Evaluations of Repetitive Behavior in Typically Functioning Adults and Intellectually Disabled Children

MISSING IMAGE

Material Information

Title:
Evaluations of Repetitive Behavior in Typically Functioning Adults and Intellectually Disabled Children Implications of the Demand Condition
Physical Description:
1 online resource (79 p.)
Language:
english
Creator:
Bosch,Amanda B
Publisher:
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Psychology
Committee Chair:
Vollmer, Timothy R
Committee Members:
Abrams, Lise
Iwata, Brian
Elder, Jennifer H

Subjects

Subjects / Keywords:
demand -- repetitive -- stereotypy
Psychology -- Dissertations, Academic -- UF
Genre:
Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract:
Repetitive behavior occurs in a wide variety of populations, settings, and topographies. Research that has focused on determining why repetitive behavior (e.g., nail biting, hair twirling, and skin scratching) occurs in typically functioning populations has concluded that the behavior is maintained by automatic reinforcement (e.g., Deaver, Miltenberger, & Stricker, 2001). However, various conditions may evoke repetitive behavior and establish the consequence produced by the behavior as a reinforcer. Woods and Miltenberger (1996) showed that typically functioning adults engaged in high levels of repetitive behavior when presented with a demand. . The absence of social consequences in their demand condition suggests that repetitive behavior in this condition may have been maintained by automatic reinforcement established as such by the presence of demands. This particular function is important because, given the presence of demands, researchers and clinicians could incorrectly conclude that the behavior is maintained by escape from demands. This possibility has important implications for the assessment and treatment of stereotypy, or repetitive behavior, in intellectually disabled children. Some previous researchers have concluded that stereotypy is escape-maintained if it occurs in the demand condition of an antecedent-based functional assessment (e.g., Durand and Carr, 1987). However, given the lack of programmed consequences, the behavior may actually be maintained by automatic reinforcement established as such by demands. Hence, we aimed to evaluate whether this function occurs in typically functioning adults and intellectually disabled children. Results of the Studies 1 and 2 showed that 22% of typically functioning adults and 33% of intellectually disabled children engaged in high levels of repetitive behavior when presented with demands. Results from Study 2 demonstrated that at least one topography of stereotypy occurred most often in the demand conditions for three individuals. Results from Study 3 confirmed that this behavior was not maintained by escape from demands, but rather by automatic reinforcement.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Amanda B Bosch.
Thesis:
Thesis (Ph.D.)--University of Florida, 2011.
Local:
Adviser: Vollmer, Timothy R.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2012-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2011
System ID:
UFE0043299:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 EVALUATIONS OF REPETITIVE BEHAVIOR IN TYPICALLY FUNCTIONING ADULTS AND INTELLECTUALLY DISABLED CHILDREN: IMPLICATIONS OF THE DEMAND CONDITION By AMANDA BOSCH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSIT Y OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Amanda Bosch

PAGE 3

3 To my parents, who tau ght me to persevere

PAGE 4

4 ACKNOWLEDGMENTS I thank Dr. Tim Vollmer for his guidan ce, his understanding, and his support; he has never been anything short of a wonderful mentor. I also would like to thank the rest of my committee members: Dr. Lise Abrams, Dr. Jennifer Elder, and Dr. Brian Iwata, for their invaluable input on this projec t. Also, I simply could not have completed this project without the hard work of my undergraduate students: Ashley Breeden, Danielle Broome, Valerie Bryant, Amanda Bullard, Ayla McClain, James Morrison, Alison Nyman, Ashley Stromberg, Diego Valbuena, and Andrea Zawoyski. I also thank Dr. Andrew Samaha for his endless assistance with data analysis in particular and computer issues in general. Furthermore, I thank my family and friends for putting up with me during qualifying exams and the completion of this project; I give thanks especially to Matthew Wrighton, who made 16 hour days of studying and writing possible.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 TA BLE OF CONTENTS ................................ ................................ ................................ .. 5 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Repetitive Behavior in Typically Functioning Individuals ................................ ......... 12 The Function of Repetitive Behavior ................................ ................................ ....... 12 Stereotypy Is Repetitive Behavior ................................ ................................ ........... 13 The Function of Stereotypy ................................ ................................ ..................... 14 Stereotypy That Occurs During Demands ................................ .............................. 16 General Purpose ................................ ................................ ................................ ..... 18 2 A N EVALUATION OF CONDITIONS THAT EVOKE REPETITIVE BEHAVIOR IN TYPICALLY FUNCTIONING ADULTS ................................ ............................... 19 Experiment 1: Purpose ................................ ................................ ........................... 19 Experiment 1: Method ................................ ................................ ............................. 19 Subjects and Setting ................................ ................................ ........................ 19 Target Behavior and Data Collection ................................ ................................ 20 Procedure ................................ ................................ ................................ ......... 21 Experiment 1: Results and Discussion ................................ ................................ .... 24 3 AN EVALUATION OF THE CONDITIONS THAT EVOKE STEREOTYPY IN INT ELLECTUALLY DISABLED CHILDREN ................................ ........................... 42 Experiment 2: Purpose ................................ ................................ ........................... 42 Experiment 2: Methods ................................ ................................ ........................... 42 Subjects and Settings ................................ ................................ ....................... 42 Target Behavior and Data Collection ................................ ................................ 43 Procedure ................................ ................................ ................................ ......... 43 Preference assessment ................................ ................................ ............. 43 Antecedent based assessment ................................ ................................ .. 44 Experiment 2: Results and Discussion ................................ ................................ .... 46

PAGE 6

6 4 AN EVALUATION OF REPETITIVE BEHAVIOR THAT OCCURS IN THE DEMAND CONDITION OF AN ANTECEDENT BASED ASSESSMENT ............... 60 Experiment 3: Pu rpose ................................ ................................ ........................... 60 Experiment 3: Methods ................................ ................................ ........................... 60 Subjects and Settings ................................ ................................ ....................... 60 Targe t Behavior and Data Collection ................................ ................................ 60 Procedure ................................ ................................ ................................ ......... 61 Experiment 3: Results and Discussion ................................ ................................ .... 62 5 GENERAL DISCUSSION ................................ ................................ ....................... 68 Overall Summary ................................ ................................ ................................ .... 68 Experiment 1 ................................ ................................ ................................ ........... 68 Experiment 2 ................................ ................................ ................................ ........... 70 Experiment 3 ................................ ................................ ................................ ........... 72 Conclusions ................................ ................................ ................................ ............ 73 LIST OF REFERENCES ................................ ................................ ............................... 76 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 79

PAGE 7

7 LIST OF TABLES Table P age 2 1 Operatio nal definitions for Experiment 1. ................................ ............................ 27 2 2 Summary data for Experiment 1 by condition. ................................ ................. 28 3 1 D emographic information for each subject ................................ ........................ 50 3 2 For each subject, topographies of stereotypy are listed. ................................ .... 51 3 3 Results of the free operant preference assessmen t are shown.. ........................ 52

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Percent of session with repetitive behavior for UG45 ................................ ......... 29 2 2 Percent of session with repetitive behavior for UG07, UG09, UG10, and UG11 ................................ ................................ ................................ .................. 30 2 3 Percent of session with repetitive behavior for UG12, UG13, UG14, and UG18.. ................................ ................................ ................................ ................ 31 2 4 Percent of session with repetitive behavior for UG20, UG21, UG22, and UG23 ................................ ................................ ................................ .................. 32 2 5 Percent of session with repetitive behavior for UG25, UG26, UG27, and UG29. ................................ ................................ ................................ ................. 33 2 6 Percent of session with repetitive behavior for UG32, UG33, UG34, and UG35. ................................ ................................ ................................ ................. 34 2 7 Percent of session with repetitive behavior for UG36, UG38, UG39, and UG40. ................................ ................................ ................................ ................. 35 2 8 Percent of session with repetitive behavior for UG43, UG44, UG45, and UG47. ................................ ................................ ................................ ................. 36 2 9 Percent of session with repetitive behavior for UG51, UG52, UG54, UG55, and UG56.. ................................ ................................ ................................ ......... 37 2 10 Percent of session with repetitive behavior for UG08, UG15, UG16, UG19, a nd UG28. ................................ ................................ ................................ .......... 38 2 11 Percent of session with repetitive behavior for UG46, UG48, UG50, and UG53. ................................ ................................ ................................ ................. 39 2 12 Percent of session with rep etitive behavior for UG24, UG42, and UG17.. ......... 40 2 13 Percent of session with repetitive behavior for UG30, UG31, UG37, UG41, and UG49.. ................................ ................................ ................................ ......... 41 3 1 Percent of session with repetitive behavior f or Gertrude, Jesus, and Pablo. ..... 54 3 2 Percent of session with repetitive behavior for Kandi, Mario, and Tobias .......... 55 3 3 Percent of session with repetitive behavior for Chad, Jasmine, Ethel, and Dante. ................................ ................................ ................................ ................. 56

PAGE 9

9 3 4 Percent of session with repetitive behavior for Marti n, Agatha, Ben, and Heidi. ................................ ................................ ................................ .................. 57 3 5 Percent of session with vocalizations (Ben) and hand stereotypy (Heidi) is displayed ................................ ................................ ................................ ............ 58 3 6 Pe rcent of session with repetitive behavior for Clyde. ................................ ...... 59 4 1 Results of the demand evaluation for Ben (top panel) and Clyde (bottom panel) are displayed. ................................ ................................ .......................... 66 4 2 al topographies are displayed. ................................ ................................ ............... 67

PAGE 10

10 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EVALUATIONS OF REPETITIVE BEHAVIOR IN TYPICALLY FUNCTIONING ADULTS AND INTELLECTUALLY DISABLED CHILDREN : IMPLICATIONS OF THE DEMAND CONDITION By Amanda Bosch Aug ust 2011 Chair: Timothy R. Vollmer Major: Psychology Repetitive behavior occurs in a wide variety of populations, settings, and topographies. Research that has focused on determining why repetitive behavior (e.g., nail biting, hair twirling, and skin s cratching) occurs in typically functioning populations has conclud ed that the behavior is maintained by automatic reinforcement (e.g., Deaver, Miltenberger, & Stricker, 2001 ). However, various conditions may evoke repetitive behavior and establish the con sequence produced by the behavior as a reinforcer. Woods and Miltenberger (1996) showed that typically functioning adults engaged in high levels of repetitive behavior when presented with a demand. The absence of social consequences in the ir demand cond ition suggests that repetitive behavior in this condition may have been maintained by automatic reinforcement established as such by the presence of demands This particular function is important because, given the presence of demands, researchers and cli nicians could incorrectly conclude that the behavior is maintained by escape from demands. This possibility has important implications for the assessment and treatment of stereotypy, or repetitive behavior, in intellectually disabled children. Some previ ous researchers have

PAGE 11

11 concluded that stereotypy is escape maintained if it occurs in the demand condition of an antecedent based functional assessment (e.g., Durand and Carr, 1987). However, given the lack of programmed consequences, the behavior may actua lly be maintained by automatic reinforcement established as such by demands. Hence, we aimed to evaluate whether this function occurs in typically functioning adults and intellectually disabled children. Results of the Studie s 1 and 2 show ed that 22 % of typically functioning adults and 33 % of intellectually disabled children engage d in high levels of repetitive behavior when presented with demands. Results from Study 2 demonstrated that at least one topography of stereotypy occurred most often in the dem and conditions for three individuals. Results from Study 3 confirm ed that this behavior was not maintain ed by escape from demands, but rather by automatic reinforcement.

PAGE 12

12 CHAPTER 1 INTRODUCTION Repetitive Behavior in Typically Functioning Individuals Re petitive behavior such as nail biting, skin picking, and hair pulling are often called habits or nervous habits and occur both in general and clinical populations (Woods, Miltenberger, & Flach, 1996). Through a survey given to 246 undergraduate students, Woods et al. determined that roughly 20% of college students reported repeatedly touching their face, shaking their legs, chewing on their mouth, and playing with pencils. Fifteen percent also endorsed repeatedly playing with their hair or jewelry. Nail biting, head scratching, knuckle cracking, and lip touching were each reported by 10% of the students sampled. Unfortunately, the authors d id not provide information about the overall prevalence of repetitive behavior (i.e. the total percentage of subject s who engaged in at least one type of habit). An earlier study using a similar method reported that all of the 142 undergraduate subjects endorsed engaging in at least one type of habit behavior (Hansen, Tishelmian, Hawkins, & Doepke, 1990). However, as W oods et al. noted, the criterion used by Hansen et al. to determine whether a subject behavior occasionally for it to count as a habit. Thus, the overall prevalence of ha bits or repetitive behavior in typically functioning individuals is not known. The Function of Repetitive Behavior Several studies have evaluated the conditions under which repetitive behavior, namely hair twirling, finger sucking, and nail biting, oc cu rred most frequently (Deaver et al. 2001; Dufrene, Watson, & Kazmerski, 2008; Stricker, Miltenberger, Anderson, Tulloch, & Deaver, 2002) R esults of these studies showed that repetitive behavior was

PAGE 13

13 maintained by automatic reinforcement (i.e., occurred i n the absence of socially mediated reinforcement) In one study that attempted to assess the function of repetitive behavior, Woods and Miltenberger (1996) exposed 44 undergraduates to three conditions designed to have a boring, anxiety provoking, or neut ral effect. Results of the assessment indicated that hair and face manipulation occurred most often in the anxiety condition and object manipulation occurred most often in the bored condition. Based on these results, Woods and Miltenberger concluded that repetitive behavior may serve an anxiety reducing or self stimulatory function (i.e. automatic negative and automatic positive reinforcement respectively ). Stereotypy Is Repetitive Behavior In a review on stereotypy, Rapp and Vollmer (2005 a ) defined ste reotypy as behavior invariant or repetiti ve in movement that persists in the absence of social consequences. Stereotypy occurs in a wide variety of topographies, including countless variations of body postures, vocalizations, and head, arm, hand, leg, or face movements (Berkson and Davenport, 1962). Stereotypy may begin as behavior that is part of typical motor development; typically developing infants engage in stereotypic behavior that gradually decreases in frequency or changes in topography ( Berkson & Tupa, 2000 ; Lovaas, Newsom, & Hickman, 1987). Stereotypy in individuals with intellectual disabilities (ID) may follow a similar course of development and fail to decrease due to a lack of a development of alternative behavior or of alternative sources o f reinforcement (Lovaas et al.) In a seminal study, Berkson and Davenport (1962) found a significant negative correlation between the incidence of stereotypy and IQ, suggesting that higher levels of stereotypy occur with lower functioning individuals. Bo dfish, Symons, Parker, and Lewis (2000) observed that individuals with autism exhibit ed more stereotypy than

PAGE 14

14 age and functioning matched peers with mental retardation without autism. One study on the longevity of stereotypy showed that it was maintained at moderate levels in a sample of 7 individuals with ID at a ten year follow up (Jones, 1999). In a preliminary investigation of stereotypy in 1962, Berkson and Davenport observed that stereotyped movements and postures occurred in two thirds of their samp le of 71 individuals with ID In a similar study, Kaufman and Levitt (1965) discovered that 57 69% of 83 individuals with ID engaged in at least one of the three topographies hypothesized to be the most common forms of stereotypy: body rocking, head rolli ng, and hand to eye stereotypy. More recently, Bodfish et al. (2000) found that 100% of individuals with autism and over 80% of individuals with mental retardation without autism engaged in stereotypic behavior. The Function of Stereotypy Rapp and Voll mer (2005 a) recently reviewed the evidence that stereotypy is maintained by automatic reinforcement. They list five sources of support gleaned from empirical research, for th e automatic reinforcement hypothesis: stereotypy persists in the absence of soci al consequences, competing sources of stimulation sometimes reduce stereotypy, blocking the p utative sensory consequences of stereotypy leads to its reduction, contingent access to stereotypy can reinforce other behavior, and restricting access to stereoty py changes the subsequent level of stereotypy. This evidence suggests that the behavior is operant, rather than reflexive or tic like and that the reinforcement is not delivered by one person to another. Over the p ast 60 years profuse evidence has indica ted the role of automatic positive reinforcement in the maintenance of stereotypy. Berkson and Davenport (1962) found that higher levels of stereotypy occurred with visually impaired individuals,

PAGE 15

15 suggesting tha t stereotypy served a self stimulatory functi on. In 1963, the same researchers (Davenport and Berkson) observed that lower levels of stereotypy occurred when preferred items were manipulated by subject s, which suggests that the stimulation provided by the objects competed with the stimulation create d by the subject stereotypy. In a literature review, Lovaas, Newsom, and Hickman (1987) provided ample evidence that stereotypy is maintained by automatic positive reinforcement in the form of perceptual or sensory stimulation. Several researchers hav e suggested that stereotypy could also be maintained by automatic negative reinforcement. Kaufman and Levitt (1965) hypothesized that stereotypy may have served a tension reducing function because it occurred more often before times of meals or rest, whic h have been observed to be periods of high tension. Forehand and Baumeister (1971) sho w ed that frustration achieved by blocking previously reinforced responses increased stereotypy in individuals with ID ; thus stereotypy could have been tension or frustr ation reducing Lovaas et al. (1987) noted that studies on the arousal theory of stereotypy showed that increased levels of stereotypy occurred under conditions of aversive stimulation. Results from animal studies have shown that stereotypy may serve a s tress reducing function ( e.g. Bildsoe Heller, & Jeppesen, 1991) ; however, little research in this area has been conducted with ID humans (Rapp and Vollmer, 2005 b ). Given these various sources of evidence, it appears likely that stereotypic behavior in in dividuals with ID may at times occur in the presence of aversive stimulation and likely is maintained by automatic reinforcement even in that context.

PAGE 16

16 Stereotypy That Occurs During Demands Several researchers have conducted analyses of stereotypy that oc curs during demands. Durand and Carr (1987) conducted evaluations of stereotypic behavior using antecedent based assessments. Because stereotypy occurred most often in the condition with increased task difficulty, they concluded that stereotypy was reinfor ced by escape from demands. To further demonstrate, they conducted a reversal phase in which escape from demands was and was not provided contingent on stereotypic behavior; for two of four subjects, levels of stereotypic behavior were higher in the conti ngent timeout condition relative to the baseline condition. Thus the researchers in the form of escape from demands. However, two of the four subjects showed marginal differences between baseline and contingent timeout phases. Furthermore, the demand condition in Experiment 1 and the baseline condition in both Experiments 2 and 3 consisted of a procedure akin to escape extinction; the children were continually present ed with tasks and a break was never presented contingent upon stereotypy. Although baseline was functionally equivalent to escape extinction, all subjects engaged in stereotypy throughout baseline sessions: as many as 21 consecutive sessions for one subje escape maintained and instead seems likely that it was maintained by automatic reinforcement established as such by aversive stimulation (task demands). In a similar study, Mace Browder, and Lin (1987) evaluated the effect of demands on the stereotypic mouthing of one subject with ID. Because the highest levels of stereotypy occurred in the demand condition with the more difficult tasks (tasks that were novel or had more freque nt response requirements), they concluded that the

PAGE 17

17 However, several of the conditions in this study involved the continuous presentation of demands without the presentation of escape, and yet ing maintained for many sessions Because the behavior persisted in conditions akin to escape extinction, it seems unlikely that the behavior was escape maintained. Instead, it would appear that this study is another case in which stereotypy wa s maintain ed by automatic reinforcement in the presence of demands but wa s incorrect ly described as escape behavior Kennedy, Meyer, Knowles, and Shukla (2000) concluded that stereotypy was multiply controlled because it occurred throughout multiple conditions of a consequence based functional analysis. Because, for two subjects, differentially higher levels of stereotypy occurred in the attention, demand, and no attention conditions havior was maintained by attention, escape, and automatic reinforcement. However, this pattern of responding would typically be thought indicative of maintenance by automatic reinforcement. In addition, they concluded that the behavior of one subject who engaged in differentially higher levels in the demand and no attention conditions was maintained by escape and automatic reinforcement. Given that stereotypy had been shown by the elevated levels in the no attention condition to be sensitive to automatic reinforcement, it is possible that the stereotypy in the demand condition was maintained by automatic reinforcement established as such by the presence of the demands. Overall, the researchers do not provide convincing evidence that the stereotypy of thei r subject s was maintained by anything other than automatic reinforcement T he possibility of behavior maintained by automatic reinforcement under conditions of aversive stimulation also has implications for consequence based

PAGE 18

18 functional analyses in genera l. If demands are an establishing operation for automatically reinforced stereotypy, a researcher or clinician could conduct a functional analysis (Iwata, Dorsey, Slifer, Bauman & Richman, 1982/1994) and conclude that behavior was maintained by escape from demands. Iwata et al. (1994) reported that 3 of 46 applications of escape extinction to self injurious behavior (SIB) thought to be maintained by escape were ineffective. Although rare, t he possibility exists that escape extinction failed in this small number of cases because SIB was not escape maintained and was instead stereotypic SIB maintained by automatic reinforcement established as such by aversive stimulation (task demands). General Purpose The purpose of the current series of experiments was t wofold The first purpose was to determine whether in some cases automatically reinforced behavior is most likely to occur under demand like situations in typically functioning and intellectually disabled individuals The second purpose was to evaluate w hether behavior that occurs in the presence of demands is maintained by automatic reinforcement.

PAGE 19

19 CHAPTER 2 AN EVALUATION OF CON DITIONS THAT EVOKE R EPETITIVE BEHAVIOR I N TYPICALLY FUNCTIONIN G ADULTS Experiment 1: Purpose Wood and Miltenberger (1996) deve loped an antecedent based assessment to evaluate the function of repetitive behavior in typically functioning individuals; however, s ubjects were only exposed to each condition once, thus limiting knowledge about trends over time Thus, the purpose of Expe riment 1 was to systematically replicate Woods and Miltenberger (1996) using repeated measures and single subject designs. Of central interest wa s evaluating whether automatically reinforced behavior can sometimes occur at the highest rates during demand ing environmental situations. Intuitively and through casual observation this seems true; for example, sometimes people bite their nails while stuck in traffic or twirl their hair while taking an exam. If behavior occurs at high l evels during demand situa tions, it is likely that demands establish the automatic consequences of the behavior as reinforcement. Experiment 1: Method Subjects and Setting Subjects were 50 undergraduate students recruited through the subject pool at the University of Florida ; sub jects received course credit for participation Consent was obtained by having subjects sign a consent form to participate in a study on behavior during various tasks. At the end of the three hour session, subjects were debriefed on the true nature of the study and allowed to choose whether their data could be included in the study. No subject declined to have their data included All sessions were conducted in a research room at the University of Florida equipped with a one way mirror. The subjects sat in a computer chair at a table

PAGE 20

20 outfitted with a desktop computer. Depending on the condition, subjects were given access to the computer, magazines, or study materials (paper, pens, and note cards). Target Behavior and Data Collection The targets include d the following topographies of repetitive behavior: hair manipulation, self scratching, tapping limb movements, object manipulation, nail biting, object mouthing, lip biting, chair swiveling, self focused vocalizations, and knuckle cracking. An operatio nal definition was written for each topography D efinitions are presented in Tab le 2 1. Topographies were similar to those used by Woods et al (1996). However, some topographies were combined to reduce the number of classes scored; ( e.g. touching face, to uching lips, and scratching head were collapsed into one class called self scratching that included contact between the hand and any body part ) Data were collected by one or more researchers who observed the subject through the one way mirror contin uously for each 5 minute session. S ubjects were blind to the true nature of the study. Subjects were told that they would be participating in a study to see how people behave when asked to perform various types of tasks T hey w ere told they would be wat ched through the one way mirror to make sure they were working but were not told what behaviors were being observed Using handheld computers, observers collected data on the duration of each topography of repetitive behavior and data were later analyzed as the percentage of session in which repetitive behavior occurred. Each topography was graphed individually and all topographies were combined in one total measure (i.e., the percent of session in which any form of repetitive behavior occurred). Two obs ervers independently collected data for 61% of sessions ( range 29% 76% across subjects). We divided each 5 min observation into 10 s intervals and the

PAGE 21

21 percent of agreement for each interval was calculated by dividing the smaller number of seconds of a beh avior scored by the larger number scored and then multiplying by 100. An overall agreement score was then calculated by averaging the agreement scores for used in beha vior analysis research. Mean agreement was 98% (range, 81% 100%) for hair manipulation, 95% (range, 69% 100%) for self scratching, 99% (range, 81% 100%) for tapping 96% (range, 55% 100%) for limb movements, 99% (range, 67% 100%) for object manipulation, 98% (range, 72% 100%) for nail biting, 99.9% (range, 94% 100%) for object mouthing, 99.8% (range, 75% 100%) for lip biting, 98% (range, 56% 100%) for chair swiveling, 99.6% (range, 79% 100%) for self focused vocalizations, 99.7% (range, 87% 100%) for knuck le cracking. Because the observers collected live data on a subject with whom they had no history, the first five sessions of interobserver agreement (IOA) were calibration sessions. During these sessions, the observers discussed the behavior of the sub ject and determined which behavior to score as which topograph y Thus, these sessions were not included in the calculation of either IOA scores or the proportion of sessions with IOA Procedure Antecedent based assessment Each subject was exposed to seve ral conditions of an antecedent based assessment in a multielement design. Antecedent based assessments were used because we were only interested in identifying repetitive behavior that occurred independently of social reinforcement (i.e., was automatical ly reinforced). Due to limited time with the subjects (3 hours of total session time), each subject was exposed to between 15 and 21 five minute sessions. Each subject was

PAGE 22

22 randomly selected to participate in one of the demand conditions; thus sessions q uasi randomly alternated between Alone, Free, and either IQ, Video Game, or Presentation (the latter three were considered possible demand conditions) Conditions were randomly chosen within a group of three sessions; no condition was repeated until each h ad been run once in a group. In each of the demand conditions, to enhance motivation to do well, subjects were told fictional high scores disguised as the scores of the previous five subjects. Programmed consequences did not occur contingent upon repetitiv e behavior in any condition. The conditions of the antecedent based assessment were: (1) Alone In this condition, the subject was left alone in the room with no materials other than those typically present in an office. The subject was told to wait for the researcher to return. This condition was designed to test if the subject would engage in repetitive behavior when given nothing to do, which would suggest that repetitive behavior was maintained by automatic reinforcement. (2) Free In this condition the subject was alone in the room and was given access to the internet and magazines; this condition was used to identify the level of repetitive behavior when the subject had preferred materials and was not asked to do anything. ( 3) Demand Each of the following demand conditions was intended to test the level of repetitive behavior exhibited during three potentially stressful tasks; high levels of repetitive behavior in these conditions could suggest that demand situations establish the stimulus product s of repetitive behavior as reinforcement. The three demand conditions were: ( 1 ) IQ. In this condition, the subject was instructed to complete difficult intelligence test type questions ; they were told the items were from an IQ test but were not told that their IQ was being measured The items

PAGE 23

23 used were visual testing items from IQ Tests to Keep You Sharp (Carter & Russell, 2002); the items involved visual tasks such as pattern recognition. The items were chosen to be difficult but not so difficult that su bjects might just give up. For 43% of the subjects who were exposed to this condition, an experimenter was in the room, but did not interact with the subject other than to hand them intelligence testing questions. (2 ) Presentation In this condition, the subject was alone in the room and asked to read a research article and prepare a summary and opinion statement. The subject was told that they would be expected to give an oral presentation to a group of graduate students at the end of the 3 hour block. (3 ) Video game In this condition, the subject was alone in the room and played a video game that was fast paced and chosen to be challenging in five minute segments. The video game used was Collapse which is a game that is available for free from a nu mber of websites. The game involved clicking on same colored blocks to clear them before the blocks reached the top of the screen. The first six subjects were exposed to all five of the experimental conditions: alone, free, IQ test, presentation, and vi de o game. Because of the limited time with each subject, we were not able to conduct a sufficient number of sessions in each condition. Hence, we chose to randomly assign future subjects to one of the demand conditions and did not include subjects 1 6 in data analyses. Data for each subject were classified via visual analysis as one of five outcomes: differentially higher in alone, differentially higher in demand, differentially higher in alone and demand but highest in alone, differentially higher in a lone and demand but highest in demand, and undifferentiated. For each data set, alone and demand conditions were visually compared to the free condition to determine if the former data paths were differentiated from the latter. In the case that data path s visually overlapped, individual

PAGE 24

24 pairs of points were examined and more than 70% of the data points for either alone or demand must have been higher than the adjacent free data points to be called differentiated. For example, data for UG45 are presented in Figure 2 1; the complete data set is presented in the top panel followed by the comparisons of alone versus free and demand versus free in the middle and bottom panels, respectively. Initial visual inspection shows that the alone data path was consist ently higher than the free data path. This can be verified by counting the number of alone data points higher than the adjacent free data points in the second panel; alone was higher in six of the seven pairs of data points (86%). To the contrary, althou gh the demand data path may appear higher upon visual inspection, analyzing the data points in the bottom panel shows that demand was higher in only four of seven pairs (57%). Hence, the demand data path was not considered differentiated and the data set was classified as differentially higher in alone. Experiment 1: Results and Discussion All subjects engaged in at least one topography of repetitive behavior Thirty three subjects ( 66 %) engaged in differentially high er levels of repetitive behavior in on ly the alone condition of the assessment (see Figures 2 2 through 2 9 for data from these subjects ). Nine subjects (18%) engaged in differentiated levels of repetitive behavior in both the alone and demand conditions relative to the free condition, but en gaged in highest levels in the alone conditio n ( s ee Figure 2 10 and 2 11 for data). Two subjects (4%) engaged in the highest levels of repetitive behavior in the demand condition; however, levels in the alone condition were also differentiated with respe ct to th e free condition (s ee the top two panels of Figure 2 12 ). One subject (2%) engaged in differentiated levels of repetitive behavior in the demand condition only ( s ee the bottom

PAGE 25

25 panel of Figure 2 12). Five of the fifty subjects (1 0%) showed undiffe rentiated results ( s ee Figure 2. 13 ). Table 2 2 displays the average percent of intervals with repetitive behavior for all subjects across conditions and topographies. The numbers in parentheses below the conditions represent the number of people who pa rticipated in each condition. Because knuckle cracking was added as a topography later in the study, the number of sessions with data for knuckle cracking is displayed (n=31). Results show that on average, repetitive behavior occurred most often in the a lone and IQ conditions, 43.29% and 21.03% respectively. In addition, self scratching, limb movements, and nail biting occurred most often overall T hese topographies occurred most frequently in the alone condition, with the exception of limb movements, wh ich occurred most often in the alone and IQ test conditions. In the vast majority of cases, repetitive behavior was most likely to occur at differentiated levels during the alone condition. Nonetheless, in some cases the behavior persisted at different ially higher levels during the demand condition. Repetitive behavior that occurred at differentially higher levels in a demand condition was presumably maintained by automatic reinforcement established as such by aversive stimulation. This function of re petitive behavior ha s implications for assessment and treatment for children who engage in repetitive or stereotypic behavior In previous research, antecedent based assessments have been conducted on the stereotypic behavior (repetitive behavior with no ap parent social function) of ID children. As mentioned previously Durand and Carr (1987) conducted a n assessment of stereotypic behavior in which tasks were presented to the children as an antecedent manipulation. Because of the presence of the tasks, the researchers concluded that the

PAGE 26

26 stereotypic behavior of these subjects was reinforced by social negative reinforcement. Given results of Experiment 1 and Woods and Miltenberger (1996), it seems possible that children with intellectual disabilities might e ngage in repetitive behavior that is maintained by automatic reinforcement during task presentation. To test this possibility, we conducted Experiment 2.

PAGE 27

27 Table 2 1. Operational definitions for Experiment 1 Behavior Operational Definition Hair Man ipulation Touching the hair with one or both hands while the hand is moving Self scratching Pressing the nail of one or two hands on the surface of the skin with indentation or gathering of the skin Tapping Repetitive (greater than once per second) hitti ng of one object against another object or body part Limb Movements Moving a limb repeatedly in a recurring and rhythmic fashion, can be judged by leg shake or body bounce/shake Object Manipulation Repeatedly moving an object in a similar manner, e.g., r olling a pencil on a desk, does not include session material manipulation or just holding an item Nail Biting Placing any part of the end of the fingers between the lips and teeth or picking at the nail with the fingers of another hand Object Mouthing Pl Lip Biting Closing the teeth around any part of the lip Chair Swiveling Moving the chair in one direction and back repeatedly Self Focused Vocalizations Making any vocal statements or sounds Knuckle Cracking Cracking any joints or pushing on the knuckle with the other hand, another body part, or an object

PAGE 28

28 Table 2 2 Summary data for Experiment 1 by condition Averages are displayed for percent of session with repetitive behavior for any topography (total) and then each individual topography across conditions. The numbers next to each condition label represent the number of subjects who participated in each condition. For knuckle cracking, which was added after the 19 th subject data are shown only for the 31 subjects for whom data on knuckle cracking were collected. Condition/ Topography Total Hair manip Self scratch Finger tap Limb move Object manip Nail bite Object mouthing Lip bite Chair spin Self vocal Knuckle crack (n=31) Alone (n=50) 43.29 3.83 5.60 5.13 8.54 5.25 11.15 0.02 0.10 7.01 0.68 0.45 Free (n=50) 16.06 0.97 4.35 0.41 4.81 1.38 2.75 0.00 0.21 1.39 0.34 0.23 IQ test (n=28) 21.03 1.38 3.23 0.72 8.73 2.56 0.95 0.00 0.17 2.08 2.36 0.18 Presentation (n=10) 12.86 1 .74 4.41 0.59 1.49 1.05 2.92 0.00 0.05 0.74 0.11 0.36 Video game (n=12) 7.45 0.30 1.29 0.02 4.96 0.00 0.33 0.01 0.49 0.10 0.06 0.00 Average 20.14 1.64 3.78 1.37 5.71 2.05 3.62 0.01 0.21 2.26 0.71 0.25

PAGE 29

29 Figure 2 1. Per cent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (black circles) conditions for UG 45 These data are presented to demonstrate the data analysis process (see the text)

PAGE 30

30 Figure 2 2 Percent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (black circles) conditions for UG07, UG09 UG10, and UG11 These four subjects engaged in differentially higher levels of repetitive behavior in the alone condition.

PAGE 31

31 Figure 2 3 Percent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (black circles) conditions for UG12, UG 13, UG14, and UG18 These four subjects engaged in differentially higher levels of repetitive behavior in the alone condition.

PAGE 32

32 Figure 2 4 Percent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (blac k circles) conditions for UG20, UG 21, UG22, and UG23 These four subjects engaged in differentially higher levels of repetitive behavior in the alone condition.

PAGE 33

33 Figure 2 5 Percent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (black circles) conditions for UG25, UG 26, UG27, and UG29 These four subjects engaged in differentially higher levels of repetitive behavior in the alone condition.

PAGE 34

34 Figure 2 6 Percent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (black circles) conditions for UG 32 UG 33 UG 34 and UG 35 These four subjects engaged in differentially higher levels of repetitive behavior in the alone condition.

PAGE 35

35 Figure 2 7 Percent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (black circles) conditions for UG 36 UG 38 UG 39 and UG 40 These fou r subjects engaged in differentially higher levels of repetitive behavior in the alone condition.

PAGE 36

36 Figure 2 8 Percent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (bla ck circles) conditions for UG 43 UG 44 UG 45 and UG 47 These four subjects engaged in differentially higher levels of repetitive behavior in the alone condition.

PAGE 37

37 Figure 2 9 Percent of session with repetitive behavior acro ss alone (white squares), free (white triangles), and demand (black circles) conditions for UG51 UG 52 UG 54 UG 55, and UG56 These five subjects engaged in differentially higher levels of repetitive behavior in the alone condition.

PAGE 38

38 Figure 2 10 Percent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (bl ack circles) conditions for UG08, UG15, UG16, UG19, and UG28 These five subjects engaged in differentially higher level s of repetitive behavior in the alone and demand conditions with highest levels in alone

PAGE 39

39 Figure 2 11 Percent of session with repetitive behavior across alone (white squares), free (white triangles), and demand (bl ack cir cles) conditions for UG46, UG48, UG50, and UG53 These four subjects engaged in differentially higher levels of repetitive behavior in the alone and demand conditions with highest levels in alone

PAGE 40

40 Figure 2 12 Percent of s ession with repetitive behavior across alone (white squares), free (white triangles), and demand (bl ack circles) conditions for UG 24 UG4 2 and UG 17 The first two subjects engaged in differentially higher levels of repetitive behavior in the alone and dem and conditions with highest levels in demand UG17 engaged in differentially higher levels of repetitive behavior in the demand condition only.

PAGE 41

41 Figure 2 13 Percent of session with repetitive behavior across alone (whi te squares), free (white triangles), and presentation (black circles) conditions for UG 30, UG31, UG37, UG41, and UG49 These five subjects displayed results that were undifferentiated.

PAGE 42

42 CHAPTER 3 AN EVALUATION OF THE CONDITIONS THAT EVOK E STEREOTYPY IN INT ELLECTUALLY DISABLED CHILDREN Experiment 2: Purpose Results of Experiment 1 showed that 20% of subjects engaged in differentially higher levels of repetitive behavior when presented with demands in the absence of social contingencies. The purpose of Exper iment 2 was to evaluate whether elevated levels of stereotypy occur in a high demand but no escape condition with ID children This finding would have implications for behavioral assessment in ID because prior antecedent based assessments of stereotypy m ay have falsely inferred a social function of the behavior (e.g., Durand & Carr, 1987). Experiment 2: Methods Subjects and Settings Subjects were fifteen individuals ages 5 20, who were diagnosed with autism or another intellectual disability and engag ed in at least one topography of stereotypic behavior. Subjects were recruited from local schools. Informed consent was obtained form that included the detailed purpose and procedure of the study. Assent was obtained from the subject if the functioning level of the subject was appropriately high. Demographic information for each subject is presented in Table 3 1. from their typical classrooms. In all but one case, the room was devoid of distracting stimuli. For Tobias, a general purpose room in his school that contained toys and other materials was used and a researcher sat near Tobias to ensure he did not get up from the research table or interact with items outside of the experiment.

PAGE 43

43 Target Behavior and Data Collection Operational definitions were constructed for any apparently stereotypic behavior; naturalistic observations were conducted with each subject until definitions wer e written for each topography. A list of all topographies of stereotypy for each subject is included in Table 3 2. D efinitions for each topography are available from the author. Data were collected by one or more researchers who observed the subject f rom an unobtrusive location in the room A operational definitions and collected at least one session of training data with each subject before collecting IOA Similar to Experiment 1, observers used handheld c omputers to collect data on the duration of each topography of behavior and data were summarized as a percentage of session with behavior ; total repetitive behavior and individual topographies were analyzed Two observers ind ependently collected data for 49 % of sessions (range 20% 67% across subjects). P ercent agreement scores were calculated using the proportional method as described in Experiment 1. Average percent agreement across all subjects and topographies wa s 90.60% (range 86 % 96% across subjec ts). Procedure Preference assessment Free operant preference assessments (Roane, Vollmer, Ringdahl, & Marcus, 1998) were used to identify items to use during the free condition of the antecedent preferred item was identifi ed in a preference assessment previously conducted for an earlier study. Items were chosen for inclusion in the preference assessment based on the recommendations of ect was prompted to

PAGE 44

44 interact with each item prior to the start of the assessment. During the sessions, the collected on engagement with an item, defined as physical contact w ith or orientation to the item. One to five 5 min sessions were conducted per subject. One 20 min session was conducted for Dante after he failed to select any item in two prior 5 min sessions. For Jasmine, who is blind and therefore unable to see the d ifferent items in an array, after one 5 min preference assessment session in which all items were presented, we conducted eight 1 min single operant sessions (Pace, Ivancic, Edwards, Iwata, & Page, 1985) to determine the extent that Jasmine would engage wi th a single item. Data were analyzed as a percentage of session in which the subject engaged with an item and percentages were averaged across sessions. Two observers independently scored engagement for 86% of all the preference assessment sessions (rang e, 50 100% across subjects). The average percent of agreement across items and subjects was 98% (range, 91% 100% across subjects) Antecedent based assessment Similar to Experiment 1, each subject was exposed to several conditions of an antecedent based a ssessment in a multielement design. The five minute sessions alternated between ignore, free, and demand conditions Condition s were randomly chosen within a group of three sessions; no condition was repeated until eac h had been run once in a group. No p rogrammed consequences occurred contingent upon stereotypic behavior in any condition. Ignore In this condition, the subject was left in the room without materials; the data collectors were present in the room, but they did not interact with the subject This condition tested whether the subject would engage in stereotypic behavior when given

PAGE 45

45 nothing to do, which would suggest that stereotypic behavior was maintained by automatic reinforcement. For Tobias, a researcher was presen t at the experimental ta ble to prompt him to sit back down if he attempted to leave the research area ; Tobias never attempted this Free This condition was identical to the ignore, except that the subject was given access to preferred materials The top four items from the pre ference assessment were included unless the subject engaged with the item less than 5% of the session during the preference assessment (for exceptions see below) T his condition was used to identify the level of repetitive behavior when the subject had pr eferred materials and was not asked to do anything. For Tobias, a researcher was present at the experimental table and provided attention to the subject if he engaged in appropriate attention seeking behavior or indicated he needed help. Demand In this condition, the researcher sat next to the subjects and prompted them to complete a task. Demands were presented continuously and there was no programmed inter trial interval; the only break in demands occurred when the researcher reset the demand, which l asted no longer than a few seconds. Tasks were chosen for each subject by asking their teachers to identify things they could do but did not like to do. Preliminary assessments were done to identify a task the subject did not complete correctly more than half of the time but for which the subject possessed prerequisite skills (e.g., grasping is a prerequisite for puzzle completion). The demand condition tested whether the subject would engage in repetitive behavior when presented with demands.

PAGE 46

46 Experime nt 2: Results and Discussion Table 3 3 displays r esults of the preference assessment as the average percent of sessions with engagement, with the exception of Jasmine, for whom percent of sessions during the 1 min single operant sessions is displayed. Ite ms chosen for inclusion in the antecedent based assessment are denoted by an asterisk. The top four items were selected for inclusion if they were engag ed with at least 5% of the session Three exceptions to this rule were made. Ethel was only provided wi th three items due to a data analysis error that underestimated the percent of engagement with the keyboard. Jasmine was given 1 extra item (the book) because she was blind and frequently pushed items out of reach. In addition, b ecause Jasmine had diffic ulty on her own and primarily used it to engage in stereotypic behavior, we chose to use the book in the subsequent assessment rather than t he See The final exception to the rule was made for Heidi. Heidi engaged with dolls, a car, a football, and a puppet most during the preference assessment. However, it was later observed that Heidi often tossed the items onto the floor seemingly b ecause of the sound the item made when it fell (based on her reaction) Thus, items w ere swapped out that could be thrown but not broken and that would make a noise when they landed; chosen. Despite this change, engagement remained high in the free sessions of the following asses sment. Results of the antecedent based assessment showed that three of the 15 subjects (20%) ha d undifferentiated results. Three subjects (2 0 %) engaged in differentially high er levels of repetitive behavior in only the ignore condition of the assessment Four subjects (2 6.67 %) engaged in differentially high er levels of repetitive behavior in both

PAGE 47

47 the ignore and free conditions relative to the demand condition Four subjects (26.67 %) engaged in differentiated levels of repetitive behavior in the ignore a nd demand condition s ; however, levels in the ignore condition were elevated above the demand condition for all four One subject (6.67%) engaged in the differentially high levels of repetitive behavior in the demand condition. Results of the antecedent b ased assessment for Gertrude, Jesus, and Pablo are displayed in Figure 3 1; these three subject s engaged in undifferentiated levels of stereotypy All three engaged in high levels of repetitive behavior across all conditions of the assessment, indicat ing t hat th e i r repetitive behavior was maintained by automatic reinforcement. This pattern was generally consistent across topographies; however, Pablo engaged in differentially more squinting in the demand condition of the assessment B ecause it was suspected that squinting might actually represent trying to see the task and not stereotypy, this topography was not further evaluated Results of the antecedent based assessment for Kandi Mario, and Tobias are displayed in Figure 3 2. All of the subject s displaye d the differentially higher levels of stereotypy in the ignore condition, suggesting that the behavior was maintained by automatic reinforcement established as such by the relatively barren environment of the ignore condition. Kandi (upper panel) engaged in the highest level of repetitive behavior in the ignore condition and moderately high levels in the other two conditions. Similar results were obtained for Mario ( middle panel). Tobias ( bottom panel) engaged in low levels of stereotypy overall, but diff erentially higher levels in the ignore condition. Results of the antecedent based assessment for Chad Jasmine, Ethel and Dante are displayed in Figure 3 3. These subjects all engaged in differentially higher levels of repetitive behavior in the ignore and free conditions relative to the demand

PAGE 48

48 condition Chad Jasmine, and Ethel engaged in the highest levels of repetitive behavior in the ignore condition, and levels in the free condition were only slightly lower, suggesting that the items available in the free condition did not adequately compete with stereotypy. Dante engaged in the highest levels in the free condition, suggesting that the items provided in the free condition may have somehow stereotypy; anecdotally it was observ ed that Dante engaged in nearly constant hand stereotypy with one hand while holding the bells in his other hand. F or all three subjects, the tasks presented in the demand condition appeared to compete with stereotypy given that stereotypy levels were lowest in this condition. Results of the anteceden t based assessment for Martin, Agatha Ben and Heidi are displayed in Figure 3 4; These subject s engaged in the differentiated level s of stereotypy in the ignore and demand conditions of the assessment, s uggesting that their behavior was maintained by automatic reinforcement established as such by both low levels of stimulati on and by demands (or that demands competed with stereotypy to a degree) Further evaluation of individual topographies showed that Ben engaged in differentially higher levels of vocalizations in the demand condition (see Figure 3 5) suggesting that this topography was uniquely sensitive to the presence of demands and stereotypy was maintained by automatic reinforcement established a s such by demands ; Ben was later included in Experiment 3 to further evaluate the demand conditions under which vocalizations would occur Heidi also displayed differentially higher levels of hand stereotypy (see Figure 3 5) and foot kicking the in demand condition; however, because Heidi wandered around the room during both the free and alone conditions, foot kicking could not technically occur during these sessions. Thus, hand stereotypy alone was chosen for later inclusion in Experiment 3 (but see

PAGE 49

49 expla nation in Chapter 4 of a subsequent disqualification from Experiment 3) Hand stereotypy was apparently maintained by automatic reinforcement established as such by demands. Results from the antecedent assessment for C lyde are displayed in Figure 3 6 Dur ing the antecedent based assessment, Clyde engaged in the differentially highe r levels of stereotypic behavior in the demand condition suggesting that his behavior was maintained by automatic reinforcement established as such by demands ; this was not only true of the total measure of stereotypy but also individual topographies including high pitched vocalizations, other vocalizations, and head hitting Due to these results, Clyde was chosen for later inclusion in Experiment 3. Results of this experime nt showed that 33% of individuals engage d in differentially high er levels of stereotypy in the demand condition of an antecedent based assessment, as compared to the free condition. Given that no consequences were manipulated, it seems likely that these s ubject reinforcement established as such by the presence of demands. Results of this study are important because they indicate a limitation of previous research. Researchers have concluded that stereotypy that occurs in the presence of demands is escape maintained (Durand & Carr, 1987; Kennedy et al., 2000 ; Mace et al, 1987 ); however, results of this study imply that stereotypy that occurs in the presence of demands and the absence of social consequences is more likel y maintained by automatic reinforcement. Experiment 3 was conducted to further evaluate the possible role of escape and automatic reinforcement. Because the presence of demands produced the highest levels of at least one form of stereotypy for three ind ividuals, these subjects were initially selected to participate in Experiment 3

PAGE 50

50 Table 3 1. Demographic information for each subject: Name, age, gender, diagnosis, medications, and task used in the antecedent based assessment are displayed. (ASD=Autism S pectrum Disorder, ID=Intellectually Disabled, S/L=Speech/Language, ADHD= Attention Deficit Hyperactivity Disorder, PDD= Pervasive Developmental Disorder ) Name Age Gender Diagnoses Medications Task Martin 16 male Cerebral Palsy, ASD, educable ID, S/L impai rment, and other physical i mpairments Abilify and Selexa Receptively identifying words Gertrude 18 female ID, ASD, and language i mpairment None Completing wood puzzle Ben 12 male Dandy Walker Syndrome, other health impairment, p hysical impairment, ADHD I ntinuv and Straterra Pointing to numbers Kandi 12 female Trisomy 9 Sy ndrome, PDD, severe ID and S/L i mpairment Zoloft Completing wood puzzle Jesus 10 male ASD, S/L impairment, visual i mpairment None Putting block s in bucket Mario 8 male ASD, profound I D, o rt hopedic i mpairment, S/L impairment, Intractable Epilepsy, Congenital Ptosis, Porencephaly, Pachygyria Topomax and Keppra Taking out puzzle piece Heidi 14 female ID, Cerebral palsy, s peech delay, legally blind, seizures asthma Cardiac arrhythmia La mictal, Preva c i d, Melatonin, and Trazadone Putting blocks in a shape sorter Agatha 15 female ID, S/L impairment legally blind, Microcephaly, h istory of seizures Carditrol, Colopine, and Miralax Putting blocks in bucket Pablo 20 male Profound ID, S/L im pairment, p hysical (orthopedic) impairment None Putting blocks in a shape sorter Jasmine 19 female Trainable ID, H ypopituitarism, visual impairment, S/L Impairment and physical disability Cortef, Solucorteff, Synthyroid, and Iron Folding a towel Dante 18 male ASD and l anguage impairment Clonodine and Abilify Pointing to 3, 5, 10 on a number line Chad 17 male Trainable ID l anguage impairment, ASD None Completing 2 digit math problems Clyde 19 male ID, ASD, and S/L i mpairment Clonodine Walking Ethel 15 female ID and S/L Impairment Levothyroxine and Methimdoze Completing a soft puzzle Tobias 6 male ID, S/L Impairment heart defect, seizures Diastat PRN Writing letters

PAGE 51

51 Table 3 2. For each subject, topographies of stereotypy are listed Subject Topograp hies Martin Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Limb Movements, Head Tics, Body Hitting, Shoe Play, Flapping, Object Manipulations, Finger Picking Gertrude Hand Stereotypy, Body Rocking, Sneering, Vocalizations, Foot Kicking, Gross Motor Movements, Tapping Ben Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Foot Kicking, Hand to Neck Stereotypy, Shirt Mouthing Kandi Hand Stereotypy, Body Rocking, Scratching, Soft Vocalizations, Pushing, Clicking, Loud Vocalizations, Mouth ing Jesus Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Foot Kicking, Mouth Stereotypy, Jumping, Tapping, Ear Covering, Arm Flapping Mario Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Wall Kicking, Mouthing, Cheek Sucking, Chin P ressing, Yelling/whining Heidi Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Foot Kicking, Stomping, Tossing, Chair Shaking Agatha Hand Stereotypy, Body Rocking, Face Rubbing, Vocalizations, Foot Kicking, Mouthing, Ear Covering, Towel Playin g, Teeth Grinding, Arm Posturing Pablo Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Foot Kicking, Mouth Stereotypy, Squinting, Inappropriate Sexual Behavior Jasmine Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Foot Kicking, Mou thing, Arm Swinging, Bouncing, Head Rolling, Ear Covering, Swearing Dante Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Foot Kicking, Arm Swinging, Bouncing Chad Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Foot Kicking, Biting, Yelling, Head Rolling, Jumping/bouncing, Disrupting Clyde Head Hitting, Banging, Skin Picking, High pitched Vocalizations, Low pitched Vocalizations, Chomping, Stomping Ethel Hand Stereotypy, Banging/throwing, Scratching, Vocalizations, Leg Flapping, Mou th Stereotypy, Yelling, Hair Playing, Huffing Tobias Hand Stereotypy, Body Rocking, Scratching, Vocalizations, Foot Kicking, Mouthing, Other Stereotypy

PAGE 52

52 Table 3 3. Results of the free operant preference assessment are shown. Numbers under the subje represent the average percent of session the individual interacted with each item, unless otherwise noted in the text Asterisks denote those items chosen for inclusion in free sessions of the antec edent based assessment. Martin (n=1) Gertrude (n=4) Ben (n=3) Kandi (n 2) Puzzle 36 Keyboard 50 34 Rain stick 100 Mr. Potato Head 27 Mirror 48 Mr. Potato Head 24 Ambulance 0 1 Lotion 18 Book 23 Keys 0 Keyboard 1 Doll s 1 Keyboard 8 5 Book 0 Book 0 Puzzle 0 Cars 2 0 0 Magna 0 Ambulance 0 phone 0 0 Bear 0 Puzzle 0 Plush 0 Rain stick 0 Polly 0 Coloring 0 Keyboard 0 Jesus (n=2) Mario (n=2) Heidi (n=2) Agatha (n=3) Bells 46 Ambulance 90 Dolls 17.3 Bells 33 Ribbon 24 phone 14 Car 16.65 Farmer 28 Keys 12 Rain stick 1 Football 10.35 Rain stick 20 Mirror 11 Mr. Potato 0 Puppet 10.15 See n' 4 Rain stick 3 Football 0 Bear 7.7 Arrow 0 0 Bob the 0 Ribbon 1 Keys 0 Antlers 0 Racecar 0 Bob the 0 Piano 0 0 Keyboard 0 Nubby ball 0 Kaleidoscope 0

PAGE 53

53 Table 3 3. C ontinued. Pablo (n=2) Jasmine (n=1) Dante (n=5) Chad (n= 3) Keyboard 92 Bob the 100 Bells 40 51 1 phone 100 Keyboard 2 51 Rain stick 1 Microphone 100 Cardboard 1 30 Gator 1 Rain stick 100 Drum 0 Keyboard 23 Bob the 0 See 100 Gator 0 Puzzle 0 Radio 0 Book 96.7 Puzzle 0 Book 0 Bells 0 Piano 96.7 0 Truck 0 phone 0 Bells 86.7 Thomas and 0 0 Ethel (n=2) Tobias (n=4) 26 Ambulance 52 Little Puzzle 25 Coloring 25 Book 12 15 Keyboard 9 Magna 4 Rain stick 9 2 Bells 7 Book 0 phone 2 Puzzle 0 Big Puzzle 0 Rain stick 0

PAGE 54

54 Figur e 3 1. Percent of session with repetitive behavior across ignore (white squares), free (white triangles), and deman d (black circles) conditions is displayed for Gertrude, Jesus, and Pablo. These subjects engaged in high levels of repetitive behavior acros s all conditions of the antecedent based assessment.

PAGE 55

55 Figure 3 2. Percent of session with repetitive behavior across ignore (white squares), free (white triangles), and demand (black circles) con ditions is displayed for Kandi, Mario, and Tobias, w ho engaged in the differentially highe r levels of repetitive behavior i n the ignore condition.

PAGE 56

56 Figure 3 3. Percent of session with repetitive behavior across ignore (white squares), free (white triangles), and demand (black circles) conditi ons is displayed for Chad, Jasmine, Ethel, and Dante, who engaged in differentially higher levels of repetitive behavior in the ignore and free conditions.

PAGE 57

57 Figure 3 4. Percent of session with repetitive behavior is displayed across ignore (white squares), free (white triangles), and demand (black circles) conditions for Martin, Agatha, Ben, and Heidi who engaged in differentiated levels of repetitive behavior during the ignore and demand conditions of the antecedent based ass essment.

PAGE 58

58 Figure 3 5 Percent of session with vocalizations (Ben) and hand stereotypy (Heidi) is displayed across ignore (white squares), free (white triangles), and demand (black circles) ; these topographies occurred at the highest levels in the demand condition.

PAGE 59

59 Figure 3 6 Percent of session with repetitive behavior is displayed across ignore (white squares), free (white triangles), and demand (b lack circles) conditions for Clyde. Clyde mo st often engaged in repetitive behavior during the demand condition of the antecedent based assessment.

PAGE 60

60 CHAPTER 4 AN EVALUATION OF REP ETITIVE BEHAVIOR THA T OCCURS IN THE DEMA ND CONDITION OF AN ANTE CEDENT BASED ASSESSMENT Experiment 3: Purpose Results of E xperiments 1 and 2 showed that differentially higher levels of repetitive behavior occurred in the presence of demands for both typically functioning adults and intellectually disabled children, respectively. Because repetitive behavior occurred in the ab sence of social contingencies, automatic reinforcement was implicated as the maintaining reinforcer. The purpose of Experiment 3 was to further evaluate whether stereotypy that occurred at differentiated levels in the demand condition of an antecedent base d assessment was maintained by escape or automatic reinforcement established as such by the presence of demands Experiment 3: Methods Subjects and Settings Subjects were three individuals who engaged in the highest levels of at least one topography of stereotypy in the demand condition of Experiment 2. Ben, Heidi, and Clyde participated ; demographic information for these subjects was provided in the previous chapter Consent for each subject was obtained prior to the previous study in the manner describ ed in Experiment 2. Sessions were conducted in a room in the rom their typical classrooms; t h is room was devoid of distracting stimuli. Target Behavior and Data Collection Operational definitions and data collection procedures w ere identical to those used in Experiment 2. Two observers independently collected data for 56% of sessions (range 53% 61 % across subjects). P ercent agreement scores were calculated using

PAGE 61

61 the proportional method as described in Experiment 1. Average perc ent agreement across all subjects and topographies was 92.29 % (range 87 % 97 % across subjects). Procedure Demand evaluation For those individuals who engaged in highe st levels of at least one stereotypic behavior in the demand condition of Experiment 2 relative to all other conditions, an additional analysis was conducted to further evaluate the function minutes and were conducted in a multi element design. Conditions w ere randomly chosen within a group of two sessions; no condition was repeated until each had been run once in a group Demand vs. f ree In the first phase of the demand evaluation, demand and free sessions were alternated. T hese sessions were identical to those conducted in Experiment 2. The purpose of this phase was to replicate the results from Experiment 2. E scape vs. free In the second phase of the demand evaluation, free sessions were alternated with escape sessions in which 20 sec breaks from dema nds were provided contingent on stereotypic behavior. Escape sessions were otherwise identical to the demand sessions of the previous experiment and phase. For Ben the contingency in the escape sessions was applied to hand stereotypy. For Clyde, escape v s. free w as conducted twice, once when vocalizations produced escape and once when head hitting produced escape Demand vs. enriched environment I n the final phase of the demand evaluation, the aversiveness of the demand was reduced via environmenta l enrichment In this phase, demand sessions that were identical to those previously conducted were

PAGE 62

62 alternated with enriched environment sessions to determine whether reducing the aversiveness of the environment would reduce stereotypy. Enriched environmen t sessions were identical to demand sessions except that access to preferred stimuli was made continuously available. Items chosen for inclusion in the enriched environment assessment were those identified as highly preferred by an assessment or natural observations. For Clyde, a preferred item and Clyde was observed to react positively to music in the classroom. used; the Magna music created by the keyboard, another preferred item. Silly putty was chosen to provide something Ben could hold in one hand while working; although Ben had not been observed with Silly Putty previously, he immediately approached the item when it was made available. Experiment 3: Results and Discussion Data are not shown for Heidi. Although Heidi engaged in differentially higher levels of stereotypy in t he demand condition of the first phase, she was not included in topography switched to foot kicking, which occurred at much higher levels during demand than during free. Ho wever, the sessions were confounded by the fact that she was seated during demands and usually not during free, so the foot kicking (although physically possible during free) was more likely in demand by mere virtue of the session structure. Unlike in ex periment 2, hand stereotypy now occurred at levels comparable to those in free, so Heidi did not participate in subsequent phases

PAGE 63

63 Results of the demand e valuation are displayed in Figure 4 1 ; results are shown for total repetitive behavior for Ben (top pan el) and for Clyde (bottom panel). In the first phase, b oth subjects engaged in differentially higher levels of repetitive behavior in the demand condition relative to the free condition; these results are true of the individual topographies as well (See F igure 4 2) These outcomes replicate the results of the previous experiment showing that stereotypy was differentially higher in the presence of stereotypy was maintained by automatic reinforcement in the presence of demands. The demand sessions involved the continuous presentation of demands, a procedure might see an initial extinction burs t followed by a gradual decrease in the level of behavior. To the contrary, the behavior of both subjects increased or maintained at relatively stable levels across the first phase. Figure 4 2 displays the results of the demand evaluation for the subject topographies ; hand stereotypy for Ben (top panel) and combine d vocalizations (middle panel) and head hitting (bottom panel) for Clyde are displayed Ben was originally chosen to participate in Experiment 3 based on his levels of vocalizations in the previous study; however, Ben engaged in similar levels of vocalizations across the demand and free sessions but much higher levels of hand stereotypy in the demand sessions relative to the free Hence, hand stereotypy was chosen as the target behav ior for the escape contingency in the second phase Clyde engaged in differentially higher levels of both general topographies of stereotypy (vocalizations and head hitting) in the demand sessions relative to the free sessions. Thus, we chose to conduct two sections

PAGE 64

64 of the second phase of the evaluation; first we applied the escape contingency to vocalizations a nd then later to head hitting. Results of the free vs. escape phase are displayed in the second phase of the top panel of Figure 4 1 for Ben an d the second and third phase s of the bottom panel of Figure 4 1 for Clyde Both subjects engaged in differentially higher levels of stereotypy in the escape sessions relative to the free sessions. Similar results were observed for the individual topogra phies in Figure 4 2. behavior was maintained by escape; however, for Ben, levels of stereotypy in the current phase were comparable to those observed in the previous phase. And, for Clyde, levels of stereotypy were actually lower in the escape sessions than in the demand sessions of the previous phase. These result s differ from what would be expec t ed with a reinforcement effect; if escape from demands had been reinforcing the shou ld have increased or at a very minimum should have decreased in the demand condition without escape. In addition, many of the stereotypic responses observ ed during the escape sessions occurred during the break from demands. Relative to levels displayed in the second phase in Figure 4 1, l ower levels of stereotypy were observed in the escape sessions when the break intervals were removed. In addition, stereotypy occur red at higher levels during the break interval s than during those times when a break wa s not presented (when demands were present) These results are important because they draw into question the reinforcing value of the escape from demands. Results of the demand versus enriched environment phase are displayed for Ben and Clyde in the last phase s of the top and bottom panels of Figure 4 1, respectively During e nriched

PAGE 65

65 e nvironment, Ben and Clyde engaged in differentially lower levels of all forms of stereotypy even though they were still work ing continuously. These results suggest that the aversiveness of the demands was an establishing operation for stereotypy, which was maintained by automatic reinforcement. Decreasing the aversiveness of the demand context may have acted as an abolishing operation to decrease the reinforcing value of st ereotypy. Alternatively, the preferred items may have provided access to behavior that effectively competed with stereotypy by producing higher quality reinforcement. Taken together, these result s show that although stereotypic behavior occ urred most often in the presence of demands behavior was not maintained by escape from demands and was instead maintained by automatic reinforcement established as such by the demand. The results of the experiment demonstrate a potential problem with ant ecedent based assessments. A researcher or clinician conducting an antecedent based assessment similar to the first phase of Experiment 3 might be tempted to conclude that the stereotypy was maintained by escape because it occurred in the presence of dema nds. Furthermore, the results of the second phase of the evaluation would also likely be interpreted as showing a demand function by a researcher or clinician conducting a functional analysis. This conclusion would also lead to a false positive because f urther analyses showed that the stereotypy was not reinforced by escape. Hence, these results hold potentially important implications for both antecedent based assessments and functional analyses of stereotypy.

PAGE 66

66 Figure 4 1. Results of the demand evaluation for Ben (top panel) and Clyde (bottom panel) are displayed. Percent of session with repetitive behavior is displayed across free (white triangles) demand (black circles), escape (white circles), and enric hed environmen t (cross hatches)

PAGE 67

67 Figure 4 2 Results of the demand evaluation for Ben topographies are displayed. Percent of session with hand stereotypy for Ben (top panel) and combined vocalizations (middle pane l) and head hitting (bottom panel) for Clyde is displayed across free (white triangles) demand (black circles), escape (white circles), and enriched environment (cross hatches) conditions

PAGE 68

68 CHAPTER 5 GENERAL DISCUSSION Overall Summary Experiment 1 determin ed that 2 4 % of typically functioning adults engaged in differentially high er le vels of repetitive behavior in the demand condition of an antecedent based assessment in which consequences were not manipulated. An overwhelming majority of individuals engage d in differentially high er levels during the alone condition. Experiment 2 determined that 20% of individuals with intellectual disabilities engage d in at least one topography of repetitive behavior ( stereotypy ) most often in the demand condition. Result s of Experiment 3 demonstrated that for those subjects who engaged in the highest levels of repetitive behavior in the demand condition, behavior was not reinforced by escape and was instead maintained by automatic reinforcement in the presence of demands. Experiment 1 Experiment 1 evaluated which conditions were most likely to evoke repetitive behavior in typically functioning adults. Adults were exposed to three conditions of an antecedent based assessment: alone, free, and demand. 66 % of subjects eng aged in differentiated levels of repetitive behavior in only the alone condition 18% of subjects engaged in differentiated levels of repetitive behavior in both the alone and demand conditions, but engaged in highest levels in the alone condition 4% of subjects engaged in the highest levels of repetitive behavior in the demand condition; however, levels in the alone condition were also differentiated One subject (2%) engaged in differentiated levels of repetitive behavior in only the demand condition. 1 0% of the subjects s howed undifferentiated results.

PAGE 69

69 Several limitations of the study should be noted. First, sessions were not run to stability due to limited time which each subject. In addition, no manipulation check was done to assess the extent to of the study, we felt that conducting a manipulation check may have increased the rpose of the study and thus made them more reactive to the procedures. Future research should evaluate the extent to which the alone and demand conditions are actually boring and aversive. The results of this study provide several implications for t he assessment of repetitive behavior or habits. First, these results have shown that repetitive behavior occurs in different contexts and is maintained by different types of reinforcement. If the goal is to reduce repetitive or habitual behavior, one shou ld first begin with an assessment similar to the one conducted here to determine whether the behavior occurs more often during low or aversive stimulation. Behavior that occurs most often during low stimulation should be addressed by increasing the ambien t stimulation in the environment, perhaps through environmental enrichment. Previous research has shown that behavior maintained by automatic reinforcement can be reduced through environmental enrichment in the form of noncontingent access to preferred ma terials (e.g., Horner, 1980; Ringdahl, Vollmer, Marcus, & Roane, 1997; Vollmer, Marcus, and LeBlanc, 1994). However, if behavior occurs most often when an individual is presented with demands or some type of aversive stimulation, treatment should address decreasing the aversiveness of the situation, perhaps by decreasing the rate of the demand or incorporating other types of preferred stimulation. Call, Wacker, Ringdahl,

PAGE 70

70 Cooper Brown, and Boelter (2004) found that decreasing task difficulty and adding non contingent attention could decrease noncompliance in the presence of demands; similar methods could be used to decrease repetitive behavior that occurs in the presence of demands. Results of Experiment 1 showed that some topographies occurred more often than others. Two of the three most frequently observed topographies were self scratching and nail biting, which are repetitive behaviors that can become harmful at extreme levels and intensities. Hence, results of Experiment 1 may have important implicat ions for clinical populations who engage in these topographies at a distressing or disruptive level. Self scratching and nail biting occurred most often in the alone condition, perhaps suggesting a self stimulatory function. As mentioned previously, envi ronmental enrichment may be an effective treatment for behavior that occurs in times of low stimulation; future research should determine whether environmental enrichment will be effective in decreasin g repetitive behavior in both nonclinical and clinical population s It function of the behavior, which more colloquially might be akin to alleviation of Experiment 2 The purpose of E xperiment 2 was to replicate Experiment 1 with intellectually disabled individuals and to evaluate whether stereotypy occurred at differentiated levels in the demand condition of an antecedent based assessment. Results of Experiment 2 were similar to thos e observed in Experiment 1 with typically functioning adults. 40 % engaged in the high est levels of repetitive behavior in the ignore condition of the assessment; 26.67% engaged in differentiated levels of repetitive behavior in the ignore

PAGE 71

71 and demand condit ion s, with highest levels in the ignore condition. 20% engaged in high, undifferentiated levels across all conditions; one subject (6.67%) engaged in the highest levels of repetitive behavior in the demand condition only. O ne final subject (6.67%) engage d in the highest levels of stereotypy in the free condition. For those subject s who engaged in lower levels of repetitive behavior in the demand condition, it is unclear what aspect of the condition suppress ed responding. It is possible that the task which in corporat ed the use of the hands for all subjects other than Clyde blocked repetitive behavior that involved the hands, e.g., hand stereotypy. competed with the reinforcement produced by the repetitive behavior. For three of the individuals who engaged in differentiated levels of stereotypy in the demand condition, at least one topography of stereotypy was differentially higher in demand than in any other condit ion. Given that consequences were not manipulated and the demand condition was akin to escape extinction, it seems unlikely that the that their behavior was maintained by automatic reinforcement established as such by the presence of demands We conducted Experiment 3 to further evaluate the function of stereotypy. Several limitations should be noted. First, brief assessments using five minutes session were conducted, w hich may not have allowed for enough time to observe an extinction effect during the demand sessions. Additionally, a formal demand assessment was not conducted at the beginning of the study. Informal assessments

PAGE 72

72 were conducted to find a task that appear ed difficult, but using a formal assessment may have ensured that an aversive task was used. Experiment 3 Experiment 3 was conducted to determine whether escape from demands or automatic reinforcement was maintaining stereotypy that occurred most often in the demand condition of an antecedent based assessment. To determine this, we conducted a series of evaluations. In the first evaluation, sessions with continuous demands were compared to session without demands, and all three subjects engaged in differ entially higher levels of stereotypy in the demand sessions. In addition, although demand sessions were procedurally identical to escape extinction, levels of stereotypy maintained or increased across the demand sessions, which is contrary to an extinctio n effect. Had stereotypy been reinforced by escape from demands, we should have seen a gradual reduction in stereotypy. It should be noted that the use of five minute sessions was a potential limitation and might not have allowed for enough time to obser ve the extinction effect. The second evaluation in Experiment 3 compared sessions with contingent escape for target stereotypy to sessions without demands. Although both subjects engaged in differentially higher levels of stereotypy in the escape sessions levels of stereotypy were not higher than in the previous demand sessions without escape. The application of the contingency did not produce an increase in responding and thus a reinforcement effect was not observed. The third evaluation in Experiment 3 compared continuous demand sessions with preferred stimulation (enriched environment) to continuous demand sessions without preferred stimulation (escape extinction). The results of this evaluation showed that

PAGE 73

73 when the aversiveness of the demand conditi on was reduced through the presentation of preferred stimuli, stereotypy decreased. In addition, because this phase also incorporated demand sessions that were procedurally identical to escape extinction, another opportunity was provided to observe the la ck of an extinction effect. Conclusions This series of experiments further confirms that the vast majority of stereotypic behavior is not maintained by socially mediated reinforcement contingencies. However, it is possible that social contingencies play ed a role in the development of behavior maintained by automatic reinforcement established as such by the presence of demands. It is possible that a subject such as Clyde initially engaged in stereotypic behavior when presented with demands and the demand s were removed. After being repeatedly paired with the reinforcing consequence of the environment becom ing demand free and thus less aversive, stereotypy may have acquired the properties of a conditioned reinfor cer (CR). Th us stereotypy could be maintaine d by automatic reinforcement via CR effects even when demands we re no longer removed. Interestingly, because extinction would be very difficult to implement had stereotypy acquired its reinforcing value in this way, punishment would be implicated for tre atment of behavior maintained by automatic reinforcement in the presence of demand s. If behavior had acquire d conditioned reinforc er properties because it was previously paired with a bettering of the environment, treatment would involve contingent worsen ing of the environment to remove the CR effect. This particular treatment recommendation is interesting because it is one of the few cases in which the results of a functional analysis would prescribe punishment as treatment. In addition, applying contin gent aversive stimulation would assist in evaluating whether behavior

PAGE 74

74 was truly operant, in which case it would decrease, or if it w ere elicited (respondent behavior), in which case it would increase. In addition, t he current studies draw into question t he use of antecedent based assessment s to infer socially mediated sources of reinforcement T he results also draw into question the conclusions of previous research that have used antecedent based assessments Read ers continue to interpret the results of studies such as Durand and Carr (1987) as indicating as escape function of be havior when the data presented here clearly refute that notion Researchers and clinicians should be skeptical when interpreting the results of studies using antecedent based ass essments. This sequence of stud ies also has important implications for researchers who conduct consequence based functional analyses but are unable to decrease behavior during extinction such as in the rare examples presented by Iwata et al ( 1994) In the rare circumstances such as those seen with Clyde in Experiment 3, b ehavior maintained by automatic reinforcement in the presence of demands will appear to be escape maintained in typical functional analyses If escape extinction is ineffective follow ing the functional analysis, efforts should be made to determine whether the behavior is not in fact maintained by automatic reinforcement. Evaluations similar to those employed in Experiment 3 could be conducted These results and implications might ha ve generality for behavior other than stereotypy and for populations other than ID individuals especially given results of Experiment 1. Typically developing adults and intellectually disabled children both appear to engage in repetitive behavior maintai ned by automatic reinforcement in the presence of demands. In addition, stereotypic SIB may be reinforced in a similar

PAGE 75

75 manner at least occasionally. Azrin, Hutchinson, and McLaughlin (1965 ) observ ed that aggression often occurs as a result of aversive st imulation; therefore aggressive behavior could also be reinforced by automatic reinforcement in the presence of demands (consider people who punch walls or punching bags during or immediately after aversive events). In conclusion, automatic reinforcement established as such by aversive stimulation is a function that may influence many populations and response topographies. This research may be viewed as a starting point to evaluate automatic reinforcement not only in austere environments but also in aversi ve environments.

PAGE 76

76 LIST OF REFERENCES Azrin, N. H., Hutchinson, R. R., & McLaughlin, R. (1965). The opportunity for aggression as an operant reinforcer during aversive stimulation. Journal of the Experimental Analysis of Behavior, 8, 171 180. Berkson, G., & Davenport, J., R.K. (1962). Stereotyped movements of mental defectives. American Journal of Mental Deficiency, 66, 849 853. Berkson, G., & Tupa, M. (2000). Early development of stereotyped and self injurious behaviors. Journal of Early Intervention, 23, 1 19. Bildsoe, M., Heller, K. E., & Jeppesen, L. L. (1991). Effects of immobility stress and food restriction on stereotypies in low and high stereotyping female ranch mink. Behavioural Processes, 25, 179 189. Bodfish, J. W., Symons, F. J., Parker, D. E., & Lewis, M. H. (2000). Varieties of repetitive behavior in autism: Comparisons to mental retardation. Journal of Autism and Developmental Disorders, 30, 237 243. Call, Wacker, Ringdahl, Cooper Brown, & Boelter (2004). An assessment of antecedent events infl uencing noncompliance in an outpatient clinic Journal of Applied Behavior Analysis, 37, 145 158. Carter, P.J., & Russell, K.A. (2002). IQ tests to keep you sharp. New York, NY: Sterling Publishing Davenport, R. K., & Berkson, G. (1963). Stereotyped moveme nts in mental defectives: II. Effects of novel objects. American Journal of Mental Deficiency, 67, 879 882. Deaver, C.M., Miltenberger, R.G., & Stricker, J.M. (2001). Functional analysis and the treatment of hair twirling in a young child. Journal of Appl ied Behavior Analysis, 34, 535 538. Dufrene, B. A., Watson, T. S., & Kazmerski, J. S. (2008). Functional analysis and treatment of nail biting. Behavior Modification, 32, 913 927. Durand, V. M., & Carr, E. G. (1987). Social influences on self stimulatory b ehavior: Analysis and treatment application. Journal of Applied Behavior Analysis, 20, 119 132. Foreh and, R. & Baumeister, A.A. (1971 ) Rate of stereotyped body rocking of severe retardates as a function of frustration of goal directed behavior Journal of Abnormal Psychology, 78 35 42 Hansen, D. J., Tishelman, A. C., Hawkins, R. P., & Doepke, K. J. (1990). Habits with potential as disorders: Prevalence, severity, and other characteristics among college students. Behavior Modification, 14, 66 80.

PAGE 77

77 Horner, R D. (1980). The effects of an environmental enrichment program on the behavior of institutionalized profoundly retarded children. Journal of Applied Behavior Analysis, 13, 473 491. Iwata, B. A., Dorsey, M. F., Slifer, K. J., Bauman, K. E., & Richman, G. S (1994). Toward a functional analysis of self injury. Journal of Applied Behavior Analysis, 27, 197 209. Iwata, B. A., Pace, G. M., Dorsey, M. F., Zarcone, J. R., Vollmer, T. R., Smith, R. G., Rodgers, T. A., Lerman, D. C., Shore, B. A., Mazaleski, J. L., Goh, H. L., Cowdery, G. E., Kalsher, M. J., McCosh, K. C., & Willis, K. D. (1994). The functions of self injurious behavior: An experimental epidemiological analysis. Journal of Applied Behavior Analysis, 27, 215 240. Jones, R. S. P. (1999). A 10 year fol low up of stereotypic behavior with eight participants. Behavioral Interventions, 14, 45 54. Kaufman, M. E., & Levitt, H. (1965). A study of three stereotyped behaviors in institutionalized mental defectives. American Journal of Mental Deficiency, 69, 467 473. Kennedy, C. H., Meyer, K. A., Knowles, T., & Shukla, S. (2000). Analyzing the multiple functions of stereotypical behavior for students with autism: implications for assessment and treatment. Journal of Applied Behavior Analysis, 33(4), 559 571. Lova as, I., Newsom, C., & Hickman, C. (1987). Self stimulatory behavior and perceptual reinforcement. Journal of Applied Behavior Analysis, 20, 45 68. Mace, F.C., Browder, D.M., & Lin, Y. (1987). Analysis of demand conditions associated with stereotypy. Journ al of Behavior Therapy and Experimental Psychiatry, 18(1), 25 31. Pace, G. M., Ivancic, M. T., Edwards, G. L., Iwata, B. A., & Page, T. J. (1985). Assessment of stimulus preference and reinforcer value with profoundly retarded individuals. Journal of A pplied Behavior Analysis, 18, 249 255. Rapp, J. T., & Vollmer, T. R. (2005a). Stereotypy I: A review of behavioral assessment and treatment. [Literature Review]. Research in Developmental Disabilities, 26, 527 547. Rapp, J. T., & Vollmer, T. R. (2005b). S tereotypy II: A review of neurobiological interpretations and suggestions for an integration with behavioral methods. [Literature Review]. Research in Developmental Disabilities, 26, 548 564. Ringdahl, J. E., Vollmer, T. R., Marcus, B. A., & Roane, H. S (1997). An analogue evaluation of environmental enrichment: The role of stimulus preference. Journal of Applied Behavior Analysis, 30, 203 216.

PAGE 78

78 Roane, H. S., Vollmer, T. R., Ringdahl, J. E., & Marcus, B. A. (1998). Evaluation of a brief stimulus preferen ce assessment. Journal of Applied Behavior Analysis, 31, 605 620. Stricker, J., Miltenberger, R. G., Anderson, C. F., Tulloch, H. E., & Deaver, C. M. (2002). A functional analysis of finger sucking in children. Behavior Modification, 26, 424 443. Vollmer, T. R., Marcus, B. A., & LeBlanc, L. (1994). Treatment of self injury and hand mouthing following inconclusive functional analysis. Journal of Applied Behavior Analysis, 27, 331 344. Woods, D. W., & Miltenberger, R. G. (1996). Are persons with nervous habit s nervous? A preliminary examination of habit function in a nonreferred population. Journal of Applied Behavior Analysis, 29, 259 261. Woods, D. W, Miltenberger, R. G., & Flach, A. D. (1996). Habits, tics, and stuttering: Prevalence and relation to anxiety and somatic awareness. Behavior Modification, 20, 216 225.

PAGE 79

79 BIOGRAPHICAL SKETCH Amanda Bosch was born in Fargo, North Dakota (ND) in 1978 and was raised in Reno, Nevada After Amanda ob tained her Bachelor of Arts in p sychology from the University of Nevad a, Reno in 2002, she moved to Atlanta, Georgia to work for two years at the Marcus Institute with children with behavior and feeding disorders. In 2004, Amanda moved back to Fargo, ND to attend graduate school at North Dakota State University, where she r ecei ved her Master of Science in clinical p sychology in 2006. Amanda lived in Gainesville, Florida for five years, working on her Ph D under the mentorship of Tim Vollmer. She received her Ph.D. from the University of Florida in the summer of 2011. Afte r graduation, Amanda move d to Lubbock, Texas to be a Visiting Professor and Postdoctoral Intern at Texas Tech University.