. (ed.).NATO/AGARD, Neuilly-sur-Seine, France, 1974.

44. Malcolm, R., et al.: The Malcolm horizon. Presented at theAnnual Scientific Meeting of the Aerospace MedicalAssociation, New Orleans, LA, 8-11 May, 1978.

45. Money, K.E.: Theory underlying the Peripheral VisionHorizon Device. DCIEM Technical Communication 82-C-57,Defence and Civil Institute of Environmental Medicine,Downsview, Ontario, Canada, Dec 1982.

46. Wick, R.L.: No flicker vertigo. Letter to the editor.Business/Commercial Aviation 51:16, July 1982.

47. Clark, B., Nicholson, M., and Graybiel, A.: Fascination: acause of pilot error. J Aviat Med 24:429-440, 1953.

48. Clark, B., and Graybiel, A.: The Break-off phenomenon--afeeling of separation from the earth experienced bypilots at high altitude. J Aviat Med 28:121-126, 1957.

'1 11 22 "

49. Miller, E.F., II, and Graybiel, A.: Comparison of five levelsof motion sickness severity as the basis for gradingsusceptibility. Aerospace Med 45:602-609, 1974. .-

50. Armstrong, H.G.: Chapter 14. Air sickness. In AerospaceMedicine. Armstrong, H.G. (ed.). Baltimore, The Williams & 1

Wilkins Company, 1961.

51. Benson, A.J.: Chapter 22. Motion sickness. In AviationMedicine, Volume I. Physiology and Human Factors.Dhenin, G. (ed.). London, Tri-Med Books Limited, 1978.

52. Kellogg, R.S., Castore, C., and Coward, R.E.: Psychophysio-logical effects of training in a full-vision simulator.Preprints of the Annual Scientific Meeting of the Aero-space Medical Association, Anaheim, CA, 12-15 May, 1980.

53. Miller, J.W., and Goodson, J.E.: Motion sickness in ahelicopter simulator. Aerospace Med 31:204-212, 1960.

54. Claremont, C.A.: The psychology of sea sickness. Psyche11:86-90, 1931.

55. Reason, J.T.: Motion sickness adaptation: a neural mismatchmodel. J Roy Soc Med 71:319-829, 1978.

56. Steele, J.E.: Motion sickness and spatial perception--atheoretical study. la Symposium on Motion Sickness withSpecial Reference to Weightlessness. AMRL-TDR-63-25, Aero-space Medical Research Laboratory, Wright Patterson AFB,OH, 1963.

57. Treisman, M.: Motion sickness: an evolutionary hypothesis.Science 197:493-495, 1977.

58. Money, K.E., and Cheung, B.S.: A mechanism for facilitationof the emetic response to poisons: the basis of motionsickness. Preprints of the Anniual Scientific Meeting ofthe Aerospace Medical Association, Bal Harbour, FL,10-13 May, 1982.

59. Reason, J.T., and Brand, J.J.: Motion Sickness. London,Academic Press, 1975.

60. Ambler, R.K., and Guedry, F.E., Jr.: Reliability and validityof the Brief Vestibular Disorientation Test under 10-rpmnand 15-rpm conditions. Aerospace Med 42:186-189, 1971,

61 Kennedy, R.S. : Motion sickness questionnaire and fieldindependence scores as predictors of success in navalaviation training. Aviat Space Environ Med 46:1349-1352,

V.,..o1975.

12 3

62. Cowings, P.S., and Toscano, W.B.: The relationship of motionsickness susceptibility to learned autonomic control forsymptom suppression. Aviat Space Environ Med 53:570-575,1982.

63. Levy, R.A., Jones, D.R., and Carlson, E.H.: Biofeedbackrehabilitation of airsick aircrew. Aviat Space Environ Med52:118-121, 1981.

64. Graybiel, A., et al.: Human assay of antimotion sicknessdrugs. Aviat Space Environ Med 46:1107-1118, 1975.

65. Johnson, W.H., Money, K.E., and Graybiel, A.: Airbornetesting of three antimotion sickness preparations. AviatSpace Environ Med 47:1214-1216, 1976. %..

/% -''a

Ih: N'. '%.

"I".

- 'a' - a.- .

RECOMMENDED READING

Handbook of Sensory Physiology, Volume II: Somatosensory System.Iggo, A. (ed.). Berlin, Springer-Verlag, 1973.

Handbook of Sensory Physiology, Volume 111/2: Muscle Receptors.Hunt, C.C. (ed.). Berlin, Springer-Verlag, 1974.

Handbook of Sensory Physiology, Volume VI: Vestibular System.Kornhuber, H.H. (ed.). Berlin, Springer-Verlag, 1974.

Handbook of Sensory Physiology, Volume VII/3: Central Processingof Visual Information. Jung, R. (ed.) . Berlin, Springer-Verlag, 1973.

Handbook of Sensory Physiology, Volume VIII: Perception. Held, .V-R. , Liebowitz, H.W., and Teuber, H.-L. (eds.) . Berlin,Springer-Verlag, 1978.

Wagman, I H. , and Dong, W.K.: Principles of Biodynamics,Volume III: Physiological Mechanisms in the Mammal UnderlyingPosture, Locomotion, and Orientation in Space. AeromedicalReview 7-75, USAF School of Aerospace Medicine, Brooks AFB,TX, 1975.

Wilson, V.J., and Jones, G.M.: Mammalian Vestibular Physiology.New York, Plenum Press, 1979. -.

Howard, I.P., and Templeton, W.B.: Human Spatial Orientation.London, John Wiley & Sons, 1966.

Peters, R.A. : Dynamics of the Vestibular System and theirRelation to Motion Perception, Spatial Disorientation, andIllusions. NASA CR-1309, National Aeronautics and SpaceAdministration, Washington, DC, 1969.

Benson, A.J. , and Burchard, E. : Spatial Disorientation inFlight--A Handbook for Aircrew. AGARD-AG-170, NATO/AGARD,Neuilly-sur-Seine, France, 1973.

Money, K.E.: Motion Sickness. Physiological Reviews 50:1-39,1970.

Reason, J.R., and Brand, J.J.: Motion Sickness. London, AcademicPress, 1975.

123.