current methodology and methods in psychophysiological studies of creative thinking

www.elsevier.com/locate/ymeth

Methods 42 (2007) 100–108

Current methodology and methods in psychophysiologicalstudies of creative thinking

N.P. Bechtereva *, S.G. Danko, S.V. Medvedev

Institute of the Human Brain, Russian Academy of Sciences, ul.acad.Pavlova 9, Sankt-Peterburg 197376, Russia

Accepted 7 January 2007

Abstract

Important points on methodology and detailed description of methods used in polymodal psychophysiological studies of human ver-bal creative thinking are presented. The psychophysiological studies were conducted with healthy volunteers during implementations ofspecially developed and adapted psychological tests aimed to bring the subjects into states of verbal creative thinking. Four different tasksets (‘‘story composition’’, ‘‘associative chains’’, ‘‘original definitions’’, ‘‘proverb sense flipping’’) were developed and applied. Positronemission tomography of regional cerebral blood flow (rCBF) and state-related quantitative electroencephalography (power and coher-ence evaluated) were used. The effectiveness of the methods is illustrated with figures.� 2007 Elsevier Inc. All rights reserved.

Keywords: Psychophysiology; Creativity; Creative thinking; Methods; Tests of creative thinking; Regional cerebral brain flow; Quantitative electroenc-ephalography

1. Introduction

Contemporary methodologies in cognitive neuroscienceallow us to answer questions about the psychophysiologyof mental performance in the normal and diseased brain.Scientific work today provide us with ample evidence thatthere is rapid progress in studying the highest form ofhuman mental activity–creative activity. The complexityof the subject under investigation and the different experi-mental paradigms used account for the fact that the datacurrently available do not always agree well. As such psy-chological tests and physiological methods should be delib-erately varied according to the aims of the study and thephenomena investigated for the simple reason that no sin-gle method is the only one for solving a problem of suchcomplexity. For the same reason, close cooperation of sev-eral research groups is not only highly desirable but is ‘‘amust’’ in order to achieve genuine progress. Furthermore,

1046-2023/$ - see front matter � 2007 Elsevier Inc. All rights reserved.

doi:10.1016/j.ymeth.2007.01.009

* Corresponding author. Fax: +7 812 2343247.E-mail address: [email protected] (N.P. Bechtereva).

clarity in the methods is a necessary condition for suchcooperation and the publication of the present volume‘‘Models for the study of creativity in animals and man’’is a timely and valuable contribution to the field in thisregard.

In the present paper we explain some essential aspects ofour methodology. We also provide a detailed description ofthe methods used in our psychophysiological studies ofhuman verbal creative thinking. All studies were conductedwith healthy volunteers, using psychological tests speciallydesigned to bring the subjects into states of verbal creativethinking. Positron emission tomography (PET) measuringregional cerebral blood flow (rCBF) and state-relatedquantitative electroencephalography (EEG) were our maintools. The effectiveness of the methods is illustrated in var-ious figures.

2. Methodology, general considerations

Studies of creativity are considered a higher level ofresearch into brain and mentality, its further progressand evolution. Due to the integration of cognitive psychol-

mailto:[email protected]

N.P. Bechtereva et al. / Methods 42 (2007) 100–108 101

ogy, neuropsychology and cognitive neurophysiologyachieved during the last decade, it has become possible toattack this problem. The latest advancements in technol-ogy, especially rCBF investigations using PET and fMRI,play a particularly important role here.

As a science, the psychophysiology of creative thinkingis still in its infancy. There is but a small number of publi-cation on the topic, most on either EEG [1–11] or PET [12–17], which is due mainly to the complexity of the problem.But even this relatively modest number of studies on thebrain’s creative mechanisms has been of great importanceto elucidating a wide range of problems, be they philosoph-ical, medical, pedagogical or technical.

The current psychophysiological literature [18] containsmore than 60 different definitions of creativity. Thus, crea-tivity cannot be considered as being rigorously defined, butthere is a certain consensus that creativity yields somethingpartly or entirely new; gives existing objects new propertiesor characteristics; allows one to imagine new potentialitiesnot conceived of before and to see or perform something ina manner different from what was thought possible or nor-mal previously. These concepts also guided our researchand the selection and design of our psychological tests.

A distinction between the creative person, the creativeproduct, the creative process, and the creative influenceof the environment has been outlined by Rhodes [19]. Eachof these factors is an important component of creativeactivity. In our studies, we have concentrated on the crea-tive process (creative thinking in particular), assuming thatcreativity is a common trait of people and, as such, more orless inherent in any healthy human being.

Tests designed to study the living brain with a conven-tional contrast analysis, are the main obstacles in thisregard. This is due to the intimate link between the creativeprocess and on-going brain activity. One should seek to dif-ferentiate between the effects caused by a creative processand on-going activity on the one hand, and those causedby stress factors of mental and emotional loads. This is nec-essary so that the research can distinguish between theeffects of quantitative brain resource mobilization andthose of system dynamics inherent in the activities. Thepsychologist faces a challenge here that is like that of Scillaand Haribda, (When sailing through a strait Odisseus hadto choose between two evils, the Scilla rock on one side ofthe strait and the storming Haribda waters on the other).

This makes it very difficult to study brain correlates ofcreative thinking. In terms of methodology, this situationdemands, that ‘‘double tests’’ be introduced. That is, atleast two variants of creativity and control tasks shouldbe applied in the same experiment. These variations haveto allow the analysis of unavoidable factors, primarilythose intimately tied to the complexity of the task. Themechanisms that are potentially important in the imple-mentation and optimization of creative thinking must beconsidered as early as possible.

A wide scope of psychophysiological investigations ,explicating underlining brain mechanisms reveal the expe-

diency and necessity of polymethodical (polymodal) phys-iological description/analysis of local and global braindynamics. Studies combining brain hemodynamics (PET/fMRI) and electrodynamics (EEG/MEG) data are cur-rently considered to be the best approach to deal with thechallenge [20]. Polymethodical results have to be treatedas complementary ones. Any divergence in the databetween these methods can be even more informative thanthe convergence because of the principal differences in thecorrelations of the measurable values in synaptic activities[21–23]. This is why we attempt to utilize both PET andEEG within the same psychological paradigms.

There is a pressing demand for psychological tests ofcreativity that are compatible with these techniques in sev-eral parameters, such as time formats, presentation of stim-uli and control, as well as fundamental ideology.

All this makes the psychophysiology of creativity andcreative thinking extremely difficult to study.

3. Psychological aspects of methodology and psychological

tests used

It is noteworthy that psychological approaches varywidely in different laboratories, which makes the investiga-tions hard if not impossible to compare.

Psychological tests (tasks) for our studies were designedon certain premises.

There was enough psychological evidence to assume thatfast cognitive leaps or intuitive insight are essentials for cre-ative thinking [24]. But we do not share the view that onlythe above mentioned transitory brain processes conditionfor a creative thinking as different from a non-creativeone. A search for brain correlates of creative thinking com-patible with temporal resolution of state-oriented physio-logical measurements (PET and state-oriented EEG)seemed justified to us, too. Proper psychological tests hereshould provide more or less uniform sustained creativeprocesses in subjects in time intervals long enough for suf-ficient data acquisition.

Psychophysiological investigations in humans aremostly based on comparisons (contrasts) between situa-tions and a within-subject comparison design in the groupunder study. Comparisons between good performers andpoor performers are possible and used. The method, how-ever, is less sensitive (group comparisons) and demands asubstantially larger number of participants because of thehigh inter-individual scatter in data. In initial psychophys-iological studies of creative thinking it is reasonable to staywithin the frames of the within-subjects design and dividetasks into more creative, less creative and non-creative.

We cannot precisely estimate all cognitive and emo-tional factors influencing the human brain during creativeactivities. Here along with multifactor contrasts in a certainexperiment, one should be able to compare the resultsobtained from the tasks which belong to the same type,e.g. verbal creative tasks, but are somewhat different(‘‘story composition’’, ‘‘associative chains’’, ‘‘original defi-

of

the

test

sets

reat

ive

task

sC

on

tro

lta

sks

Su

bsi

dia

ryac

tivi

tyin

du

ced

Ph

ysio

logi

cal

met

ho

ds

app

lied

and

the

nu

mb

ero

fsu

bje

cts

.S

tory

com

po

siti

on

usi

ng

wo

rds

fro

mre

mo

tese

man

tic

area

s1.

Res

titu

tio

no

fa

text

fro

ma

set

of

wo

rds

wit

ho

ut

chan

gin

gth

eir

ord

er—

PE

T—

16;

EE

G—

26

.S

tory

com

po

siti

on

usi

ng

wo

rds

fro

mn

ear

sem

anti

car

eas

2.R

emem

ber

ing

and

sub

seq

uen

tre

pro

du

ctio

no

fsa

mp

les

of

wo

rds

.F

orm

atio

no

fa

logi

cal

wo

rdch

ain

bet

wee

ntw

op

rese

nte

dw

ord

sfr

om

rem

ote

sem

anti

car

eas

1.A

ttac

hm

ent

of

5m

ore

wo

rds

fro

mth

esa

me

sem

anti

cca

tego

ryto

ap

rese

nte

dw

ord

—P

ET

—9;

EE

G—

35

2.R

ead

ing

alo

ud

.O

rigi

nal

defi

nit

ion

toa

no

un

,u

sin

ga

sup

ple

men

tw

ord

fro

ma

rem

ote

sem

anti

car

ea

1.D

efin

itio

no

fa

no

un

usi

ng

asu

pp

lem

enta

ryw

ord

fro

mth

esa

me

sem

anti

cfi

eld

1.In

du

ctio

no

fp

osi

tive

emo

tio

ns

usi

ng

emo

tio

nal

lyva

len

ced

wo

rds

and

the

inst

ruct

ion

tosu

stai

nth

ein

tern

alem

oti

on

alst

ate

corr

esp

on

din

gto

the

wo

rd2.

Ind

uct

ion

of

neg

ativ

eem

oti

on

sin

the

sam

ew

ay

PE

T—

16;

EE

G—

30

.R

adic

alch

ange

of

ap

rove

rbse

nse

via

sub

stit

uti

on

of

an

eww

ord

inst

ead

of

am

issi

ng

on

e

1.In

sert

ion

of

am

issi

ng

wo

rdin

ap

rove

rb1.

Dir

ecte

dd

etec

tio

no

fer

rors

inth

ete

xts

pre

sen

ted

1.In

cid

enta

ld

etec

tio

no

fer

rors

inth

ete

xts

pre

sen

ted

1.E

xtra

atte

nti

on

acti

vate

db

yco

un

tin

gas

teri

sks

add

ed1.

Dir

ecte

dd

etec

tio

no

fer

rors

inth

ete

xts

pre

sen

ted

2.E

xtra

atte

nti

on

acti

vate

db

yco

un

tin

gas

teri

sks

add

ed

EE

G—

22E

EG

—25

EE

G—

20E

EG

—37

102 N.P. Bechtereva et al. / Methods 42 (2007) 100–108

nitions’’, ‘‘proverb sense flipping’’). It might be reasonablealso to design tests that anticipate ways of introducingcontrolled loads on emotions, attention and error detec-tion as substantial factors in creative processes.

We performed our studies with healthy volunteers,aged 17–28. Only males participated in the PET studies,while both males and females were involved in the EEGstudies.

Since a number of tests and experimental paradigmswere used, an overview of the test sets is presented inTable 1. All the texts were shown to subjects on a com-puter display at a distance of 1 m from their eyes.

3.1. Test set A

In each of the tasks, a word set consisting of 16 words(eight infinitives and eight nouns in the singular nomina-tive case) organized as a matrix (four lines were arrangedin four columns) were presented to a volunteer for 90 s.The letters were black against the white background. Allthe words consisted of two or three syllables and occurredonly in one task. The first creative task was to compose astory using a word list with the words from different andremote semantic areas, such as: ‘to begin, to want, glass,roof, mountain, to keep silence, book, to leave, sea, night,to open, cow, to throw, to notice, to disappear, mush-room’. Composing such a story was a rather difficult task(further marked D—‘Difficult’). In creative Task 2 (Con-dition E—‘Easy’), a subject was to compose a story usingthe word list with the words from one semantic area suchas: ‘school, to understand, task, to learn, lesson, answer,to get, to write, mark, to ask, class, to answer, question,to solve, to listen, teacher.’ The D Condition was sup-posed to involve more creativity than the E Condition.

Moreover, to control the syntactic and memory-relatedaspects of the tasks, two more control tasks were added.In control task 1 (marked R—Reading) the subject wasto compose a sentence by merely changing the wordforms1 and adding auxiliaries for the following set: ‘man-ager, to suggest, to call, boss, department, to demand, tocompose, program, to solve, problem, to increase, pro-ductivity, to try, to reach, level, plan’. In the control task2 (marked W—Words), the subject was to memorize thepresented list of words, such as: ‘air, to see, meeting, coun-try, to do, to remember, wave, airplane, to arise, to stand,to look, substance, to enter, word, to say, accident’ andreproduce them thereafter.

The subjects participating in this experiment had toundergo a preliminary psychological screening and onlythose took part who used successive strategy (composedthe whole story in mind before reproducing it).

The results obtained with this test set are illustrated inFig. 1.

Tab

le1

An

ove

rvie

w

Tes

tse

tC

A1 2

B1

C1

D1

1 Russian is an inflexional language.

Fig. 1. Statistically significant rCBF differences obtained with the test set A. D—composition of a story using words from remote semantic fields; E—composition of a story using words from the same semantic field; R—restitution of a text changing word forms; W—memorizing a word set. Furtherphysiological analyses of rCBF distributions in the regions of interest suggest that Brodmann Area (BA) 39 is most intimately involved in creativethinking. EEG differences here did not appear to be statistically significant. More detailed information see [32,13,15].


3.2. Test set B

This test was applied to the second (‘‘insight’’) group ofvolunteers. In each task a set consisting of 12 nouns (singu-lar in the nominative case) organized as matrix (four lineswere arranged in three columns) were presented to a volun-teer for 90 s. The letters were also black against the whitebackground. All the words consisted of 1–4 syllables andoccurred only in one task. In the creative task D (‘Diffi-cult’) a subject was expected to fill in the gaps betweenthe words from different semantic areas offered to him/her. The subject was to say aloud (in PET) or to oneself(in EEG) results of associative word chain formation.For instance, in case of the chain: ‘glass, river, magnet,time, parquet, book, snow, wax, ceiling, package, wood-pecker, armchair’, the subject was to transform them into:‘glass-reflection-water-river, river-sand-element-electron-magnet, magnet-attraction-movement-time’, etc. In thefirst control task E (‘Easy’), the subject was to list five itemsaloud (in PET) or to oneself (in EEG) for each presentedword: ‘clothes, tree, utensils, transport, structure, footwear,bird, science, vegetable, metal, dance, flower’. In the secondcontrol task R (‘Reading’), the subject was to read aloud(in PET) a chain of words, such as: ‘city, case, car, word,window, hooter, slope, group, groats, sofa, key, tin’. Thecontrol task 2 was not used in the EEG experiments.

The results obtained from this test set are illustrated inFig. 2.

3.3. Test set C

In the creative task, the subjects were to create theirown, personal, and original definitions (as many as possi-ble) to the presented nouns. The nouns serve either emo-tionally positive, negative or neutral ones. Thesupplementary neutral word was taken from a differentremote semantic area. The neutral word pairs were:‘‘essence (to be defined), sand (to be used)’’, ‘‘pedestal,foil’’. The emotionally positive were ‘‘love, groats’’; ‘‘kiss,cap’’. The emotionally negative were ‘‘death, paste’’,‘‘social outcast, ore’’. Some examples of definitions pro-posed by the subjects were as follows: ‘‘Essence of aliens– are thoughts that run out of their heads as sand, whenthey stand on their heads in order to think.’’; ‘‘Love isthe state when the lifeless groats can turn back into cornand grow into a wonderful plant, changing the laws ofnature’’.

In the control task the subject was to produce severaldefinitions to a noun using a supplementary word fromthe same semantic area: ‘‘chips, firewood’’ or ‘‘clothespress,linen’’. There were no demands on the originality of thedefinitions in this task.

Fig. 2. Statistically significant rCBF differences obtained with the test set B. D—formation of associative word chains words from remote semantic fields;E—to add 5 words (items) from the same semantic field to a word presented; R—reading aloud words out of the text presented. Further physiologicalanalyses of rCBF distributions in the regions of interest suggest that BA40 is most intimately involved in this kind of creative thinking (insight strategy?).EEG differences in the contrasts did not appear to be statistically significant. More detailed information see [11,14,15].

O-K

47L

P-K

47,11L

N-K

17, 18, 19R9,10,11,47L

EEG power EEG coherence rCBF

O-K

P-K

N-K

Fig. 3. Statistically significant differences of local EEG power and coherence of b2 range (left) and rCBF (right) between the creative task and the controlone obtained with the test set C. O—a search for original definitions to concepts using in definitions words from remote semantic fields in fixed timeinterval (the basic task); K—a search for definitions (non original) to concepts using in definitions words from the same or nearby semantic field (controltask); P—task similar to the ‘‘O’’ but in condition of positive emotions induction; N—type of task similar to ‘‘O’’ but in a condition of negative emotionsinduction. Arrows pointing up show higher power of the EEG in the first task, in given brain area. Solid lines connecting brain areas show a higher level ofcoherence in the first task. The results illustrate masking effects of emotion induction on EEG correlates of creative thinking (the contrasts P–K and N–Kcompared to the contrast O–K and rCBF data). They also illustrate that rCBF and EEG data are complementary. Diverging rCBF/EEG data allow, incertain situations, to separate effects of augmentation in synaptic acts density (rCBF data) and in synaptic acts temporal synchronization/desynchronization (EEG data). The results were previously published in parts [22,33,34].



The word pairs were presented for 60 s each. The sub-jects were requested to sustain corresponding emotionalfeeling during the trial if they considered the defined nounas emotionally arousing.

The results obtained with this test set are illustrated inFig. 3.

3.4. Test set D

Both in the creative and control tasks well-known Rus-sian proverbs were presented to subjects with the last wordmissing. In the creative task the subjects were to createtheir own, original ending to the proverb with a word,changing their well-known meaning for a quite differentone. In the control task (memory retrieval task) the sub-jects were to remember commonly used ending of theproverb.

Each task (creative or control) consisted of 21 trials andlasted nearly 3 min. One trial lasted for 2800 ms, with thetext presentations starting 300 ms after the trial started.All the trials were separated by pause intervals (from2500 to 4500 ms). The subjects were to give their repliesat the onset of a pause.

Some subsidiary simultaneous activities were tested inthe experiments too. Errors in the proverb texts were intro-duced in one of the variants to activate the error detectionmechanisms. The errors were in the form of rearrange-ments of the order of the letters in the words or letter

Directeddetection 1

Incidentaldetection

CRD-CR

CR-K

CRD-K

KD-K

Fig. 4. Statistically significant differences of EEG power in c-range obtained wwith error detection in the text; CRA—the creative task with counting additionerror detection in the texts; KA—the control task with counting additional sywere applied in the frames of one session (presented randomly) in the same groreproducibility of the main effect (contrast Cr-K) for different groups of subjedetection and extra attention activation. The results were partly published [35

replacements without changing the word length. Theinstructions varied. The subjects were either instructed tocount the number of errors along with main tasks (theexperiment with directed or ‘‘conscious’’ detection) or theydid not get any special instruction about errors (an exper-iment with incidental or ‘‘unconscious’’ detection). Thus,the complete set of tasks in one experiment (4 tasks) of thistype consisted of 84 trials.

To compare the effects of the error detection and theeffects of extra attention activation, another modificationof the tests was used. In the text field presented, asteriskswere included (and no errors) with the instruction to countthem in case they were detected. The complete set of tasksin an experiment with such comparison (6 tasks) consistedof 126 trials. The sequence of tasks was randomized for allthe subjects.

The results obtained from this test set are illustrated inFig. 4.

4. PET in creativity studies: Some methodological aspects

and methods used

Tomographic methods of local brain hemodynamicscontrol (PET and fMRI) are widely accepted nowadaysin psychophysiological studies of cognition and emotions.They provide unprecedented non-invasive access to pro-cesses in the structures of the living brain with a rather high

CRA-CR

CR-K

CRA-K

KA-K

Directeddetection 2

Extra attention

ith the test set D between: CR—the creative task; CRD—the creative taskal symbols on the screen; K—the control task; KD—the control task withmbols on the screen. Tests ‘‘Directed detection 2’’ and ‘‘Extra attention’’up of volunteers. Symbols are the same as in Fig. 3. The results illustrate

cts. They also illustrate difference in intensities of effects of directed error,36].


spatial accuracy, which makes their use in psychophysio-logical studies of creativity highly desirable.

PET and fMRI methods are standardized in their appli-cation and no particular methodological comments areneeded here except, perhaps, for one remark.

Both methods – fMRI and PET—are quite differentbut they provide similar information about changes oflocal brain hemodynamics. These changes reflect theinvolvement of brain regions in the activities under study.Though fMRI methods are less expensive and simpler,PET seems more adequate for studying brain correlatesof creativity. The reason for this is that creative processesare rather delicate activities, which makes identifying theirneural correlates hard to detect when other strongly pro-nounced activations are also simultaneously present. ThefMRI procedures are inevitably noisy and the subjecthas to be placed in a closed, uncomfortable space. There-fore, PET seems more adequate here and gives hope ofmore reliable detection of the brain correlates ofcreativity.

In our experiments, the subjects lay supine on the coachof the PET camera with their heads fixed in the headholder. The right forearm was positioned in a way whichenabled the injection of 15–0-water through a catheterinserted into the right cubital vein. Prior to each 60 s scan-ning, the instruction about the task performance wasrepeated to the subject. The presentation of the test textsbegan prior to the injection of 15–0 water and ended withthe end of the scanning.

All PET scans were performed on ScanditronixPC2048-15B camera (15 parallel slices with an in-planespatial resolution of 6.5 mm FWHM in the center ofthe FOW and an interslice distance of 6.5 mm; for moretechnical details see [25,26]). rCBF was measured by using15–0-labeled water by autoradiography—following anintravenous bolus injection of 50–60 mCi of H2

15O. A60-s PET scan was acquired 15–20 s after the bolus injec-tion. The data were analyzed with the SPM 99 software[27]. Following the realignment of images from each sub-ject in order to correct any changes in the head positionbetween scans, the images were transformed into standardanatomical space used in SPM 99. In order to increase thesignal-to-noise ratio and to accommodate normal vari-ability in functional and gyral anatomy the images weresmoothed with a gaussian filter 16 · 16 · 16 mm wide[28]. The resulting activity data were normalized for differ-ences in global flow by scaling voxel by voxel to a globalmean of 50 ml/dl/min [29].

Task-specific activations were assessed by statisticalcomparisons between conditions using t-statistics. For thedetected activation to be statistically significant, the impor-tance threshold for the resulting statistical parametric maps(SPMt) was set at P < 0.05 in voxel based analysis. Ana-tomical identification of activations was made on the basisof the Talairach and Tournoux brain atlas [30] using MMspace utility and Talairach space utility (http://www.ihb.spb.ru/�pet_lab).

As far as we know, PET has not been used often forcreativity research. Methods of brain mapping used herecould be compared with those used by others [16]. Theresults obtained by way of fMRI can be consideredcomparable to ours, especially since, in this rare case,the psychological tests used were similar. Both tech-niques offered the opportunity to record not only gen-eral but also local rearrangements of brain function[see also 31].

5. EEG methodology and methods used in our creativity

studies

Along with PET, we consider, at the moment, electro-encephalography, the most helpful and affordable instru-ment in multimodal psychophysiological studies ofcreativity.

Quantitative EEG confirmed gamma frequency EEG tobe a sensitive indicator of brain activation in various situ-ations of brain-mind interactions. EEG analysis in a total-ity of frequency bands has become a source of reliableinformation.

EEG is particularly useful in order to obtain more spe-cific information that is hard or impossible to acquire fromhemodynamics data. Thus,EEG is a complementary tool toPET and fMRI. It is not only the relatively high temporalresolution that provides such complementary informationin an event-related EEG paradigm. When using EEGparameters averaged on time intervals similar in durationto PET scan intervals, one also obtains important comple-mentary information about brain states and activitiespresumably related to the functioning of the brain’sself-regulating mechanisms [34,35].

When discussing brain functional states (FS) that arespecific to certain kinds of cognitive processes, we referto quasi stationary components of brain maintenancethat are relatively stable in time. It is most likely thatthere are close links between certain FS and spatial pat-terns of quasi-stationary components of EEG dynamics(revealed via time averaging of oscillatory EEG parame-ters). Thus, one of the ways to substantiate a certain FSis to obtain evidence of significant and reproducible dif-ferences of time averaged EEG parameters in compari-sons with a reference FS. Oscillatory EEG can be usedfor this purpose which can be denoted as a ‘‘state-relatedEEG’’ and thus be distinguished from event-relatedEEG.

It is noteworthy that by applying PET we assume thatthere are quasistationary components in local brainhemodynamics during a scan (30–60 s.). Thus, a searchin state-related EEG for complementary information onbrain correlates of creative thinking is a reasonableundertaking.

State-related EEG in creativity studies can also be usedas a tool for evaluating the adequacy of our experimentalparadigms as well as our theoretical ideas regarding thepsychological structure of the subject’s mental activities.

http://www.ihb.spb.ru/~pet_lab




Since the EEG studies are neither very expensive nor timeconsuming, large numbers of subjects can be involved.

For EEG recordings, 19 active electrodes (the full inter-national 10–20 system) and coupled earlobe reference elec-trodes were used. Our EEG data was recordedmonopolarly in the bandwidth of 0–70 Hz with the use ofa Mitsar EEG computerized electroencephalograph (NPFMitsar, St. Petersburg, Russia) that has a sampling rateof 250 Hz. For the data processing, a rejector filter tunedto a frequency of 50 Hz with a suppression band of0.1 Hz was used.

Fragments contaminated by artifacts were excludedfrom further processing. Quantitative EEG characteristicswere calculated by means of the commercial WinEEG soft-ware package, version 3.1 (NPF Mitsar, St. Petersburg,Russia). Mean estimations of the EEG absolute spectralpower for each derivation were calculated for each subjectin the given state. The estimations were made for the spec-tral components averaged over the bands delta (1.5–4 Hz),theta (4–7 Hz), alpha1 (7–10 Hz), alpha2 (10–13 Hz), beta1(13–18 Hz), beta2 (18–30 Hz), and gamma (30–40 Hz).Arrays of the estimations of the EEG absolute power werenormalized using Y = logX transformation. Mean values ofthe coherence function were also calculated for each of thebands. Arrays of the estimations of the coherence functionwere normalized with the formula Y = log(X2/(1�X2)).

The statistical analysis of the arrays was aimed atrevealing significant differences between the EEG parame-ters in the compared states The significance of differencesbetween the state mean values was determined using awithin-subject design. Plans D · S · Z were used toanalyze the EEG power spectrum parameters, where Dis the factor of frequency band, S is the factor of state,and Z is a topographical factor (derivations are gradationsof this factor). Because of the large number of variables,plans S x Z were used to analyze the EEG coherenceindividually for each frequency band in which derivationpairs were used as gradations of the topographical factorZ.

In determining the significance of the main effects ofthe factors and their interactions, the Greenhouse–Geis-ser correction was used, followed by post-hoc FisherLSD test.

The electrophysiological approach is comparable to theone used in other laboratories. However, they show varia-tion from the so-called ‘‘old ones’’.

6. Conclusions

The experience obtained from using the methodsdescribed in this article confirms that we are on theright track in procuring some knowledge about thebrain mechanisms of creative thinking. A multi-facetedpolymodal approach to measuring physiological indicesduring adequate psychological test performance providethe basis for such attempts. We know more now aboutthe zones in the brain most likely involved in creative

components of verbal brain activities. We also havenow some reference data to evaluate the contributionsof emotion, error detection and attention in the brainactivities.

References

[1] C. Martindale, D. Hines, Biol. Psychol. 3 (1975) 91–100.[2] C. Martindale, N. Hasenfus, Biol. Psychol. 6 (1978) 157–167.[3] . Martindale, in: R.J. Sternberg (Ed.), Handbook of Creativity,

Cambridge University Press, 1999, pp. 137–153.[4] H. Petsche, Int. J. Psychophysiol. 24 (1996) 145–159.[5] H. Petsche, S. Kaplan, A. von Stein, O. Filz, Int. J. Psychophysiol. 26

(1997) 77–97.[6] M. Molle, L. Marshall, W. Lutzenberger, R. Pietrowsky, H.L. Fehm,

J. Born, Neurosci. Lett. 12 (1996) 61–64.[7] M. Molle, L. Marshall, B. Wolf, H.L. Fehm, J. Born, Psychophys-

iology 36 (1999) 95–104.[8] O.M. Razoumnikova, Cogn. Brain Res. 10 (2000) 11–18.[9] O.M. Razumnikova, Neurosci. Lett. 362 (2004) 193–195.

[10] N. Jausovec, K. Jausovec, Brain Topogr. 12 (2000) 229–240.[11] S.G. Danko, M.G. Starchenko, N.P. Bechtereva, Hum. Physiol. 29

(2003) 502–504.[12] I. Carlsson, P. Wendt, J. Risberg, Neuropsychologia 38 (2000) 873–

885.[13] N.P. Bechtereva, M.G. Starchenko, V.A. Klyucharev, V.A.

Vorob’ev, S.V. Pakhomov, S.V. Medvedev, Hum. Physiol. 26(2000) 516–522.

[14] N.P. Bechtereva, S.G. Danko, M.G. Starchenko, S.V. Pakhomov,S.V. Medvedev, Hum. Physiol. 27 (2001) 390–397.

[15] N.P. Bechtereva, A.D. Korotkov, S.V. Pakhomov, M.S. Roudas,M.G. Starchenko, S.V. Medvedev, Int. J. Psychophysiol. 53 (2004)11–20.

[16] S.V. Medvedev, N.P. Bechtereva, S.G. Danko, M.G. Starchenko,N.V. Shemyakina, S.V. Pakhomov, Int. J. Psychophysiol. 54 (2004)34.

[17] P.A. Howard-Jones, S.J. Blakemore, E.A. Samuel, I.R. Summer, G.Claxton, Brain Res. Cogn. Brain Res. 25 (2005) 240–250.

[18] C.W. Taylor, in: R.J. Sternberg (Ed.), The Nature of Creativity:Contemporary Psychological Perspectives, Cambridge UniversityPress, 1988, pp. 99–121.

[19] M. Rhodes, Phi Delta Kappan 42 (1961) 305–311.[20] B. Horwitz, D. Poeppel, Hum. Brain Mapp. 17 (2002) 1–3.[21] P.L. Nunez, R.B. Silberstein, Brain Topogr. 13 (2000) 79–96.[22] S. Medvedev, N. Bectereva, S. Danko, Int. J. Psychophys. 61 (2006)

323.[23] S. Danko, N. Bectereva, N. Shemyakina, S. Medvedev, S. Pahomov,

Int. J. Psychophys. 61 (2006) 324.[24] A. Koestler, The Act of Creation, Macmillan, London, 1964.[25] S. Holte, L. Eriksson, M. Dahlbom, Eur. J. Nucl. Med. 15 (1989)

719–721.[26] A. Evans, C. Thompson, S. Marrett, E. Meyer, M. Mazza, IEEE

Trans. Med. Imaging 10 (1991) 90–98.[27] K.J. Friston, A.P. Holmes, K.J. Worsley, J.P. Poline, C.D. Frith,

R.S.J. Frackowiak, Hum. Brain Mapp. 2 (1995).[28] K.J. Worsley, S. Marett, P. Neelin, A.C. Vandal, K.J. Friston, A.C.

Evans, Hum. Brain Mapp. 4 (1996).[29] A.R. Mclntosh, C.L. Grady, J.V. Haxby, J.M. Maisog, . Horwitz,

C.M. Dark, Hum. Brain Mapp. 4 (1996).[30] J. Talairach, P. Tournoux, Co-planar stereotaxic atlas of the human

brain, Georg Thieme Verlag, Thieme Medical Publishers, Stuttgart,New York, 1988.

[31] M. Jung-Beeman, E.M. Bowden, J. Haberman, J.L. Frymiare, S.Arambel-Liu, R. Greenblatt, J.P. Reber, J. Kounios, Public LibrarySci. Biol. 2 (2004) E97.

[32] M.G. Starchenko, V.A. Vorobiev, V.A. Klyucharev, N.P. Bechtereva,S.V. Medvedev, Hum. Physiol. 26 (2000) 125–128.


[33] N.V. Shemyakina, S.G. Danko, Hum. Physiol. 30 (2004) 22–29.

[34] N.V. Shemyakina , in: F. Jagla, I. Riechansky (Eds.), Interna-tional C.I.A.N.S Conference 2005, Book of Abstracts, WORD&-IMAGE Publishing & Translation Agency, Bratislava, Slovakia,2005, p. 72.

[35] N.P. Bechtereva, N.V. Shemyakina, M.G. Starchenko, S.G. Danko,S.V. Medvedev, Int. J. Psychophysiol. 58 (2005) 227–234.

[36] N.P. Bechtereva, S.G. Danko, M.G. Starchenko, N.V. Shemyakina,in: F. Jagla, I. Riechansky, (Eds.), International C.I.A.N.S Confer-ence 2005, Book of Abstracts, WORD&IMAGE Publishing &Translation Agency, Bratislava, Slovakia, 2005, p. 22.

current methodology and methods in psychophysiological studies of creative thinking

Documents