the occurrence of autistic children's self-stimulation as a function of familiar versus...

14
Journal of Autism and DevelopmentalDisorders, Vol. 16, No. 1, 1986 The Occurrence of Autistic Children's Self- Stimulation as a Function of Familiar versus Unfamiliar Stimulus Conditions I Mark A. Runco 2 University of Hawaii at Hilo Marjorie H. Charlop Claremont McKenna College and Clarernont Graduate School Laura Sehreibman University of California, San Diego The present study was conducted to determine whether certain stimulus con- ditions were associated with high and low rates of autistic children's self- stimulation. Six autistic boys were assessed in situations varying along three dimensions: famih'arity or unfamiliarity of setting, learning task, and therapist. Each child was observed in 10 lO-min stimulus conditions, and trained observers recorded the occurrence of self-stimulation within each con- dition. The results of a 2 • 2 x 2 ANOVA indicated that self-stimulation occurred significantly more often with an unfamiliar than with a familiar therapist. Unfamiliar versus familiar setting and task were not significant effects, and there were no significant interactions. Also, significant differences were found within each condition, with self-stimulation increasing in fre- quency as the sessions progressed. Finally, there was a significant and negative correlation between the occurrence of self-stimulation and correct respon- ding. These findings suggest several treatment strategies for facilitating a generalized suppression of autistic children's self-stimulation. ~This research was supported by U.S. Public Health Service Research Grants MH 28231 and MH 28210 from the National Institute of Mental Health. 2Address all correspondence to Mark A. Runco, Psychology, Social Sciences, College of Arts and Sciences, University of Hawaii, 1400 Kapiolani Street, Hilo, Hawaii 96720. 31 0162-3257/86/0300-0031505.G0/0 1986PlenumPublishing Corporation

Upload: mark-a-runco

Post on 10-Jul-2016

213 views

Category:

Documents


1 download

TRANSCRIPT

Journal of Autism and DevelopmentalDisorders, Vol. 16, No. 1, 1986

The Occurrence of Autistic Children's Self-

Stimulation as a Function of Familiar versus

Unfamil iar Stimulus Condit ions I

Mark A. Runco 2 University o f Hawaii at Hilo

Marjorie H. Charlop Claremont McKenna College and Clarernont Graduate School

Laura Sehreibman University o f California, San Diego

The present study was conducted to determine whether certain stimulus con- ditions were associated with high and low rates o f autistic children's self- stimulation. Six autistic boys were assessed in situations varying along three dimensions: famih'arity or unfamiliarity o f setting, learning task, and therapist. Each child was observed in 10 lO-min stimulus conditions, and trained observers recorded the occurrence of self-stimulation within each con- dition. The results of a 2 • 2 x 2 ANOVA indicated that self-stimulation occurred significantly more often with an unfamiliar than with a familiar therapist. Unfamiliar versus familiar setting and task were not significant effects, and there were no significant interactions. Also, significant differences were found within each condition, with self-stimulation increasing in fre- quency as the sessions progressed. Finally, there was a significant and negative correlation between the occurrence of self-stimulation and correct respon- ding. These findings suggest several treatment strategies for facilitating a generalized suppression of autistic children's self-stimulation.

~This research was supported by U.S. Public Health Service Research Grants MH 28231 and MH 28210 from the National Institute of Mental Health.

2Address all correspondence to Mark A. Runco, Psychology, Social Sciences, College of Arts and Sciences, University of Hawaii, 1400 Kapiolani Street, Hilo, Hawaii 96720.

31

0162-3257/86/0300-0031505.G0/0 �9 1986 Plenum Publishing Corporation

32 Runco, Charlop, and Schreibman

Self-stimulation is a behavior commonly displayed by autistic children (e.g., Koegel, Egel, & Dunlap, 1980; Lovaas, Koegel, Simmons, & Long, 1973; Ritvo & Freeman, 1978; Ritvo, Ornitz, & LeFranchi, 1968; Rutter, 1978). Self-stimulatory behaviors are usually defined as repetitive, stereotyped, idiosyncratic motor behaviors that typically function only to provide sen- sory feedback (e.g., Lovaas, Litrownik, & Mann, 1971). Examples include arm and hand flapping, body posturing, squinting or gazing from the cor- ners of the eyes, and quick, darting movements (e.g., Schreibman & Mills, 1983). Other self-stimulatory behaviors involve object manipulation such as spinning, tapping, and feeling the edges of a coin, pencil, or small toy (Lovaas et al., 1973; Schreibman & Mills, 1983). Self-stimulation is quite problematic in that it interferes with appropriate responding and has been implicated in the failure of acquisition and generalization of new behaviors (Risley, 1968). It appears to be inversely related to appropriate responding (Koegel & Covert, 1972) and responsivity to previously controlling stimuli (Lovaas et al., 1971).

Self-stimulation has been viewed as an operant behavior maintained by sensory feedback (Rincover, 1978; Rincover, Newsom, Lovaas, & Koegel, 1977), and sensory extinction procedures have been shown to produce decreases in the behavior (Rincover, 1978). Unfortunately, sensory extinc- tion procedures are not always easy to design or implement. As a result, behavior therapists often resort to procedures that focus on the external con- sequences of self-stimulation. Examples of these procedures include differen- tial reinforcement of incompatible behaviors (Repp, Deitz, & Spier, 1974), punishment (Koegel, Firestone, Kramme, & Dunlap, 1974; Lovaas, Schaef- fer, & Simmons, 1965), and overcorrection (Azrin, Kaplan, & Foxx, 1973). These procedures are, however, often time-consuming and tedious to ad- minister (e.g., overcorrection or DRO) or pose ethical problems (e.g., cer- tain forms of punishment).

In an attempt to develop expedient and less restrictive means of reduc- ing self-stimulation, the study of antecedent variables may provide informa- tion relevant to the control of this behavior. Previous research has demonstrated the relationship between occurrence of specific behaviors and certain stimulus conditions (Cataldo, Bessman, Parker, Pearson, & Rogers, 1979; Charlop, 1981; Charlop, Schreibman, Mason, & Vesey, 1983) and the effectiveness of antecedent setting control (Whalen, Henker, Collins, Finck, & Dotemoto, 1979). Additionally, the literature on the generalization of treat- ment effects suggests that although gains may be made in the therapy en- vironment (i.e., suppression of self-stimulation), these gains may not generalize to extratherapy situations (Baer, Wolf, & Risley, 1968; Lovaas et al., 1973; Stokes & Baer, 1977; Wahler, 1969; Walker & Buckley, 1972). It may be important to determine whether certain stimulus situations (i.e., unfamiliar environments) set the occasion for an increase in self-stimulation,

Autistic Self-Stimulation 33

making generalization even more difficult to achieve. Thus, the study of stimulus conditions may provide direct implications for the generalized sup- pression of self-stimulation.

The present study was desighed to determine whether certain stimulus conditions are associated with high and low rates of autistic self-stimulation. Specifically, children were assessed in situations varying along three dimen- sions: familiarity or unfamiliarity of setting, familiarity or unfamiliarity of learning task, and familiarity or unfamiliarity of therapist. These dimensions were chosen because much of autistic children's behavior has been demonstrated to vary as a function of these variables (i.e., Charlop, 1981; Koegel & Rincover, 1977; Lovaas et al., 1973; Rincover & Koegel, 1975).

METHOD

Subjects

Six boys, each diagnosed as austistic by at least two independent agen- cies, participated in this study. The children were diagnosed according to the National Society for Autistic Children's criteria for the syndrome of autism (Ritvo & Freeman, 1978). The mean chronological age of the children was 6 years 6 months, with a range of 3 years 8 months to 10 years 11 months. The mean Merrill-Palmer mental age was 3 years 6 months, with a range of 2 years 3 months to 5 years. The mean Vineland Social Maturity Scale age was 3 years 8 months, with a range of 2 years 11 months to 4 years 5 months. Five of the children were echolalic, and one (Child 1) was functionally mute. All of the children engaged in high rates of self-stimulatory behaviors. Table I presents the chronological, mental, and social ages, and the most frequent self-stimulatory behaviors for each child. These children regularly attended a university-affiliated behavior therapy clinic.

Settings

Two settings, a familiar and an unfamiliar room, were employed. The familiar setting was the experimental room in which each child received behavior therapy three times a week. This room was located in a university building and was approximately 2.9 m • 4.4 m. It was carpeted and con- tained a table, two chairs, a cabinet, and a wall with a one-way mirror.

The unfamiliar setting was one of four rooms in the same building. This room was either (a) a storeroom (12 m x 12 m), which was filled with shelves and electrical equipment but had no windows; (b) a student lounge (9 m x

34 Runco, Charlop, and Schreibman

Table I. Chronological, Mental, and Social Ages, and Representative Self-Stimulatory Behaviors

Child CA MA SA Self-stimulatory behavior

1 6-1 5-0 4-5

2 7-6 4-3 4-4

3 5-3 2-6 3-4

4 10-11 2-3 4-0

5 3-8 2-11 2-11 6 5-8 4-6 3-0

Toe walking, mouthing objects, body rocking, gazing askance

Waving hands, flapping wrists, facial grimacing, gazing askance, slapping or tapping back of hand

Body bouncing, jumping and rocking, tapping objects, digital exploration of toys

Body rocking and bouncing, flapping arms, mouthing ob- jects, gazing askance

Body posturing and rocking, panting, facial grimacing Tapping objects, rolling head, hand posturing, body rock-

ing, snapping fingers

8.5 m), which contained comfortable furniture, a refrigerator, and one wall of windows; (c) a small student office (3 m • 2.4 m) with a desk and bookshelves, but no window or carpeting; or (d) a large experimental room (6.5 m x 5.5 m) with carpeting, comfortable furniture, a counter and sink, and a one-way mirror. The children had never seen these four settings prior to the study. Furthermore, unfamiliarity was maintained throughout this pro- ject by using new settings for all consecutive conditions.

Tasks and Therapists

The children were presented with a familiar or unfamiliar task by one of several familiar or unfamiliar persons in the settings described above. The familiar task was a receptive labeling task that the child had previously mastered and demonstrated, on the average, 90% correct responding. Dur- ing these sessions, the child sat directly in front of the therapist and was ask- ed to touch various body parts. The therapist would sit quietly and wait for the child to display eye contact. Then, the therapist would ask the child to "Touch your nose," "Touch your head," and so on. Correct responding was defined as the child's accurate identification of a body part within 5 sec of the therapist's command. This behavior was reinforced with food (VR:5 schedule) and praise (CRF schedule). A verbal "No" was presented contingent upon incorrect responses or failure to respond. No consequences were pro- vided for self-stimulation throughout the entire investigation.

The unfamiliar task was a receptive labeling task for which the child had received no previous training. During sessions with this task, "Mother" and "Father" adult doll figures and "son" and "daughter" child doll figures

Autistic Self-Stimulation 35

were used as stimuli. In order to ensure unfamiliarity of the task, the labels of the stimuli were all given in Spanish. "Madre" and "padre" were the labels used for the mother and father dolls, and "hijo" and "hija" were the labels for the son and daughter dolls. These dolls were approximately 10 cm in length, made of rubber, and colored in a lifelike style. The children sat directly in front of the therapist, who held a small board in his/her lap. The therapist began the trial by placing the dolls on the lap board. The child was instructed first to attend to the therapist (by giving eye contact and sitting still) and then was instructed to give the therapist one or the other of the dolls. Cor- rect responses were reinforced with food (VR:5 schedule) and praise (CRF), while incorrect responses were followed by a verbal "No." Again, no con- tingencies were administered for any self-stimulatory behaviors.

Several graduate students who had conducted behavior therapy with a given child for approximately 1 year served as familiar therapists. The un- familiar therapists were adults who had never interacted with the child but who had been previously instructed on how to present each task. Training of the unfamiliar therapists consisted of an experimenter modeling the task procedures, and then role playing in which the unfamiliar therapist played both the therapist and the child. Thus, for the purposes of this study, the unfamiliar therapists' behavior sufficiently resembled that of the familiar therapists'. Each child worked with a different therapist in each of the obser- vations.

Each child was Observed in 10 10-min stimulus conditions. Each con- dition was presented three times for a total of 30 observations. Eight of the 10 conditions consisted of a combination of unfamiliar or familiar setting, task, and therapist. In order to compare the frequency of self-stimulation occurring in these conditions with unstructured settings, two additional con- ditions consisted of the child alone with toys, in either the familiar setting or the unfamiliar setting. The 10 conditions are listed in Table II. The order of presentation of the conditions was controlled so that the same condition was not presented to a child consecutively, and so that no two children were presented with the conditions in the same order. A 10-rain time period was chosen to closely approximate the amount of time typically devoted to one task during work sessions in the clinic and school. At least one 10-rain break occurred between conditions, and no more than three conditions were con- ducted in one day.

During the unfamiliar task conditions, the "fa ther /mother" doll set or the "son/daughter" doll set was randomly chosen and then held constant for particular stimulus conditions. For example, Child 1 was presented with "mother / fa ther" dolls during all conditions in which unfamiliar tasks were presented by an unfamiliar therapist (stimulus conditions 7 and 8), while the "son/daughter" dolls were used during all familiar therapist conditions

36 Runco, Charlop, and Schreibman

Table II. Mean Percent of Occurrence of Self-Stimulation for All Children Within the First, Second, ~tnd Third Segments of Each Condit ion

Condit ion 1st Segment 2nd Segment 3rd Segment

1. Familiar therapist

Familiar room Familiar task

2. Familiar therapist Unfamiliar room Familiar task

3. Familiar therapist Familiar room Unfamiliar task

4. Familiar therapist Unfamil iar room Unfamiliar task

5. Unfamiliar therapist Familiar room Familiar task

6. Unfamiliar therapist Unfamiliar room Familiar task

7.

8.

Unfamil iar therapist Familiar room Unfamiliar task

Unfamil iar therapist Unfamil iar room Unfamiliar task

16.3 20.0 27.6

28.6 31.4 29.5

29.2 31.5 34.6

28.6 30.4 36.0

45.6 51.7 56.1

43.2 56.3 70.0

51.3 50.1 58.5

55.0 58.6 66.1

9. Alone w / t o y s 62.5 63.1 75.6 Familiar room

10. Alone w / t o y s 75.0 77.7 83.1 Unfamiliar room

= 43.5 47.1 53.7

Autistic Self-Stimulation 37

(stimulus conditions 3 and 4). Child 2, however, was presented with the "son/daughter" dolls by an unfamiliar therapist while the familiar therapist presented only the "mother/father" dolls.

Data Recording

Two observers were trained to score self-stimulation with a continuous 10-second partial interval recording procedure (Powell, Martindale, & Kulp, 1975). The observers practiced scoring these behaviors with videotapes of other autistic children until they agreed on more than 80% occurrences and nonoccurrences. Then they independently scored the experimental sessions. Reliability was calculated by dividing the number of intervals in which the observers agreed by the total number of intervals. When the setting allowed, they observed from behind the one-way mirror, and otherwise they observ- ed from the most unobtrusive position possible (usually behind and to one side of the child). The same observers were used throughout the experiment for every child, and hence the familiarity of the observer was constant across children and conditions. The behaviors scored for each child are presented in Table I. These particular self-stimulatory behaviors were chosen because parents, teachers, and therapists reported that they occurred most frequent- ly. Interrater reliability for the experimental sessions exceeded 74% for the occurrence of self-stimulation and 76% for nonoccurrences. In addition, cor- rect responding on the receptive labeling tasks was scored (before conse- quences were given) in eight experimental conditions by additional observers. The data were then transformed into percentages by dividing the number of correct responses by the total number of discrete trials. Interrater reliability for correct responding exceeded 86% for the entire experiment.

RESULTS

Setting, Task, and Therapist Effects

In order to answer the primary question, that of the effect of familiar versus unfamiliar stimulus conditions upon the frequency of self-stimulatory behaviors, the first eight experimental conditions were analyzed with a 2 • 2 • 2 ANOVA for repeated measures. The results indicated that self- stimulation occurred significantly more often with an unfamiliar rather than a familiar therapist (F(1, 5) = 240.9, p < .001). Unfamiliar versus familiar setting and task did not produce significant effects, nor were there any signifi- cant interactions.

38 Runco, Charlop, and Schreibman

Additionally, to compare the occurrence of self-stimulation in unstruc- tured versus structured stimulus situations, the average of the two "alone" conditions was compared to the average of the four familiar therapist con- ditions and then to the average of the four unfamiliar therapist conditions, with t tests for dependent samples. "Therapist" was the only variable employed in this comparison because the other variables (i.e., setting and task) did not produce significant effects. The results indicated that the rate of self- stimulation was higher in the alone conditions than in the familiar therapist conditions and the unfamiliar therapist conditions, but only the former dif- ference was statistically significant (t(5) = 4.69, p < .01).

Figure 1 shows the mean percentage of self-stimulation for all the children in each of the conditions. Notice that for the aggregate sample, the mean percentage of self-stimulation was comparatively higher during task conditions in which an unfamiliar therapist presented the task. The most self- stimulation occurred, however, in the alone conditions.

Figure 2 presents the mean percentage of occurrence of self-stimulatory behaviors for the individual children, and shows that the individual children demonstrated the same trend as the aggregate data (Figure 1) in that the highest mean percentages of self-stimulation generally occurred in the alone conditions. Additionally, the data of the individual children illustrate the higher mean percentages of self-stimulation in the conditions in which the unfamiliar therapist was present (Conditions 5, 6, 7, and 8) than in the con- ditions in which the familiar therapist was present (Conditions 1, 2, 3, and 4).

Within-Condition Differences

Differences within the 10-min conditions were evaluated by compar- ing the rate of self-stimulation for each third (3 min, 20 sec) of each condi- tion. With the exception of Child 2, for the individual children as well as the aggregate, the mean percentage of self-stimulation increased as each ses- sion progressed. The mean percentage of self-stimulation across all condi- tions for all children was 43.5 during the first segment, 47.1 during the second, and 53.7 for the third. The difference between the first and second segments was not statistically significant, but the differences between the second and third segments and the first and third segments were (p < .01). These dif- ferences were also found looking specifically at the unfamiliar therapist con- ditions. The mean percentage of self-stimulation for each of the conditions can be seen in Table II.

Because the within-condition differences were significant, with the first third of each session having the lowest mean percentage of self-stimulation, an additional 2 x 2 • 2 repeated-measures ANOVA was conducted using just the frequency of self-stimulation in the first third of each segment. In

Autistic Self-Stimulation 39

I l l Z 0 _1

I -

uJ

I -

- I

I.I. Z D

I -

I11 "I" I -

_1

I.i,.

I �9 ]

I , I

I E �9

- I I

A ' k v 1

I I

S-IVAU3J_NI -40 39V.J_N3Ol::13d NV31AI

o ~ =

~ ' ~ ~ ~

~r ~ c~ ~ . . . , " ~

~ a

40 Runco, Charlop, and Schreibman

100~ Child 1 100 Child 2

q I I 60 6 0 ~

40 40

20 20

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

,oo 1 Ch.d3 lOO C.,,d.

ldllll 80 Z

6 0 60

40 40

~, 20 20

Z I,LI 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

0

100 -- Child 5 100_ Child 6

.~ 60 60

40 40

20 20

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

C O N D I T I O N S Fig. 2. Means for each child in each condition. The conditions were (1) familiar therapist, familiar room, familiar task; (2) familiar therapist, unfamiliar room, familiar task; (3) familiar therapist, familiar room, unfamiliar task; (4) familiar therapist, unfamiliar room, unfamiliar task; (5) unfamiliar therapist, familiar room, familiar task; (6) unfamiliar therapist, unfamiliar room, familiar task; (7) unfamiliar therapist, familiar room, un- familiar task; (8) unfamiliar therapist, unfamiliar room, unfamiliar task; (9) alone, familiar room, with toys; (10) alone, unfamiliar room, with toys.

add i t i on to ma in ta in ing the lowest mean percentage o f se l f - s t imula t ion , the first th i rd was chosen for fur ther analysis because dur ing this segment the chi ldren were p r o b a b l y r e spond ing more to the fami l i a r i ty or un fami l i a r i t y o f the s t imulus cond i t ions t han to the d iscovery tha t no consequences would be provided contingent upon self-st imulation. The results o f this analysis were near ly ident ica l to the previous A N O V A . Tha t is, the un fami l i a r versus fami l ia r therap is t i ndependen t var iab le was the on ly s ignif icant ma in ef-

Autistic Self-Stimulation

fect on the rate of self-stimulation (F(1, 5) = 43.5, p < were no interactions.

41

.001), and there

Correlation Between Self-Stimulation and Correct Responding

A Pearson product-moment correlation was computed for all children's rate of self-stimulation and correct responding on familiar and unfamiliar tasks. This correlation was negative and significant (r = -.35, p < .01, two- tailed). Spearman rank correlations between the children's rate of self- stimulation and correct responding on familiar tasks (r = -.47, p < .05, two-tailed) and unfamiliar tasks (r = -.39, p < .01, two-tailed) suggest that self-stimulation may have interferred more with correct responding on familiar than on unfamiliar tasks.

For the individual children, Spearman rank correlations between self- stimulation and correct responding on both tasks were negative except for Child 1 (r = .27), who made only a few correct responses (M = 13.6% cor- rect), and Child 5 (r = . 17), who made only a few incorrect responses (M = 80.4~ correct). These correlations were not statistically significant.

DISCUSSION

The results of this study suggest that the self-stimulatory behaviors of the autistic children were influenced by certain stimulus conditions. The most self-stimulation occurred during conditions in which the child was provided with the opportunity to play with toys, but in which no formal tasks were presented and the therapist was not present. During the other conditions, when either familiar or unfamiliar formal tasks were presented, more self- stimulation occurred when an unfamiliar therapist was present. The familiari- ty or unfamiliarity of the task and setting, in contrast, had a negligible ef- fect on the occurrence of self-stimulation.

One possible explanation for these findings is that the children in this study may have responded to those aspects of the environment that control contingencies (e.g., person working with the child), even though contingen- cies were not administered during experimental conditions. These children may have emitted relatively low rates of self-stimulation in the familiar therapist conditions because that therapist had administered consequences for these inappropriate behaviors during behavior therapy sessions before the start of the study. Thus, these therapists had previously established con- trol over these children's behavior. Recall also the significant within condition

42 Runco, Charlop, and Schreibman

differences, with self-stimulation increasing as each session progressed. This finding further supports the significance of previously established control. A similar effect is often found in parent-training procedures. Koegel, Schreib- man, Britten, Burke, and O'Neill (1982), for example, found that autistic children who received treatment only from their own parents maintained treat- ment gains for several months; however, these treatment gains did not generalize to other people. That is, the treatment gains were not evident when the children were with anyone but their own parents. The children respond- ed as if they discriminated who was likely to provide consequences for their behaviors.

Additionally, high percentages of self-stimulation in the alone condi- tions may be due to the absence of any potential contingency manager. The higher mean percentage of self-stimulation in the alone unfamiliar setting condition, as opposed to the familiar setting, may be indicative of the children's response to an environment with no history of consequences associated with it. This particular finding should be emphasized because it suggests that "familiarity" is not solely a function of a therapist's previous presence. Still, this issue should be examined in more detail. In particular, the effect of a familiar individual who had previously administered conse- quences could be compared with the effect of a familiar individual who had not previously administered consequences. This would help clarify the exact nature of "familiarity."

Another important finding was that self-stimulation was inversely related to correct responding. This is consistent with previous research (Lovaas et al., 1971; Koegel & Covert, 1972). Lovaas et al. described the relationship of self-stimulation and correct responding as one of competing reinforcers. That is, either the children engaged in self-stimulation, did not respond cor- rectly to the task at hand, and did not sample the extrinsic reinforcers, or they responded correctly to the task, did not engage in self-stimulatory behavior, and were provided with the reinforcers. Apparently autistic self- stimulation (and thereby correct responding) may be a function of both antecedent stimulus conditions and extrinsic contingencies.

Of relevance here is that two children demonstrated a positive but insignificant relationship between correct responding and self-stimulation. Child 1 seldom responded correctly, while Child 5 seldom erred. Moreover, Child 1 demonstrated the highest percentages of self-stimulation, while Child 5 demonstrated the lowest. These findings are difficult to explain in that a high rate of self-stimulation may interfere with correct responding, or, alternative- ly, very low-functioning children who have great difficulty in learning may resort to higher levels of self-stimulation to attain sensory reinforcement because they would not often receive food or praise. In either case, one must be cautious in generalizing from these data in that the negative correlation

Autistic Self-Stimulation 43

between self-stimulation and correct responding may not apply to extreme cases.

The present results suggest several alternative treatment strategies to facilitate a generalized suppression of self-stimulation. If the children indeed respond appropriately in situations that are associated with the administra- tion of contingencies for their self-stimulation, then one might want to in- crease the similarity between therapy and nontherapy situations. Employing multiple therapists might help since therapists often administer contingen- cies in a relatively distinctive manner (Lovaas & Simmons, 1969; Stokes & Baer, 1977; Stokes, Baer, & Jackson, 1974). Additionally, requiring treat- ment to be performed in a variety of settings has enhanced generalization beyond the therapy room (Griffiths & Craighead, 1972; Lovaas & Simmons, 1969). The problem with these procedures is that it is impossible to know a priori when sufficient exemplars have been employed. Another possible procedure would be to employ one therapist but ensure that the therapist administers contingencies in an irregular, intermittent manner. Or a variety of consequences could also be employed (Charlop, Burgio, Iwata, & Ivan- cic, 1983; Egel, 1981). Assuming the child will continue to respond, this pro- cedure would circumvent the problem of anticipating the number of exemplary therapists needed for that child. Also, the unfamiliarity of the treatment procedures would be ensured. The result might be a generalizable suppression of self-stimulatory behaviors.

REFERENCES

Azrin, N. H., Kaplan, S. J., & Foxx, R. M. (1973). Autism reversal: Eliminating stereotyping self-stimulation of retarded individuals. American Journal o f Mental Deficiency, 78, 241-248.

Baer, D. M., Wolf, M. M., & Risley, T. (1968). Some current dimensions of applied behavior analysis. Journal o f Applied Behavior Analysis, 1, 91-97.

Cataldo, M. F., Bessman, C. A., Parker, S. H., Pearson, J. E. R., & Rogers, M. C. (1979). Behavioral assessment for pediatric intensive care units. Journal o f Applied Behavior Analysis, 12, 83-97.

Charlop, M. H. (1981, August). The occurrence of echolalia in unfamiliar settings. Paper presented at the meeting of the American Psychological Association, Los Angeles.

Charlop, M. H., Burgio, L., Iwata, B. A., & Ivancic, M. (1983, May). The use o f punisher variation in the treatment o f developmentally disabled children. Paper presented at the meeting of the Association for Behavior Analysis, Milwaukee.

Charlop, M. H., Schreibman, L., Mason, J., & Vesey, W. (1983). Behavior-setting interac- tions of autistic children: A behavioral mapping approach to assessing classroom behaviors. Analysis and Intervention in Developmental Disabilities, 3, 359-371.

Egel, A. L. (1981). Reinforcer variation: Implications for motivating developmentally disabled children. Journal o f Applied Behavior Analysis, 14, 345-350.

Griffiths, H., & Craighead, W. E. (1972). Generalization in operant speech therapy for misar- ticulation. Journal of Speech and Hearing Disorders, 37, 457-468.

44 Runco, Charlop, and Schreibman

Koegel, R. L., & Covert, A. (1972). The relationship of self-stimulation to learning in autistic children. Journal of Applied Behavior Analysis, 5, 381-387.

Koegel, R. L., Egel, A. L., & Dunlap, G. (1980). Learning characteristics of autistic children. In W. Sailor, B. Wilcox, & L. Brown (Eds.), Methods of instruction with severely han- dicapped students (pp. 259-301). Baltimore: Brookes.

Koegel, R. L., Firestone, P. B., Kramme, E. W., & Dunlap, G. (1974). Increasing spontaneous play by suppressing self-stimulation in autistic children. Journal of Applied Behavior Analysis, 7, 521-528.

Koegel, R. L., & Rincover, A. (1977). Research on the difference between generalization and maintenance in extra-therapy responding. Journal of Applied Behavior Analysis, 10, 1-12.

Koegel, R. L., Schreibman, L., Britten, R. L., Burke, J. C., & O'Neill, R. E. (1982). A com- parison of parent training to direct child treatment. In R. L. Koegel, A. Rincover, & A. L. Egel (Eds.), Educating and understanding autistic children (pp. 260-279). San Diego: College Hill Press.

Lovaas, O. I., Koegel, R. L., Simmons, J. Q., & Long, J. S. (1973). Some generalization and follow-up measures on autistic children in behavior therapy. Journal of Applied Behavior Analysis, 6, 131-166.

Lovaas, O. I., Litrownik, A., & Mann, R. (1971). Response latencies to auditory stimuli in autistic children engaged in self-stimulatory behavior. Behaviour Research and Therapy, 9, 39-49.

Lovaas, O. I., Schaeffer, B., & Simmons, J. Q. (1965). Building social behavior in autistic children by use of electric shock. Journal of Experimental Research and Personality, 1, 99-109.

Lovaas, O. I., & Simmons, J. Q. (1969). Manipulation of self-destruction in three retarded children. Journal of Applied Behavior Analysis, 2, 143-157.

Powell, J., Martindale, A., & Kulp, S. (1975). An evaluation of time-sample measures of behavior. Journal of Applied Behavior Analysis, 9, 463-469.

Repp, A. C., Deitz, S. M., & Spier, N. A. (1974). Reducing stereotypic responding of retarded individuals by differential reinforcement of other behaviors. American Journal of Men- tal Deficiency, 79, 279-284.

Rincover, A. (1978). Sensory extinction: A procedure for eliminating self-stimulatory behavior in autistic children. Journal of Abnormal Child Psychology, 6, 299-310.

Rincover, A., & Koegel, R. L. (1975). Setting generality and stimulus control in autistic children. Journal of Applied Behavior Analysis, 8, 235-246.

Rincover, A., Newsom, D. D., Lovaas, O. I., & Koegel, R. L. (1977). Some motivational pro- perties of sensory stimulation in psychotic children. Journal of Experimental Child Psychology, 24, 312-323.

Risley, T. (1968). The effects and side effects of the use of punishment with an autistic child. Journal of Applied Behavior Analysis, 1, 21-34.

Ritvo, E. R., & Freeman, B. J. (1978). National Society for Autistic Children definition of the syndrome of autism. Journal of Autism and Childhood Schizophrenia, 8, 162-167.

Ritvo, E. R., Ornitz, E. M., & LeFranchi, S. (1968). Frequency of repetitive behaviors in early infantile autism. Archives of General Psychiatry, 19, 341-347.

Rutter, M. (1978). Diagnosis and definition of childhood autism. Journal of Autism and Childhood Schizophrenia, 2, 139-161.

Schreibman, L., & Mills, J. I. (1983). Infantile autism. In T. H. Ollendick & M. Hersen (Eds.), Handbook of child psychopathology (pp. 123-142). New York: Plenum.

Stokes, T. F., & Baer, D. M. (1977). An implicit technology of generalization. Journal of Ap- plied Behavior Analysis, 10, 349-367.

Stokes, T. F., Baer, D. M., & Jackson, R. L.(1974). Programming and generalization of greeting responses in four retarded children. Journal of Applied Behavior Analysis, 7, 599-610.

Wahler, R. G. (1969). Setting generality: Some specific general effects of child behavior therapy. Journal of Applied Behavior Analysis, 2, 239-246.

Walker, H. M., & Buckley, N. K. (1972). Programming generalization and maintenance of treat- ment effects across time and across setting. Journal of Applied Behavior Analysis, 5, 209-224.

Whalen, C. K., Henker, J., Collins, B. E., Finck, D., & Dotemoto, S. (1979). A social ecology of hyperactive boys: Modification effects in structured classroom environments. Jour- nal of Applied Behavior Analysis, 12, 65-81.