Journal of Autism and Development Disorders, Vol. 19, No. 2, 1989

The Effects of Haloperidol on Discrimination Learning and Behavioral Symptoms in Autistic Children 1 Lowell T. Anderson and Magda Campbell 2

Department of Psychiatry, New York University Medical Center Phillip Adams

New York State Psychiatric Institute Arthur M. Small, Richard Perry, and Juanita Shell

Department of Psychiatry, New York University Medical Center

This double-blind and placebo-controlled clinical trial in autistic children had three objectives: (a) to replicate earlier findings that haloperidol administration is associated with a significant reduction o f behavioral symptoms; (b) to further assess its safety when given on a short-term basis; and (c) to assess whether it has an effect on discrimination learning. Forty-five children, 2. 02 to 7.58years old (M = 4.49), completed this crossover design, with random assignment to treatment sequences. Haloperidol was shown to be a powerf u l therapeutic agent when administered f o r 4 weeks and free o f side effects; at doses ranging f r o m 0.25 to 4.0 m g / d a y (M = 0.844), there was a clinically and statistically significant reduction o f a variety o f symptoms. Under the given conditions, the children failed to learn on either haloperidol or placebo.

~This work was supported in part by National Institute of Mental Health grant MH 32212 and was aided by Social and Behavioral Sciences Research grant 12-108 from March of Dimes Birth Defects Foundation (Dr. Campbell). The authors thank Dr. Dennis Grega and Mr. Charles W. Hamilton for participating in this research; Dr. Michael Palij for analyzingthe discrimination learning data; Dr. John E. Overall for statistical consultation; and McNeil Pharmaceutical for Haldol and matching placebo tablets. 2Address all correspondence to Magda Campbell, Department of Psychiatry, New York University Medical Center, 550 First Avenue, New York 10016.

227 0162-3257/89/0600-0227506.00/0 9 1989 Plenum PublishingCorporation

228

Anderson, Campbell, Adams, Small, Perry, and Shell

INTRODUCTION We found in two controlled studies in young autistic children that haloperidol, when administered on short-term basis at dally doses ranging from 0.5 to 4.0 mg, has significant clinical effects, without side effects, and facilitates learning in the laboratory (Anderson et al., 1984; Campbell et al., 1978). Using multiple independent raters and multiple dependent measures, including objective rating scales, we found that administration of this potent dopamine blocking agent is associated with reductions in withdrawal, stereotypies, and abnormal object relations (Anderson et al., 1984; Campbell et al., 1978). In the laboratory, haloperidol facilitated the acquisition of words (Campbell et al., 1978) and had a facilitative effect on successive discrimination learning (Anderson et al., 1984). Because facilitation of discrimination learning was independent of reduction of stereotypies and motor activity in the laboratory (Anderson et al., 1984), we speculated that discrimination learning was facilitated as a result of the drug's effect on attentional mechanisms. Most importantly, we found that this effect on learning was retained when the children were switched from haloperidol treatment to placebo. Since the same learning task was used throughout the study, we did not know if this retention was specific to previously learned discrimination, or a general improvement in learning. It is possible that children were simply able to retain the previously learned discrimination. Therefore, the present study tested the enhancement of general learning ability by comparing learning scores of a group that had been on haloperidol to a group that had only been on placebo while learning a new discrimination task. Short- and long-term retention and generalization effects were also investigated in this trial. Furthermore, we wished to replicate our earlier findings concerning clinical-behavioral effects of haloperidol and its safety when given over a period of 1 month.

METHOD Sample The diagnosis of infantile autism was made independently by three child psychiatrists according to DSM-III (American Psychiatric Association, 1980) criteria. Exclusion criteria have been reported elsewhere (Anderson et al., 1984). The sample consisted of 45 autistic children, 35 boys and 10 girls. Their ages ranged from 2.02 to 7.58 years (M = 4.49 years, SD = 1.16 years). Intellectual functioning ranged from profound to borderline retardation: In the total sample, 42 subjects (95.5%0) performed intellectually at a mild or lower level of retardation; of these, 29 subjects (64.4%) were either profound-

Effects of Haloperidol in Autistic Children

229

ly or severely retarded. All subjects were patients of the Therapeutic Nursery of Bellevue Hospital, Psychiatric Division, Children's4npatient Service. Complete medical diagnostic workups, as well as psychiatric observation, psychological, neurological, pediatric, hearing and speech evaluations, and laboratory studies, were carried out on all children, prior to enrollment in the study. Written informed consent was obtained from the children's parents or legal guardians after the research procedure had been fully explained.

Design A double-blind and placebo-controlled crossover design was employed. Following the 2-week placebo period, during which baseline ratings were carried out, the children were randomly assigned to one of three treatment sequences (Groups) shown in Table I: Group I, haloperidol-placebo-placebo (HPP); Group II, placebo-haloperidol-placebo (PHP); and Group III, placebo-placebo-haloperidol (PPH). There were three testing periods, each of 4-week duration: Phase A, Phase B, and Phase C. Thus, the entire study, for each child, was of 14-week duration. Two discrimination tasks were given in each phase, each task lasting for 2 weeks. A total of five different discrimination tasks were used in the study, with Task 1 repeated in the first and final phases to test long-term retention. The two tasks in each phase were presented in counterbalanced order such that half of the subjects in each condition were randomly assigned to begin with Task 1 and the rest with Task 2. Tasks 1 and 2 were used in Phase A; Tasks 3 and 4 in Phase B; and Tasks 5 and 1 in Phase C. Counterbalancing of the order of haloperidol/placebo treatments across the three phases ensured that treatment effects and task differences were not confounded. Short-term generalization of learning was investigated by comparing performance in learning of related kinds of tasks. Specifically, generalization was investigated by between- and within-group comparisons of performance on Tasks 1 versus 3 and Tasks 2 versus 4. Long-term generalization was evaluated by between-group comparisons on Task 5, and long-term retention was investigated by within-group comparisons on Task 1 (Phases A and C). The use of different kinds of tasks allowed investigation of the effects of haloperidol on different types of abilities called for by these tasks.

Tasks The tasks were three successive (matching forms, dot discrimination, and animate/inanimate figure discrimination), and two simultaneous (position: sequential learning 1, and color: sequential learning 2) discrimination tasks. For each of the three phases the tasks were administered as follows:

230

Anderson, Campbell, Adams, Small, Perry, and Shell

Phase A Matching Discrimination. Subjects are presented with either matched or mismatched pairs of four geometric forms (Square, Triangle, Cross, Circle). Subjects are reinforced when a lever-press response is made to matched pairs (s + ) but not to mismatched pairs ( s - ) . Position Discrimination. One of three colors (red, blue, or yellow) is projected onto all three translucent screens (stimulus onset). Each o f the screens when pressed activates a microswitch and this is recorded as a response by the computer. Only one of the three screens, however, is correct and generates a reinforcer when activated (s +). A 10-sec intertrial interval (all lights are off) ( s - ) occurs between each 15-see stimulus onset. Phase B Dot Discrimination. Either a 10-dot pattern or a blank screen is projected onto a center panel. Reinforcement is given when a lever-press response is made to the 10-dot pattern ( s + ) but not to the blank screen ( s - ) . Subjects who respond correctly on at least 80~ of a block o f 48 trials advance to the next level which is 10 dots (s + ) versus 1 dot ( s - ) in the next session. Subsequent sessions each add one dot to the s - if the minimum criterion of 80~ correct is reached. Color Discrimination. Each of the three screens is transilluminated with a different color (red, blue, yellow). The screens change colors for each o f the 18 trials given in a session. One of the three colors is correct and a reinforcer is delivered when the appropriate screen is depressed (s +). No intertrial interval ( s - ) is utilized for this task. Phase C Animate~Inanimate Figure Discrimination. Slides o f line drawings o f either people or inanimate objects are projected onto the center panel. A lever pressed in response to projection o f a person (or people) generates a reinforcer ( s + ) , but a lever pressed when an inanimate object (or objects) is projected is not reinforced ( s - ) . Matching Discrimination. Repeated. The Matching, Dot, and Animate/Inanimate Discrimination tasks consist o f 48 trials (24 s + , 24 s - ) . Stimulus exposure o f s + trials lasts for at most 7 see; s - stimuli are exposed for at least 7 see. The Position Discrimination task contains 20 trials. A trial lasts 15 see with 10-see intertrial intervals. Color discrimination consists of 18 trials with each trial lasting 15 see. The color task uses no intertrial intervals.

Effects of Haloperidol in Autistic Chndren

231

A total o f 12 sessions were given for each of the tasks over a 2-week period. Half of the sessions were given during the first Week and half in the second week. Stimuli for the lever-press tasks were counterbalanced and equally represented over the 12 sessions. Stimuli for the screen-press tasks consisted of nine sets of counterbalanced color and screen combinations. Three of these nine are repeated in each of the last 3 sessions to bring the total to 12.

Training During the placebo baseline period, children are trained to approach the reinforcement dispenser. Subjects are then trained to press the response screens when they are color-illuminated and to press the lever when the center screen is illuminated with white light (from an empty slide projector). Training is accomplished by the method of successive approximations. Responses are reinforced on a continuous reinforcement schedule during 15-min training sessions. When subjects begin to average at least one response per trial, a 3:1 ratio schedule is instituted. During training, subjects are exposed to all possible color and screen combinations. Subjects are trained to distinguish between the different response modes called for by the different tasks. Screen-press behavior is required when screens are color illuminated, and lever-press behavior is the correct response mode when the center panel is white-illuminated. Once a child starts the treatment phase (placebo or haloperidol) training begins on a successive discrimination learning paradigm. For the position and color discrimination tasks, a multiple 3: I ratio schedule is employed. The Matching, Dot, and Animate/Inanimate tasks use a continuous singlereinforcement schedule. The s + and s - were described in the Design section. Responses made during the last 5 sec of each s - period delay onset of the next stimulus exposure for 5 sec. Thus, stimulus exposure does not occur until the subject has not responded for at least 5 sec. This procedure is employed in order to reduce the likelihood of supervenient (or supernumerary) reinforcement of screen- or lever-pressing behavior during the s - period.

Apparatus The experimental apparatus is computer-operated; it consists of a BRS/LVE Human Test Console on which are mounted three Manipulandi projector screens and a heavy-duty lever. A Kodak Carousel slide projector projects the visual stimuli for the Matching, Dot, and Animate/Inanimate Discrimination tasks. The three 5-inch x 3-inch screens used for color and position discrimination are translucent panels illuminated by 6-volt painted

Anderson, Campbell, Adams, Small, Perry, and Shell

232

b a y o n e t b u l b s . These a r e p o w e r e d b y a 6-volt p o w e r s u p p l y c o n t r o l l e d b y the m i c r o c o m p u t e r using a M e d A s s o c i a t e s interface f o r timing, c o n t r o l , a n d o u t p u t f o r m a t t i n g . A G e r b r a n d s U n i v e r s a l F e e d e r dispenses r e i n f o r c e r s (M & M candies, sweetened cereal) to the children. Pressure-sensitive switches are m o u n t e d u n d e r a n i n d o o r - o u t d o o r carp e t every 6 inches in a n a r e a s u r r o u n d i n g the test c o n s o l e (Muller, C r o w , & C h e n e y , 1979). O u t p u t f r o m the switches feeds into the M e d A s s o c i a t e s i n t e r f a c e a n d t h e n into the c o m p u t e r to r e c o r d overall activity levels. Sitting in the s a m e r o o m as the child, a n o b s e r v e r presses a b u t t o n d u r ing each occurrence o f stereotypic b e h a v i o r . D u r a t i o n o f stereotypic b e h a v i o r is r e c o r d e d b y the c o m p u t e r t h r o u g h a c o n t r o l p a n e l o p e r a t e d b y t h e experimenter.

Behavioral

Ratings

A s s h o w n in T a b l e I, d u r i n g W e e k s 1, 2 (baseline ratings) a n d 6, 10, a n d 14 ( t r e a t m e n t s ) t h e c h i l d r e n were r a t e d a n d v i d e o t a p e d in a s e m i s t r u c t u r e d interview a t fixed times, as d e s c r i b e d in p r e v i o u s r e p o r t s ( A n d e r s o n et al., 1984). T h e children were rated i n d e p e n d e n t l y b y two child psychiatrists o n t h e C h i l d r e n ' s P s y c h i a t r i c R a t i n g Scale ( C P R S ; Psychopharmacology Table I. Double-Blind, Placebo-Controlled Crossover Design* Discrimination learning tasksc Group b

Baseline period 2 weeks Placebo

Phase 1 (A) 2 weeks 2 weeks

Phase 2 (B) 2 weeks 2 weeks

Phase 3 ((2) 2 weeks 2 weeks

Placebo

Placebo

Haloperidol

I

HPP

(n = 7) $1

M

$2

D

M

X

(n = 7) M

Sl

D

$2

X

M

Placebo II PHP

Haloperidol

(n = 7) $I

M

$2

D

M

X

(n = 7) M

$1

D

$2

X

M

Placebo III PPH XX

XX

Placebo

Placebo

Haloperidol

(n = 7)$1

M

$2

D

M

X

(n = 7) M

$1

D

$2

X

M

XX

XX

XX

~ behavioral part of this study was completed by 45 children; 42 children completed the discrimination learning part. bn = 14 in each treatment sequence. cS, = sequential learning 1 (position discrimination); $2 = sequential learning 2 (color discrimination); M = matching discrimination; D = dot discrimination; X = animate/inanimate discrimination; XX = behavioral, and abnormal movement ratings, videotaped.

Effects of Haloperidol in Autistic Children

233

Bulletin, 1985) and the Clinical Global Impressions Scale (CGI; Psychopharmacology Bulletin, 1985). The child's nursery teacher completed the Conners Parent-Teacher Questionnaire (PTQ; Psychopharmacology Bulletin, 1985) on the same schedule. Untoward drug effects were monitored daily and rated on the Treatment Emergent Symptoms Scale developed by the Psychopharmacology Branch of the NIMH (Psychopharmacology Bulletin, 1985) and on a haloperidol side effect checklist.

Medication

Haloperidol or identical matching placebo tablets (supplied by McNeil Pharmaceutical Laboratories) were administered daily at fixed times in the morning by a research nurse blind to the type of medication. Compliance was fully achieved. Dosage level of haloperidol or matching placebo was individually regulated; the starting dose was 0.5 mg/day. Increments were done at regular intervals until therapeutic and/or untoward effects (behavioral or other) were observed; in this manner optimal dosage level was determined. The dose of 4.0 mg/day was not to be exceeded. The first 2 weeks of each post-washout period were reserved for dosage regulation. The optimal dosage attained at the beginning of the study, for each subject, was also to be the optimal dose at each replication, in order to ensure comparability across conditions.

RESULTS Dosage

The therapeutic doses of haloperidol ranged from 0.25 to 4.0 rag/day (M =0.844, SD = 0.570) or from 0.016 to 0.184 mg/kg/day (M = 0.047, SD = 0.028). There were no significant differences in the three treatment groups in drug dosage. At therapeutic doses, which were individually regulated, no side effects were observed in this short-term study, where each child received haloperidol for 4 weeks only. Above therapeutic doses, excessive sedation and acute dystonic reaction were the most common side effect with haloperidol. All laboratory studies remained within normal limits.

Children's Psychiatric Rating Scale The data in this crossover design were reordered so that the effects of haloperidol could be assessed as a main effect in a repeated-measures ANOVA

234

Anderson, Campbell, Adams, Small, Perry, and Shell Table II. Reordering of Data to Evaluate Effects of

Haloperidol, Tested as a Main Effect in the Repeated Measures ANOVA~ Group 1 2 3

Originaldata order H P1 PI

PI H P2

Reorderingfor analysis

P~ P2 H

H H H

P1 P2 P1

P2 PI P2

~ = haloperidol; P = placebo. Subscripts indicate the placebo order. using the baseline rating as covariate. The grouping factor (Order of Presentation) had three levels: H P P , P H P , and P P H . The within-subject factor (Treatment) also had three levels: haloperidol, first and second placebo. The data for each of the three groups were reordered for analysis as shown in Table II (in which subscripts indicate the placebo order) after which the efficacy of haloperidol was tested as a main effect. In the reordered layout, order and carry-over effects (if any) should contribute to the interaction term in the two-way Groups x Treatments A N O V A , The main effect for haloperidol was significant for all three CPRS factors (Overall & Campbell, 1988), the sum of all 14 CPRS items appropriate for autistic children (Campbell & Palij, 1985), and for 7 individual scale items. A similar repeatedmeasures A N O V A compared the haloperidol treatment condition with baseline phases. The mean CPRS factor ratings across groups and phases are presented in Table III (Autism factor), Table IV (Hyperactivity factor), and Table V (Conduct Problem factor). These tables show both reduction in severity associated with haloperidol and a tendency toward increased severity after drug withdrawal. However, the Groups x Treatments interaction in the reordered layout did not reach statistical significance for this or any of the other clinical measures analyzed, thus failing to confirm the presence of a true drug withdrawal effect in this study.

Table III. Baseline-Adjusted Mean

of CPRS Autism Factor Scores.(N = 45)* Groupb

1

Phase 2

3

HPP 4.88 5 . 3 2 5.21 PHP 5.16 4 . 3 4 5.12 PPH 5.16 4 . 8 9 4.11 *Effect for haloperidol; F(2, 84) = 17.49, p < .0001. bn = 15 for each group.

Effects of Haloperidol in Autistic Children

235

Table IV. Baseline-Adjusted Mean of CPRS Hyperactivity Factor Scores (N = 45) = Phase Group b HPP PHP PPH

1

2

3

4.07 4.44 4.42

5.09 3.79 4.70

5.00 4.82 3.92

aEffect for haloperidol: F(2, 84) = 13.22, p < .0001. bn = 15 for each group.

Clinical Global Impressions CGI ratings were, similaxly, analyzed using repeated-measures ANOVA. The baseline rating was used as a covariate in the analysis of Severity of Illness. The main effect for haloperidol was highly significant for all ratings: Severity of Illness, F(2, 84) = 30.39, p < .0001; Improvement, F(2, 84) = 39.29, p < .0001; Efficacy, F(2, 84) = 56.99, p < .0001. The means of Severity ratings across groups and phases are shown in Table VI. For all CGI variables, inspection of means shows that all changes were in the direction of decreased symptoms during haloperidol treatment.

Conners Parent- Teacher Questionnaire Item 5 on the original scale (Fidgeting) was deleted and replaced with a rating of Stereotypies to conform more nearly to the behavioral syndrome of autistic children. The repeated measures analysis of covariance using baseline rating as covariate indicated significant reduction in the PTQ Total score,

Table V. Baseline-Adjusted Mean CPRS Conduct Problem Factor Scores (N = 45) ~ Phase Group b HPP PHP PPH

1

2

3

1.95 1.99 1.89

2.78 1.92 2.18

2.61 2.02 1.69

~ for haloperidol; F(2, 84) = 3.29, p < .04. bn = 15 for each group.

236

Anderson, Campbell, Adams, Small, Perry, and Shell Table VI. Baseline-Adjusted Mean

CGI SeverityScores (N = 45)* Phase Groupb 1 2 3 HPP 4.89 5.71 5.66 PHP 5.43 4.43 5.40 PPH 5.50 5.27 4.14 *Effectfor haloperidol:F(2, 84) = 30.39, p < .0001. An = 15 for each group. F(2, 68) = 4.11, p < .02. Only 1 o f the 10 items in the scale was significantly reduced with haloperidol: Temper outbursts (Item 10) with F(2, 68) = 3.78, p < .03. D i s c r i m i n a t i o n Learning

Successive Discrimination Tasks O f the 45 children, 29 completed all the learning tasks. The effects of haloperidol on the learning measures was assessed by mixed-model ANOVAs including three levels of subject grouping (HPP, P H P , PPH), and, as appropriate, within-subject factors of Drug Condition, Sessions, and Display Type (reinforcement available and not available). Only the matching forms tasks produced results indicative of learning. However, this effect was independent of haloperidol administration. Even though there was both a statistically significant increase in the number of correct responses across phases, F(11, 36) = 12.76, p < .001, and a simultaneous decrease in the number of incorrect responses, (F(1, 36) = 12.11, p < .05, the overall response rates were very low and the changes across phases trivial in magnitude. For example, the proportion o f correct responses from Phase I (52.4070) to Phase III (57.907o) showed little improvement and remained close to chance. Subjects also showed evidence o f gradual learning across sessions on the matching forms task but this effect, although statistically significant, was minimal, close to chance, and, again, independent of haloperidol administration. On this task, the proportion o f correct responses to the matched displays increased from 51.7~ in Session 1 to 55.2070 in Session 12. Performance on both the Dots and Animate/Inanimate Discrimination tasks was characterized by a low (near chance) level of responding with higher scores on tasks that required that the child not respond. Laboratory meas-

Effects of Halopefldol in Aufls/ic Children

237

ures of motor activity and stereotypies showed no significant effects of haloperidol during any of the three successive disriminafion tasks. Simultaneous Discrimination Tasks The effects of haloperidol on the Color and Position Discrimination tasks were assessed by analyzing several dependent measures (number of correct and incorrect responses, motor activity, and stereotypies) with analysis of variance including one between-subjects factor (Groups) and one withinsubjects factor (Sessions). There is little evidence suggesting that subjects were able to learn either of the simultaneous discrimination tasks and there was no evidence that haloperidol was superior to placebo in facilitating learning on these tasks. For example, on the Color Discrimination task subjects made more errors than correct responses and the average number of responses per trial (1.3) was below the 3:1 ratio of the reinforcement schedule. This low rate of response indicates that on average subjects did not make enough responses to obtain a reinforcement. Analysis of the Position Discrimination task indicated that haioperidol had no facilitative effect on the number of correct or incorrect responses within a session, and no ameliorative effects on either motor activity or stereotypies within sessions. For example, the overall ratio of correct responses on the Position Discrimination task was 58.17~ well above the chance level of 33.307o (three panels were simultaneously presented to the subject). However, this ratio did not change either across sessions (Session 1 = 57.2607o and Session 12 = 57.0007o) or with drug treatment phase indicating that haloperidol did not improve performance relative to placebo.

COMMENTS The results show that haloperidol is a powerful drug in reducing, both statistically and clinically, maladaptive behaviors in many autistic children who require pharmacotherapy. The children were calmer without being sedated, showed decreases in hyperactivity, temper tantrums, withdrawal, and stereotypies, and showed increase in relatedness. These data are in agreement with our previous findings (Anderson et al., 1984; Campbell et al., 1978). As in our previous studies, by careful regulation of dosage for each individual child, none of the 45 patients had untoward drug effects, at therapeutic doses. We emphasize that the sample consisted of inpatients whose symptoms range from moderate to severe, and who are often referred to our

238

Anderson, Campbell, Adams, Small, Perry, and Shell

Therapeutic Nursery because they cannot be maintained at home and in various day programs or special schools. Thus, in three carefully designed and well-controlled studies the results show that haloperidol, a potent antidopaminergic drug, is an effective therapeutic agent for many autistic children who are in an enriched psychosocial environment. This finding is important both clinically and theoretically. Even though the evidence of dopamine involvement in this condition is sparse and inconclusive (Cohen, Caparulo, Shaywitz, & Bowers, 1977; for review, see Anderson & Hoshino, 1987; Campbell, Anderson, Green, & Deutsch, 1987; Campbell, Perry, Small, & Green, 1987; Spencer & Campbell, in press; Young, Kavanagh, Anderson, Shaywitz, & Cohen, 1982), indirect evidence based on clinical trials of dopamine agonists (Campbell, Perry et al., 1987) and of dopamine antagonists, such as the present and previous studies (for review, see Campbell, 1987; Anderson et ai., 1984; Campbell et al., 1978; Fish, Shapiro, and & Campbell, 1966) are strongly supportive of such involvement. Under the conditions of this study, and employing the tasks specified above, haioperidol did not have generalized faclitating effects on discrimination learning. We can only speculate why haloperidol failed in this respect: Was the learning period too short or the tasks too difficult for our patients or was the reason the limited effect of the drug itself? Further research, employing a different design and tasks may provide us with an answer to these questions. However, it is important that haloperidol administration did not have an adverse effect on learning during the 4-week period, and this itself is important information regarding a population where the majority is of subnormal intellectual functioning, having severe learning difficulties.

REFERENCES American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders Ord ed.). Washington, DC: Author. Anderson, L. T., Campbell, M., Grega, D. M., Perry, R., Small, A. M., & Green, W. H. (1984). Haloperidol in infantile autism: Effects on learning and behavioral symptoms. American Journal of Psychiatry, 141, 1195-1202. Anderson, G. M., & Hoshino, Y. (1987). Neurochemical studies of autism. In D. J. Cohen & A. M. Donnellan (Eds.), Handbook of autism andpervasive developmental disorders (pp. 166-191). New York: Wiley. Campbell, M. (1987). Drug treatment of infantile autism: The past decade. In H. Y. Meltzer (Ed.), Psychopharmacology: The third generation of progress (pp. 1225-1231). New York: Raven Press. Campbell, M., & Palij, M (1985). Behavioral and cognitive measures used in psychopharmacological studies of infantile autism. Psychopharmacology Bulletin, 21(4), 1047-1053. Campbell, M., Anderson, L. T., Green, W. H., & Deutsch, S. I. (1987). Psychopharmacology. In D. J. Cohen & A. M. Donnellan (Eds.) (R. Paul, Assoc. Ed.), Handbook of autism and pervasive developmental disorders (pp. 545-565). New York: Wiley.

Effects of Haloperidol in Autistic Children

239

Campbell, M., Anderson, L. T., Meier, M., Cohen, I. L., Small, A. M., Samit, C., & Sachar, E. J. (1978). A comparison of haloperidol, behavior therapy and their interaction in autistic children. Journal of the American Academy of Child Psychiatry, 17, 640-655. Campbell, M., Perry, R., Small, A. M., & Green, W. H. (1987). Overview of drug treatment in autism. In E. Schopler & G. B. Mesibov (Eds.), Neurobiological i~'ues in autism (pp. 341-356). New York: Plenum Press. Cohen, D. J., Caparulo, B. K., Shaywitz, B. A., & Bowers, M. B. (1977). Dopamine and serotonin metabolism in neuropsychiatrically disturbed children. Archives of General Psychiatry, 34, 545-550. Fish, B., Shapiro, T., & Campbell, M. 0966). Long-term prognosis and the response of schizophrenic children to drug therapy: A controlled study of trifluoperazine.American Journal of Psychiatry, 123, 32-39. Muller, P. G., Crow, R. E., & Cheney, C. D. (1979). Schedule-induced locomotor activityin humans. Journal of the Experimental Analysis of Behavior, 31, 83-90. Overall, J. E., & Campbell, M. (1988). Behavioral assessment of psychopathology in children: Infantile autism. Journal of ClinicalPsychology, 44, 708-716. Psychopharmacology Bulletin.(1985). Rating scalesand assessment instruments for use in pediatric psychopharmacology research ]Special feature]. 21(4), 765-770. Spencer, E. K., & Campbell, M. (in press). Infantileautism. In T. Dinan (Ed.), Principlesand practices of biologicalpsychiatry. London: Clinical Neuroscience Publishers. Young, J. G., Kavanagh, M. E., Anderson, G. M., Shaywitz, B. A., & Cohen, D. J. (1982). Clinical neurochemistry of autism and associated disorders. Journal of Autism and Developmental Disorders, 12, 147-165.