Comparison of Sensory and Edible Reinforcers

MISSING IMAGE

Material Information

Title:
Comparison of Sensory and Edible Reinforcers
Physical Description:
1 online resource (46 p.)
Language:
english
Creator:
Fahmie, Tara Ann
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:
Iwata, Brian
Committee Members:
Stehouwer, Donald J
Vollmer, Timothy R
Griffin, Cynthia C

Subjects

Subjects / Keywords:
acquisition -- autism -- maintenance -- preference -- reinforcer
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:
Previous research has shown that individuals with developmental disabilities generally prefer edible items over leisure items. Other research has reported that “sensory” (leisure) items facilitate acquisition and maintenance of behavior better than do edible items for individuals with autism. Although these findings seem contradictory, the data are not comparable because studies on preference were conducted with subjects without autism, whereas data on performance were collected for subjects with autism. The current study examined preference and performance of children with and without autism using sensory and edible reinforcers. Results showed that edibles were more preferred (Experiment 1) and resulted in higher rates of responding under maintenance conditions (Experiment 3) in subjects with and without autism. Edible and sensory items resulted in equal rates of response acquisition (Experiment 2) for both samples and for subjects with different patterns of preference.
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 Tara Ann Fahmie.
Thesis:
Thesis (Ph.D.)--University of Florida, 2012.
Local:
Adviser: Iwata, Brian.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-08-31

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
lcc - LD1780 2012
System ID:
UFE0044067:00001


This item is only available as the following downloads:


Full Text

PAGE 1

1 COMPARISON OF SENSORY AND EDIBLE REINFORCERS By TARA A. FAHMIE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPH Y UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Tara A. Fahmie

PAGE 3

3 To my wonderful family, Anthony, Deborah, and Christopher

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank the many people who have encouraged my professional development, preserved my sanity, and h elped me to accomplish so many personal goals along the way. Foremost, I need to thank my family. I thank my mother for showing me so much love and support. I thank my father for reminding me to put down my work every once in a while. I thank my brother fo r his unfaltering confidence in me. I also thank my lab mates (my second family) for their immeasurable contributions to my graduate career. In particular, I thank Kasey Stephenson for demonstrating amazing strength of character, Jen Hammond for her helpfu l advice and encouragement, and Katie Jann for her contributions throughout the course of this project. I would also like to thank all of the instructors of my behavior analysis classes; I consider myself beyond lucky to have had the opportunity to work wi th so many amazing instructors. And last, I would like to thank the members of my dissertation committee, Drs. Cynthia Griffin, Donald Stehouwer, and Timothy Vollmer for their time and assistance. Of course, none of this would have been possible without th e immense support of my advisor, Dr. Brian Iwata. His guidance throughout the past five years has been invaluable, and I know that his influence will extend far past my graduation from the University of Florida.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 7 LIST OF FIGURES ................................ ................................ ................................ ......................... 8 ABSTRACT ................................ ................................ ................................ ................................ ..... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 10 Reinforcer Classes ................................ ................................ ................................ .................. 11 Comparison of Sensory and Edible Re inforcers ................................ ................................ ..... 12 2 EXPERIMENT 1: PREFERENCE FOR LEISURE AND EDIBLE ITEMS ......................... 17 Method ................................ ................................ ................................ ................................ .... 17 Subjects, Setting, and Materials ................................ ................................ ...................... 17 Response Measurement and Interobserver Agreement ................................ ................... 18 Procedures ................................ ................................ ................................ ............................... 18 Results and Discussion ................................ ................................ ................................ ........... 19 3 EXPERIMENT 2: ACQUISITION OF RESPONDING FOR EDIBLE AND LEISURE ITEMS ................................ ................................ ................................ ................................ .... 24 Method ................................ ................................ ................................ ................................ .... 24 Subjects, Setting, and Materials ................................ ................................ ...................... 24 Procedures ................................ ................................ ................................ ....................... 25 R esponse Measurement and Interobserver Agreement ................................ ................... 26 Results and Discussion ................................ ................................ ................................ ........... 27 4 EXPERIMENT 3: MAINTENANCE OF RE SPONDING FOR EDIBLE AND LEISURE ITEMS ................................ ................................ ................................ ................... 32 Method ................................ ................................ ................................ ................................ .... 32 Subjects, Setting, and Materials ................................ ................................ ...................... 32 Response Measurement and Interobserver Agreement ................................ ................... 32 Procedures ................................ ................................ ................................ ....................... 33 Baseline ................................ ................................ ................................ .................... 33 Schedule training ................................ ................................ ................................ ...... 33 Maintenance ................................ ................................ ................................ ............. 34 Results and Discussion ................................ ................................ ................................ ........... 35

PAGE 6

6 5 GENERAL DISCUSSION ................................ ................................ ................................ ..... 38 LIST OF REFERENCES ................................ ................................ ................................ ............... 43 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 46

PAGE 7

7 LIST OF TABLES Tab le page 2 1 Subject characteristics. ................................ ................................ ................................ ....... 22 2 2 Ranki ngs for the HPE and HPL items in the combined assessments. ............................... 22 3 1 Most to least prompt hierarchy ................................ ................................ .......................... 30

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Selection percentages for the HPE and HPL items in the Edible only, Leisure only, and Combined assessments. ................................ ................................ ............................... 23 3 1 Number of steps mastered under Leisure and Edible conditions. ................................ ...... 31 4 1 Rate of target responding across Baseline, Leisure, and Edible conditions. ..................... 37

PAGE 9

9 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 COMPARISON OF SENSORY AND EDIBLE REINFORCERS By Tara A. Fahmie Augus t 2012 Chair: Brian Iwata Major: Psychology Previous research has shown that individuals with developmental disabilities generally prefer edible items over leisure items. Other research has reported that sensory ( leisure ) items facilitate acquisition and maintenance of behavior better than do edible items for individuals with autism Although these findings seem contradictory, the data are not comparable because studies on preference were conducted with subjects without autism, whereas data on performa nce were collected for subjects with autism. The current study examined preference and performance of children with and without autism using sensory and edible reinforcers. Results showed that edibles were more preferred (Experiment 1) and resulted in high er rates of responding under maintenance conditions (Experiment 3) in subjects with and without autism. Edible and sensory items resulted in equal rates of response acquisition (Experiment 2) for both samples and for subjects with different patterns of pre ference.

PAGE 10

10 CHAPTER 1 INTRODUCTION Identifying reinforcers for adaptive behavior is an important step in developing effective behavioral intervention s for individuals with disabilities who often have deficits in the areas of language, self care, academic, and social skills. Decisions regarding the type of reinforcer used may be influenced by a number of variables, including the availability, practicality, and, most important, efficacy of the reinforcer. Basic research has revealed that the efficacy of parti cular events as reinforcers largely depends on individual conditioning histories and present levels of deprivation, both of which suggest an individualized approach to the selection of reinforcers. Such an approach commonly is accomplished in applied setti ngs through stimulus preference assessments (Pace, Ivancic, Edwards, Iwata, & Page, 1985) which are used to predict the value of various stimuli as reinforcers for an individual. Pace et al. (1985 sensory items (e.g., lights), odors (e.g., dried hibiscus), foods (graham cracker), and social events (e.g., hugs). Items that were approached the most (at least 8 of 10 presentat ions) by each subject were considered high preference items and were shown to maintain a higher level of adaptive responding than items approached least (less than 5 of 10 presentations). Results also revealed between subject differences in preference for specific stimuli. These results were replicated across different assessment formats (e.g., multiple stimulus and duration based formats); the majority of outcomes from preference assessment research have supported the idiosyncratic nature of preference bet ween subjects (e.g., DeLeon & Iwata, 1996) and within subjects across time (Hanley, Iwata, & Roscoe, 2006). Nevertheless certain broad classes of stimuli (e.g., social, edible, leisure) have proven generally effective as reinforcers across a large number of behaviors

PAGE 11

11 and subjects, and thus several studies (e.g., Rincover & Newsom, 1985) have sought to compare stimulus classes with respect to reinforcer efficacy and durability. Reinforcer Classes Reinforcers have been categorized in behavioral research alon g a variety of dimensions, including the delivering agent (social, nonsocial), contingency (positive, negative), conditioning history (primary, secondary), and stimulus form (attention, item, activity). Most research on acquisition has been conducted using social positive reinforcement (attention, tangible items, and activities). In the first reported application of operant conditioning with a human Fuller (1949) delivered a sugar milk solution contingent on the arm movements of an adolescent mal e labeled of adaptive behaviors, such as communication (Barton, 1970) daily living skills (Wolf, Risley, & Mees, 1964) and compliance (Whitman, Zakaras, & Chardos, 1971) For example, Whitman et al. used cereal, chocolate, and diet soda as reinforcers for the completion of single step instructions in two children with profound disabilities who were reported by their teachers as unwilling to follow vocal instructions. Physical guidance was used initially to ensure contact with the reinforcer but was later faded from training. Results showed a marked increase in compliance with instructions used during training as well as generalization to novel instructions that were not used during training for both subjects. Sensory stimulation in the form o f vibration, light, and sound also has been shown to reinforce simple arbitrary responses (e.g., lever pressing) in animals (e.g., Kish, 1 955; Kish & Baron, 1962) and humans (Bailey & Meyerson, 1969; Fehr, 1979) Some of the earliest studies to apply sensory reinforcers to socially meaningful behavior were conducted by Rice and colleagues (Rice & McDaniel, 1966; Rice, McDaniel, & Stallings, 1967) and involved increasing

PAGE 12

12 could be used to establish more complex behaviors, such as vocal behavior (Fineman, 1968; Fineman & Ferjo, 1969) and conditional dis criminations (Rincover & Ne wsom, 1985) Comparison of Sensory and Edible Reinforcers Direct comparisons of the effects of sensory and edible reinforcers on response acquisition (Ferrari & Harris, 1981; Rehagen & Thelen, 1972; Rincover & Newso m, 1985) have shown somewhat inconsistent results. Rehagen and Thelen compared rates of button pressing in 20 individuals with moderate to severe disabilities who received 10 s access to either sensory (vibrating massager) or edible (sugared cereal) stimu li on a continuous schedule of reinforcement. The effects of each reinforcer class were evaluated in a group design, with 10 subjects assigned to the sensory group and 10 to the edible group. Results showed no significant difference in response rates acros s groups, indicating relatively equal efficacy across edible and sensory reinforcers. However, t he group design did not permit a direct within subject comparison of the effects of each reinforcer, which is problematic if certain individuals are prone to g reater reinforcer sensitivity by reinforcer type. In addition, n o individual data were reported so the extent to which group mean s w ere representative of responding by individuals is unknown. Ferrari and Harris (1981) observed individual differences in the effectiveness of edible (fruit, candy, and cereal) and sen sory (music, strobe lights, and vibrating massager) reinforcers in 4 children with autism. A within subject design was used and consisted of two phases, which reinforcers (single sensory and edible items were alternated across sessions) on a fixed ratio (FR) 6 schedule of reinforcement. In phase 2, the experimenters trained a two choice object discrimination using continuous reinforcement. One reinforcer type wa s used to train an initial set of discriminations and, following mastery with the initial set, the second reinforcer type was

PAGE 13

13 used to train a second set of discriminations. Phase 1 r esults showed overall higher levels of responding for vibrating items in t wo subjects and for edible items in the remaining two subjects. P hase 2 results were the same with vibration (two subjects) and edibles (two subjects) resulting in faster acquisition and better maintenance of the trained discrimination. The authors also n oted that no difference in satiation effects was observed across stimulus classes, although data were not presented on this measure. Rincover and Newsom (1985) also compared the effects of edible items (various snack s ) and sensory items or events (auditor y, visual, and tactile stimulation) on the outcomes of discrimination training in three children with autism. Four conditions were included, each involving the repeated presentation of a two choice visual discrimination task. The same edible item was deliv ered contingent on each correct response in the Single Edible condition, whereas different edible items were rotated as reinforcers in the Multiple Edible condition; the Single Sensory and Multiple Sensory conditions were identical except that sensory item s were used. Accuracy of responding was measured by calculating the percentage of trials with a correct response within 10 s of task presentation. Durability of responding was measured by counting the number of trials completed before meeting a satiation c riterion or until a 300 trial cap was met. Responding on the discrimination task was more accurate and more durable for all subjects when the consequences for correct responses were multiple sensory items, as compared to multiple edible items. Only small d ifferences in favor of sensory items were obtained when single sensory items were compared to single edible items: Single sensory items resulted in slightly higher accuracy but no better durability compared to single edible items. Results from more recent studies on the assessment of preference (Bojak & Carr, 1999; DeLeon, Iwata, & Roscoe, 1997) however, indicate that edible items are generally preferred to

PAGE 14

14 leisure items in individuals with intellectual disabilities. DeLeon et al. conducted a multiple stimulus withou t replacement (MSWO) preference assessment (DeLeon & Iwata, 1996) for edible items alone and leisure items alone with 14 individuals with moderate to profound mental retardation. The authors subsequently included the most highly preferred items from each class in a combined preference assessment. Twelve o f their subjects showed a pronounced preference for the edible items over leisure items. Bojak and Carr replicated this general finding in four adults with severe mental retardation. The authors additionally attempted to alter the establishing operation fo r edibles by conducting combined MSWO assessments immediately following meals. Resulting preferences were not changed by this manipulation (i.e., preferences remained in favor of edible items), indicating a general resistance to satiation effects for edibl e items. No comparable manipulation was conducted for leisure items. It should be noted that the leisure items in these studies primarily included auditory, tactile, visual, or olfactory features, and thus, these items should be considered comparable to th previous research, both classes of items will be referenced by the more commonly used term, In summary although results of some studies have shown that edibles are generally preferred (Bojak & Carr, 1999; DeLeon, et al., 1997) results of others (Rincover & Newsom, 1985) suggest that leisure items may be more effective reinforcers. Several possible explanations exist for these seemingly discr epant results. First, edible items may be preferred, but also may be more prone to satiation relative to leisure items. Satiation was unlikely to occur during the brief preference assessments conducted by Bojak and Carr and by DeLeon et al., so it is uncle ar whether the highly preferred edible items would have maintained responding longer than the

PAGE 15

15 less preferred leisure items when used over extended periods of time. Second, results of the Rincover and Newsom study may have reflected subject characteristics. The authors did not compare preference between reinforcer classes, so perhaps subjects in that study happened to prefer leisure over edible reinforcers. In addition, all subjects in the Rincover and Newsom study were diagnosed with autism whereas subject s in the studies conducted by Bojak and Carr and by DeLeon et al. were not. Several authors have noted that sensory stimuli might be particularly potent and durable reinforcers for individuals with autism because these individuals frequently engage in ster eotypic behaviors that persist throughout the day (e. g., Ferrari & Harris, 1981) Early accounts suggested that these particularly rigid and repetitive forms of behavior were likely maintained by the sensory stimulation produced as a direct result of the behavior; this hypothesis was bolstered by evidence t hat stereotypy decreased when the resulting sensory consequences were removed (e.g., "sensory extinction"; Rincover, Cook, Peoples, & Packard, 1979) In addition, a large proportion of studies in which functional analyses were conducted of stereotypy have shown that the behavior typically is not maintained by social reinforce ment (Hanley, Iwata, & McCord, 2003) adding f urther evidence that stereotypy produces automatic reinforcement in the form of sensory stimulation. The use of sensory events has been broadly recommended as an important component apy); however, evidence in support of this recommendation is largely anecdotal and has little empirical basis (Rogers & Ozonoff, 2005) The purpose of the current studies was to extend research on preference, effica cy, and durability of leisure and edible items as reinforcers for individuals with and without autism. In Experiment 1, we replicated the procedures of DeLeon et al. (1997) to compare preference for

PAGE 16

16 leisure and edible items. Experiment 2 was a comparison o f the effects of leisure and edible reinforcement on the acquisition of a 6 step response chain. We compared the effects of leisure and edible reinforcement on the maintenance of a simple motor response in Experiment 3.

PAGE 17

17 CHAPTER 2 EXPERIMENT 1: PREFER ENCE FOR LEISURE AND EDIBLE ITEMS Method Subjects, Setting, and Materials Twelve individuals diagnosed with autism or an intellectual disability participated in one or more studies. All subjects attended a special education school approximately 7 hrs a day at the time of participation. Table 2 diagnoses, and participation. A diagnosis of autism or of a specific intellectual disability (e.g., Down was made by a qualified physician. A diagnosis of intellectual disability (not otherwise specified) was listed as the primary a physician or school psychologist All as sessments took place in a research room located close to the s ubjects Preference for 16 stimuli was assessed for each subject. Eight edible items were selected arbitrarily from a list t hat included sweet, salty, and sour snacks similar (but not identical) to those typically used during acquisition training in the school All edible items were bite sized and able to be consumed easily within 15 s. Examples included candies (sweet and sour ), crackers, cereals, pretzels, peanuts, and pickles. Eight leisure items were selected based on recommendations from an on site occupational therapist and included a sample of visual, auditory, and tactile stimuli. The sensory function was a prominent fea ture of each leisure item sampled. Inclusion of a specific edible or leisure item in any individual assessment was based upon consumption (edible) o r interaction (leisure ) with the item during a brief pre experimental exposure. Informal interviews with the were not being used on a daily basis in the classroom or the home settings.

PAGE 18

18 Response Measurement and Interobserver Agreement The primary dependent measure for all preference assessments was select ion, defined as approach or physical contact with an item. Data were summarized by dividing the number of times an item was selected by the number of times that item was available as a choice. Items were then ranked based on these selection percentages. T he highest percentage was given a rank of 1 and the lowest a rank of 8. If two items had identical selection percentages, the y were ranked equally by calculating the mean of the two ranks. Reliability was assessed by having a second observer simultaneousl y but independently compared on a trial by trial basis, and an agreement was scored if both observers listed the same item selected for a given trial. Trials with agreeme nt were divided by total trials and the resulting quotient was multiplied by 100. The mean agreement across subjects was 100% for leisure and edible assessments and 98.8% (range, 92.5% to 100%) for combined assessments. Procedures Preference assessments w ere conducted using a multiple stimulus without replacement (MSWO) format, similar to that described by DeLeon and Iwata (1996) Prior to the first trial of each assessment, the subject was given an opportunity to sample the items to be used in the assessment. An array of eight items then was presented in front of the subject. The Edible only assessment included all eight edible items; the Leisure only assessment included all eight leisure of an item resulted in access to that item for 15 s, removal of that item from the array, rotation of This procedure continued until all items were selected. Attem pts to select more than one item at

PAGE 19

19 The Edible only and Leisure only assessments were conducted five times each. The top four ranked edible items (HPE) and the top four ranked leisure items (HP L ) from these assessments subsequently were presented in a Combined array of edible and leisure items, which also was presented five times. This procedure resulted in 15 assessments per subject. Results and Discussion Table 2 2 shows rankings for the HPE and HP L items in Combined preference assessments for the 12 subjects (Wes, Billy, Nick, Martin, Dan, Carl, Ken, Mark, Elliot, Cade, Caleb, and Henry) who participated. All edible items were ranked higher than all leisure items in the combined assessment for 8 subjects (66.7%); these subjects showed an exclusive preference for edible over leisure items. Edible items were ranked higher than all but one leisure item in the combined assessments for 2 additional subjects (16.7%). Th ese subjects showed a general, albeit not exclusive, preference for edible items over leisure items. Rankings for the remaining two subjects (Caleb and Henry) showed mixed preference for edible and leisure items. Figure 2 1 shows selection percentages for the same top four ranked edible and leisure items in the Leisure only Edible only and Combined assessments. Logged percentages are displayed because the equation used to derive selection percentages re sults in a negatively accelerating decrease in percen tage for later selections from the array (i.e., if items consistent ly were selected 1 st 2 nd 3 rd 4 th 5 th 6 th 7 th and 8 th across all 5 assessments, percentages equal ed 100%, 50%, 30%, 25%, 20%, 16.7%, 14.3%, and 12.5%, respectively). All 48 leisure it ems shown in Figure 2 1 were selected on a lower percentage of trials in the Combined assessment relative to the Leisure assessment. This was true for all subjects, despite differences in the relative rankings of leisure and edible items in the Combined as sessment. By contrast, only 21 of the 48 edible items (43.8%) shown in Figure 2 1 were selected on a lower percentage of trials in the Combined assessment relative to the Edible only assessment.

PAGE 20

20 The remaining 27 edible items (56.2%) were selected on a hig her percentage of trials. Decreases in the percentages of individual items indicate either a consistent decrease in their preference ranking or a general decrease in the consistency of rankings. Thus, although some variability in individual preferences may have been introduced in the Combined assessment (decreases in percentages for some leisure and some edible items), this was more frequently accounted for by changes in the rankings within each class as opposed to across classes. These results replicate t hose reported by DeLeon et al. (1997) and by Bojak and Carr (1999) in that most subjects, 9 of 12, showed an exclusive preference for edible items. This extends the results of previous comparisons in that the gene ral effect was shown in subjects with autis m as well as subjects with intellectual disabilities. Preference for edible items may be particularly strong because these items represent primary sources of reinforcement with an extensive phylogenic and ontogenic history. Leisure items generally are cons idered secondary reinforcers, though some evidence exists to suggest that sensory stimulation represents a primary source of reinforcement (Kish, 1966) Kish first ng that sensory stimuli showed similar patterns of satiation and habituation as did primary reinforcers and that sensory reinforcers showed generality across species. Nevertheless, given that the sensory reinforcers in the current study could only be obtai ned through engagement with leisure items, this source of stimulation may have been more effortful to obtain than that of edibles. Perhaps a stronger establishing o peration for edible items exists because meals are scheduled at spec ific points in the day, and generally are restricted to three, whereas all other sources of sensory stimulation (visual, auditory, tactile) are continuously experienced throughout the day. Although we did not attempt to control for background levels of food consumption or sensory

PAGE 21

21 stimulation, we did ensure that all items used in the preference assessments were unavailable to the subjects outsid e of experimental sessions. In the current study, stimuli were selected so that a range of sensations was sampled (e.g., sweet, salty, and sour foods as well as tactile, visual, and auditory toys) and so that stimuli were unique to the experimental setting. Thus, item selection was not based on reports from the highly preferred) items. Although the procedure used in the current study provided a more even distribution across sensory modalities, it may have limited the generality of preference outcomes for subjects with high ly idiosyncratic preference (e.g., a spe cific toy to which subject has access at home).

PAGE 22

22 Table 2 1. Subject characteristics. Subject Age Diagnosis Experiment Ken 5 Autism spectrum disorder (ASD) 1, 2, 3 Elliot 7 Autism 1 Henry 12 ASD 1, 2, 3 Dan 16 ASD 1, 3 Caleb 17 ASD 1, 2, 3 Mark 18 ASD 1 Carl 20 ASD 1 Martin 10 Intellectual disability (ID) 1 Billy 12 Dandy Walker syndrome 1, 3 Wes 19 Down Syndrome, Klinefelter Syndrome 1 Nick 20 ID 1, 2, 3 Cade 22 ID Attention deficit hyperactivity disorder, Microcephaly 1 Table 2 2. Ranki ngs for the HPE and HP L items in the combined assessments. Subject Highly Preferred Edible (HPE) items Highly Preferred Leisure (HP L ) items Nick 1, 2, 3, 4 5, 6, 7, 8 Billy 1, 2, 3, 4 5, 6, 7, 8 Wes 1, 2, 3, 4 5, 6, 7, 8 Martin 1, 2, 3, 4 5, 6, 7, 8 D an 1, 2, 3, 4 5, 6, 7, 8 Carl 1, 2, 3, 4 5, 6, 7, 8 Ken 1, 2, 3, 4 5, 6, 7, 8 Mark 1, 2, 3, 4 5, 6.5, 6.5, 8 Elliot 1, 2, 3, 5 4, 6, 7.5, 7.5 Cade 2, 3, 4.5, 4.5 1, 6, 7, 8 Caleb 1, 3, 5, 6 2, 4, 7, 8 Henry 2, 3, 7, 8 1, 4, 5, 6

PAGE 23

23 Figure 2 1. Se lection percentages for the HPE and HP L items in the Edible only Leisure only and Combined assessments.

PAGE 24

24 CHAPTER 3 EXPERIMENT 2: ACQUIS ITION OF RESPONDING FOR EDIBLE AND LEISURE ITEMS Method Subjects, Setting, and Materials Four individuals from Experim ent 1 (Caleb, Henry, Nick, and Ken) participated in Experiment 2. These individuals were selected for participation because they represented two different patterns of preference (exclusive preference for edibles and mixed preference) and two different diag noses (autism and intellectual disability). This study was conducted in the same therapy room as that used in Experiment 1. Sessions were conducted one to two times per day, three to five days per week. Prior to the experiment, four block (Lego ) structur es were constructed, each including one baseboard and five blocks of four different colors and shapes. A six step task analysis for each structure specified the order and the position in which the baseboard and blocks were to be arranged in the structure. The first step consisted of pulling the baseboard out from the group of unstructured blocks and placing it directly in front of the subject. Steps two through six each involved picking up and placing one block in a specific location. This task was selected for use in Experiment 2 because it seemed to share properties with many adaptive chains, yet it allowed for multiple variations that could be equated in difficulty. Several criteria were established to ensure that each structure was equally difficult Fi rst, each structure contained the same number of materials. Second, each completed structure had based on several criteria. T he experimenter initially placed all five bloc ks in a vertical column ; this starting position was equal to zero points. Each subsequent offset (left, right, forward, or backward movement) or rotation from the starting position added one point to the difficulty score. Each

PAGE 25

25 structure was created to have a difficulty score of 5 points. In addition, the two blocks that were identical in shape and color always were positioned in parallel in the final structure. The materials necessary to complete one Lego structure, plus a model of a completed Lego struct ure, were available to the subjects during training. The four high preferred edible (HPE) stimuli and the four high preferred leisure stimuli (HP L ) from the Combined preference assessments of Experiment 1 were rotated across correct trials in the Edible a nd Leisure conditions, respectively. A picture board showing the edible and leisure stimuli to be used as reinforcers was visible to the subject during the respective sessions. Procedures The experimenter used forward chaining and most to least prompting to train the 6 step task analysis in a manner identical to that described by Libby, Weiss, Bancroft, and Ahearn (2008). Each session consisted of one probe trial followed by 10 training trials. Each trial commenced reinforcement, or blocking during the probe trial. A probe trial was completed when the subject made an error or when 15 s elapsed with no responding. The experimenter ended the probe trial by removing the materia ls. A forward chaining procedure was used during training trials. Each step was taught in a sequential order, beginning with step one. A subject progressed to the next step of the task analysis following two consecutive correct independent responses on a given step (step mastery criterion). A subject was retrained on a previously mastered step following two consecutive errors on the previously mastered step. A most to least manual guidance procedure with a delay feature also was used during training trials (see Table 3 1 for a prompt hierarchy). Training of a step always began with immediate hand over hand guidance. Manual guidance subsequently was decreased by one level

PAGE 26

26 following two consecutive correct prompted responses at the designated prompt level. Th e less intrusive prompts (forearm, upper arm, elbow) all were preceded by a 2 second delay, during which independent correct responses (or errors) could occur. Manual guidance was increased by one level if the subject emitted two consecutive errors at the designated prompt level. An error was defined as placing (or attempting to place) a material in the wrong position or in the wrong order, or allowing 15 s to elapse without making contact with any materials. Errors always resulted in immediate hand over ha nd guidance during the trial in which the error was made. Reinforcers were delivered contingent on each correct, prompted or independent, response during training. During the Edible condition, correct responding on a training step resulted in the delivery of praise and 15 s access to a single edible item. During the Leisure condition, correct responding on a training step resulted in the delivery of praise and 15 s access to a single leisure item. Following reinforcer deliveries, the experimenter demonstrat ed all steps required to build the structure prior to initiating the next training trial. The effects of edible and leisure items on response accuracy were compared using a multielement design, in which Edible and Leisure conditions were alternated across sessions. Two structures were taught simultaneously one structure under each condition. When the subject independently completed all six steps of a structure across two consecutive trials (structure mastery criterion), two new structures were introduced, o ne structure under each condition. If a structure was mastered in one condition but not in the other, training continued with both structures until the mastery criterion was met for both structures. Response Measurement and Interobserver Agreement Trained observers recorded dependent variables with a paper and pencil. The level of prompting (hand over hand, forearm, upper arm, light touch/shadow, or independent [no

PAGE 27

27 prompt]) required to complete each training step was recorded on every trial. Data were summa rized as the number of steps performed independently in each session. A second independent observer collected the same data for a mean of 58.5% of the sessions (range, 42.3% to 70%) for each subject. Interobserver agreement was calculated on a trial by tri al basis. An agreement was scored if both observers recorded the same prompt level required on the same training step for a given trial. A disagreement was scored if observers recorded different prompt levels or different training steps for a given trial. The total number of agreements w as divided by the total number of trials (11 per session) and the quotient was multiplied by 100. The mean percentage agreement across subjects was 97.7% (range, 80% to 100%). Results and Discussion Figure 3 1 shows the num ber of steps mastered across sessions in the Leisure and Edible conditions for Caleb, Henry, Nick, and Ken. Three subjects (Caleb, Henry, and Nick) showed similar rates of acquisition across leisure and edible conditions; that is, the same number of sessio ns was required to master each structure in both conditions for each subject. Caleb mastered both sets of structures in four sessions (40 trials) each. Henry mastered the first set of structures in 6 sessions (60 trials) and the second set in 3 sessions (3 0 trials). Nick mastered the first set of structures in 7 sessions (70 trials). For the second set of structures, Nick mastered the structure in the Edible condition slightly faster (5 sessions; 50 trials) than the structure in the Leisure condition (7 ses sions; 70 trials). Ken showed a more variable pattern of acquisition: H e mastered the first set of structures faster under Leisure conditions (6 sessions compared to 15 sessions under Edible conditions), but mastered the second set of structures faster un der Edible conditions (2 sessions compared to 6 sessions under Leisure conditions). The difference in relative speed of acquisition across sets

PAGE 28

28 build on the right of the structure. For both structures that were mastered more slowly, Step 3 of the task analysis resulted in the most errors (84.2% of errors on probe trials), and this step involved placing a Lego on the left of the structure. By contrast, for the structures mastered faster, Step 3 involved placing a Lego on the right of a structure and relatively fewer errors were made (42.8% of errors on probe trials). towards the right of the structure provides further e vidence that neither reinforcer class was significantly more effective than the other. Thus, despite initial differences in relative preference for leisure and edible items, all subjects showed no difference in the absolute efficacy of reinforcers when use d during acquisition of a response chain. This result is congruous with previous research showing that differences in relative preference were not predictive of differences in absolute efficacy of reinforcers (Roscoe, Iwata, & Kahng, 1999) Likewise, DeLeon et al. (1997) showed that leisure items that had been displaced by edibles in a combined preference assessment were nonetheless effective reinforcers for adaptive behavior; however, rates of responding for less preferred leisure items were never compared to those for more preferred edible items. Thus, results of the current experim ent extend the findings of DeLeon et al. by including a direct comparison between reinforcer classes. Our results are perhaps not surprising given that all items used in the current experiment were first confirmed to be highly preferred during the single c lass ( Leisure only and Edible only) preference assessments of Experiment 1. The current results might also suggest that the instructional procedure ( most to least physical guidance and praise ) was sufficiently effective to mask any difference in reinforce r effectiveness. Inclusion of a condition in which no leisure or edible reinforcers were delivered following

PAGE 29

29 correct responses would have allowed for an evaluation of the necessity of these reinforcers in the acquisition of the chain F uture research shoul d include such a condition to control for the influence of instructional components (i.e., guidance) and non targeted reinforcers (i.e., praise) on correct responding Despite the absence of a control condition, the current results suggest that no differen ce existed between reinforcer classes when used in combination with what many would consider standard instructional techniques.

PAGE 30

30 Table 3 1. Most to least prompt hierarchy Prompt Level Prompt Type 1 Hand over hand guidance 2 2 s delay, forearm guidance 3 2 s delay, upper arm guidance 4 2 s delay, light touch of elbow

PAGE 31

31 Figure 3 1. Number of steps mastered under Leisure and Edible conditions.

PAGE 32

32 CHAPTER 4 EXPERIMENT 3: MAINTE NANCE OF RESPONDING FOR EDIBLE AND L EISURE ITEMS Method Subjects, Setting, and Materials Six individuals (Ken, Nick, Dan, Billy, Caleb, Henry) participated in Experiment 3. These individuals were representat ive of different diagnoses ( Table 2 1) and different patterns of preference ( see Experiment 1 results). Sessions were conducted in the same location as previously described, one to six times per day, three to five days per week. The stimuli necessary to complete a simple target response were present throughout all conditions. Reinforcers were the same HPE and HP L items used in Exp eriment 1 Response Measurement and Interobserver Agreement A simple and arbitrary response, such as pressing a button or putting a chip in a bank, was selected as a target response for each subject to expedite training and focus on maintenance. Target res ponses for each subject are listed in Table 4 1. Observers recorded the frequency of target responses on a handheld PDA and summarized responses as a rate. Each delivery and removal of a reinforcer also was recorded and was used to calculate the total dur ation of reinforcer deliveries. These durations were subtracted from the total session time prior to calculating the rate of responding. Interaction with leisure items was t with a leisure contact with a leisure item for at least 3 s. Data on interaction were summarized as a percentage by dividing the number of seconds of interaction by the total number of seconds presented (15 s for each reinforcement interval) and multiplying by 100. Consumption of edible items was scored when a subject placed the item in his or her mouth and was summarized as the percentage

PAGE 33

33 of edible items consumed. Percen t of interaction and consumption was above 95% across al l sessions and for all subjects, so these data were not graphed. A second observer recorded data on 52.7% of sessions. To calculate interobserver agreement, sessions were divided into 10 s intervals, by interval basis. Interobserver agreement on target responses and consumption (edible condition only) was calculated by dividing the smaller number of responses by the larger number of responses in each interval, summing those quotients, and dividing the sum by the total number of intervals. Scores were converted to percentages by multiplying by 100. Interobserver agreement on reinforcer interaction ( leisure condition only) was scored by dividing the sma ller duration of interaction by the larger and multiplying by 100. The mean agreement across subjects was 90.4% (range, 70.6% to 100%) for target responses. Procedures Baseline During baseline, target stimuli were present throughout the session, but no rei nforcers or feedback were delivered. Baseline was conducted until a steady, low or zero rate of target responding was observed. Schedule training In order to equate reinforcer access time across leisure and edible conditions, subjects were trained to respo nd to a fixed ratio (FR) 10 schedule of reinforcement and reinforcer delivery time was equal to 15 s across maintenance conditions (see below). During schedule training, the experimenter was seated at a table across from the subject. Target stimuli were placed in front of the subject; one type of reinforcer (edible or leisure, depending on the condition) was located behind the target stimuli, visible to the subject. The experimenter modeled the target response (e.g., put a chip in the bank) and then promp ted the

PAGE 34

34 leisure item) was delivered contingent on each target response (FR1). The schedule requirement subsequently was increase d to FR3, FR5, and FR10 following two consecutive reinforcer deliveries at a given schedule ratio. Verbal, model, and physical prompts were used as necessary to assist in schedule thinning, but only unprompted target responses resulted in an increase in the subsequent schedule requirement Following training with one reinforcer type (e.g., edible), training occurred with the other reinforcer type (e.g., leisure ). Training on each reinforcer type was terminated when the subject completed two consecutive FR10 schedule requirements without ex perimenter prompts. Maintenance In the Edible condition, completion of 10 target responses resulted in 15 s access to a single edible item. In the Leisure condition, completion of 10 target responses resulted in 15 s access to a single leisure item. The fo ur HPE and four HPL items were rotated across schedule completions. The materials necessary to engage in a target response were restricted during all reinforcer access periods. If an edible was consumed before the time period elapsed (which occurred in nea rly every access period across all subjects), an empty plate remained in front of the subject for the remainder of the 15 s. The comparison was conducted initially using a multielement design, in which reinforcer conditions were alternated across sessions. Each session was 5 min in duration and four or six sessions were conducted p er day. The multielement design served as a starting point because it provided a rapid way to compare responding across reinforcer conditions. If a clear difference within subjec ts and a consistent difference across subjects emerged, there would be no need for a lengthier comparison. However, if no difference in response rates was observed across conditions in the multielement design, a reversal design was implemented to more clos ely

PAGE 35

35 approximate conditions of repeated exposure to each reinforcer class. Each session was 15 min in duration and two or three sessions of each condition were conducted daily. Results and Discussion Figure 4 1 shows the rate of target responding across Ba seline, Leisure Maintenance, and Edible Maintenance conditions of Experiment 3. All subjects showed little to no responding in baseline, when no consequences were programmed for responding. Following baseline, all subjects showed rapid acquisition of the t arget response during Training (not depicted on graph). All subjects except Ken showed little difference between response rates in the Leisure and Edible Maintenance conditions when session length was short (5 min) and when reinforcer conditions were alte rnated across sessions. Ken engaged in a mean of 22.3 (range; 7.4 to 34.8) responses per minute during Leisure conditions and 39.4 (range; 28.2 to 54.1) responses per minute during Edible re sponding compared to leisure reinforcers. When Leisure and Edible conditions were increased to 15 min and alternated in a reversal design, all remaining subjects except one (Nick) showed higher rates of responding under Edible conditions. Nick engaged in a mean of 4.4 (range, 3.8 to 5.6) responses per minute under Leisure conditions as compared to 4.1 (range, 3.4 to 4.5) responses per minute under Edible conditions. Dan engaged in decreasing rates of responding during the first Leisure condition and increas ing rates during the first Edible condition ; however, these results were not replicated in the second Leisure and Edible conditions and no further replications were permitted because Dan abruptly left the school at which the experiment took place. Billy, Caleb, and Henry all showed some evidence of satiation (decreasing rates across successive sessions) during one or both Leisure conditions but higher and more stable rates of responding during the Edible conditions indicating that edible items were more durable reinforcers for these subjects.

PAGE 36

36 Nonetheless, leisure reinforcers resulted in relatively equal levels of maintenance compared to edibles for most subjects when session length was short and when these reinforcers were alternated across sessions. The se results are similar to those reported by Roscoe et al. (1999), in which high preference and low preference reinforcers resulted in similar levels of responding when presented under a single schedule of reinforcement. Results of the current study howeve r, provide a weaker demonstration of this effect because all reinforcers initially were identified as highly preferred in a single class MS WO assessment. Perhaps a more interesting finding was that responding across the multielement phase showed no decreme nt s for any subject That is, when reinforcers were alternated across sessions, responding was relatively stable over time. A lternating preferred reinforcers in this manner is perhaps an ideal strategy in promoting the durability of a response when short s essions are an option M ore evidence of satiation was detected when session length was increased and repeated exposure to the same reinforcer class was programmed using a reversal design. Even given these manipulations, however, decreases in response rate s were only shown during 5 of the 16 total phases conducted across subjects Perhaps the lack of satiation shown in the current analysis was a function of the rotation of four edible and four leisure items across reinforcer deliveries Previous research ha s shown that the presentation of varied reinforcers results in better response maintenance than the presentation of a single reinforcer (Egel, 1981) The presentation of varied reinforcers in the current study was an attempt to replicate the conditions under which Rincover and Newsom ( 1985) showed important differences in the effectiveness of leisure and edible items. Nevertheless, the current analysis did not replicate these differences but instead showed the opposite effect. When differences in the durability of reinforcers were obtai ned, they favored edible items over leisure items.

PAGE 37

37 Figure 4 1. Rate of target responding across Baseline, Leisure and Edible conditions.

PAGE 38

38 CHAPTER 5 GENERAL DISCUSSION Our results showed no evidence in either subject sample favoring sensory leisure item s in three different contexts: preference, acquisition, and maintenance. Thus, these results directly contradicted the findings of Rincover and Newsom (1985). In addition, the majority of subjects showed exclusive preference (Experiment 1) and more durable maintenance (Experiment 3) for edible items compared to leisure items. Edible items, however, did not have an advantage when used to reinforce correct responding during acquisition of a behavioral chain (Experiment 2). Although consuming edible items is an entirely sensory event, the current results support the distinction between leisure and edible reinforcers in that general disparities in preference and performance were shown across reinforcer classes. However, it may be important to consider the featu res of leisure and edible reinforcers that could contribute to such disparities. In the current study, leisure items were presented to the subject and removed 15 s later. Two features of this arrangement may have biased results in favor of edible stimuli. First, a larger magnitude of reinforcement was perhaps accessed during leisure conditions because the sensory event (i.e., engagement with the leisure item) typically occupied the entire 15 s reinforcer access period. Edible stimuli, by contrast, typicall y were consumed within the first 3 5 s. The potential disparities in magnitude of reinforcement may have contributed to the results of Experiment 3: Satiation effects were perhaps more evident in the leisure condition due to an overall larger magnitude of sensory stimulation per reinforcer delivery compared to the edible condition. This hypothesis could be tested by yoking reinforcer access in the leisure condition to that observed in the edible condition. However, if the items included in the leisure condi tion did in fact provide a larger magnitude of reinforcement compared to items included in the edible condition, results of the preference assessments in Study 1 should have favored leisure

PAGE 39

39 items. Given that our results showed the opposite effect, it remai ns unclear whether a difference in magnitude of reinforcement is an important feature of leisure and edible comparisons. It might be interesting to examine whether increases in the magnitude of sensory (leisure) reinforcement could produce a shift in prefe rence for leisure over edible stimuli. A second distinctive feature of the leisure stimuli used in the current study was the necessity accompanied by its subseque nt removal (following 15 s access) to allow for the continuation of the preference assessment (Experiment 1), training (Experiment 2), or free operant responding (Experiment 3). Although no subjects engaged in problem behaviors when reinforcers were remove d, some subjects engaged in behaviors (e.g., clasping the items more tightly, holding the items under the table) suggesting that reinforcer removal may have had some aversive characteristics (cf. Roane, Vollmer, Ringdahl, & Marcus, 1998). Edible items, by contrast, were naturally consumed and digested, and thus were never physically removed by the therapist. A component analysis of this feature of leisure and edible items might be conducted in future research. Sensory events might be delivered in a more dir ect manner (e.g., flashing lights on a screen), without leisure items as their vehicle, so that the removal of a toy is not required. Alternatively, edible items might be delivered in a more lingering manner (e.g., licking a lollipop), requiring therapist removal. Either of these manipulations would allow for a more thorough analysis of the effect of reinforcer removal on sub ject preference and performance The findings reported by Rincover and Newsom (1985), which suggested the superiority of sensory leis ure items and events for autistic children, remain unexplained. In addition, it is yet unclear how the results of that study may be reconciled with the outcomes of studies on the assessment of preference showing that individuals with disabilities generally prefer edibles. We

PAGE 40

40 attempted to test three hypotheses regarding this discrepancy. First, we determined whether highly preferred edibles were more prone to satiation than less preferred leisure items. Our results showed contrary evidence, in that edibles m aintained responding better than did leisure items for the majority of subjects. Although we did not attempt to control for establishing operations, the amount of noncontingent reinforcement (edible or leisure ) available to our subjects throughout their ty pical day perhaps differed from that experienced by the subjects in the study conduct ed by Rincover and Newsom That is, perhaps satiation occurred more quickly with leisure items in the current study because our subjects were in classrooms enriched with m ultiple sources of sensory stimulation (frequent leisure activities and academic periods). Environmental enrichment has become an important component of services for individuals with disabilities over the past three decades, so perhaps the environment expe rienced by our subjects contained more of these sources of noncontingent leisure activity as compared to the environment experienced by the subjects in the study conducted by Rincover and Newsom. Second, we sought to test the hypothesis that the results of the study conducted by Rincover and Newsom (1985) were influenced by subject preferences (i.e., subjects in that study happened to prefer leisure items). However, none of the subjects who participated in Experiment 1 showed this pattern of preference, and so whether or not subjects showing this pattern also would show more accurate and durable responding for leisure items remains unknown. Finally we included individuals both with and without autism in the current studies to determine whether subject diagn osis influenced the preference for, or effectiveness of, leisure and edible items. We found no noticeable differences between the outcomes for students with versus without autism across all analyses. However, other subject characteristics that were not mea sured in the current studies may have influenced the results obtained by Rincover and

PAGE 41

41 Newsom (1985) For example, Rincover and Newsom noted that each of their subjects engaged in high levels of stereotypic behavior, though no data were presented on stereot ypy. Although some of our subjects (Caleb, Elliot, Carl, Mark, Martin, Dan, and Billy) engaged in stereotypy, all did so at low to moderate levels. Several possible modifications to the current procedures may have made a replication of Rincover and Newsom (1985) more tenable. First, the edible items selected for inclusion in the Combined assessment of Experiment 1 could have been those selected on the fewest percentage of trials during the Edible only assessment (i.e., low preference edibles) as opposed to the largest percentage of trials (i.e., high preference edibles). This manipulation would have made it more likely that the high preference leisure items would have retained higher rankings during the Combined assessments, and thus, this manipulation may h ave resulted in overall preference for leisure over edible items. Given this outcome, a comparison of highly preferred leisure and less preferred edible items during acquisition and maintenance could have been accomplished. Results in favor of leisure item s also may have been obtained by selecting subjects with highly idiosyncratic and rigid leisure preferences (e.g., those who have a favorite toy with them at all times) and by including those items in the assessments of Experiment 1. It should be noted, h owever, that either of these strategies would have biased results in favor of leisure items. Despite our inability to reconcile the findings of previous research (Bojak & Carr, 1999; DeLeon et al., 1997; Rincover & Newsom, 1985), the current study provides implications for effective practice. Edible items represent a generally preferred class of reinforcers that maintain responding somewhat longer than do leisure items. These advantages provide evidence in favor of including edible items in acquisition trai ning. O ne additional advantage of edible reinforcers is that they are delivered more easily and efficiently, which may have an overall beneficial effect

PAGE 42

42 on training in the long run. Some researchers (e.g., Ferrari & Harris, 1981; Rincover & Newsom) have em phasized the relative benefits of leisure stimuli in behavioral training, stating that leisure stimuli are more natural reinforcers, less deleterious to health, and require more active engagement. However, the cost of using less preferred and less durable leisure reinforcers should be considered in light of these proposed benefits. If behavioral acquisition is an important goal, and if gains are expected to be maintained over a long period of time, edible items should be included in preliminary preference a ssessments. Nevertheless, leisure items may be preferred for use with individuals with strict dietary restrictions or for settings in which a budget does not allow for the continued purchase of edibles. Results of these studies showed that leisure items, a l though not generally as preferred as edibles, were equally effective reinforcers for response acquisition (when paired with most to least guidance and praise) and maintained responding as well as edibles when session length was short. Thus, the inclusion of leisure items in training programs is only encouraged when the use of edibles is not convenient.

PAGE 43

43 LIST OF REFERENCES Bailey, J., & Meyerson, L. (1969). Vibration as a reinforcer with a profoundly retarded child. Journal of Applied Behavior Analysis, 2 135 137. Barton, E. S. (1970). Inappropriate speech in a severely retarded child: a case study in language conditioning and generalization. Journal of Applied Behavior Analysis, 3 299 307. Bojak, S. L., & Carr, J. E. (1999). On the displacement of leisur e items by food during multiple stimulus preference assessments. Journal of Applied Behavior Analysis, 32 515 518. DeLeon, I. G., & Iwata, B. A. (1996). Evaluation of a multiple stimulus presentation format for assessing reinforcer preferences. Journal of Applied Behavior Analysis, 29 519 533. DeLeon, I. G., Iwata, B. A., & Roscoe, E. M. (1997). Displacement of leisure reinforcers by food during preference assessments. Journal of Applied Behavior Analysis, 30 475 484. Egel, A. (1981). Reinforcer variatio n: implications for motivating developmentally disabled children. Journal of Applied Behavior Analysis, 14 345. Fehr, M. (1979). Visual, auditory, and vibratory stimulation as reinforcers for profoundly retarded children. Rehabilitation Psychology, 26 20 1 209. Ferrari, M., & Harris, S. L. (1981). The limits and motivating potential of sensory stimuli as reinforcers for autistic children. Journal of Applied Behavior Analysis, 14 339 343. Fineman, K. R. (1968). Shaping and increasing verbalizations in an a utistic child in response to visual color stimulation. Perceptual and Motor Skills, 27 1071 1074. Fineman, K. R., & Ferjo, J. (1969). Establishing and increasing verbalizations in a deaf schizophrenic child through the use of contingent visual color reinf orcement. Perceptual and Motor Skills 29, 647 652. Fuller, P. R. (1949). Operant conditioning of a vegetative human organism. The American Journal of Psychology, 62 587 590. Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of prob lem behavior: A review. Journal of Applied Behavior Analysis, 36 147 186. Hanley, G. P., Iwata, B. A., & Roscoe, E. M. (2006). Some determinants of changes in preference over time. Journal of Applied Behavior Analysis, 39 189 202. Hanley, G. P., Iwata, B A., Thompson, R. H., & Lindberg, J. S. (2000). A component analysis of "stereotypy as reinforcement" for alternative behavior. Journal of Applied Behavior Analysis, 33 285 297. Kish, G. B. (1955). Learning when the onset of illumination is used as the r einforcing stimulus. Journal of Comparative and Physiological Psychology, 48 261.

PAGE 44

44 Kish, G. B. (1966). Studies of sensory reinforcement. Operant behavior: areas of research and application. New York: Appleton Century Crofts, 76 369 377. Kish, G. B., & Bar on, A. (1962). Satiation of sensory reinforcement. Journal of Comparative and Physiological Psychology, 55 1007. Lalli, J. S., Casey, S., & Kates, K. (1995). Reducing escape behavior and increasing task completion with functional communication training, e xtinction, and response chaining. Journal of Applied Behavior Analysis, 28 261 268. 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 individual s. Journal of Applied Behavior Analysis, 18 249 255. Rehagen, N., & Thelen, M. (1972). Vibration as positive reinforcement for retarded children. Journal of Abnormal Psychology, 80 162 167. Rice, H. K., & McDaniel, M. W. (1966). Operant behavior in veget ative patients. The Psychological Record 16, 279 281. Rice, H. K., McDaniel, M. W., & Stallings, V. D. (1967). Operant Behavior in vegetative patients: II. The Psychological Record 17, 449 460. Rincover, A., Cook, R., Peoples, A., & Packard, D. (1979). S ensory extinction and sensory reinforcement principles for programming multiple adaptive behavior change. Journal of Applied Behavior Analysis, 12 221 233. Rincover, A., & Newsom, C. D. (1985). The relative motivational properties of sensory and edible re inforcers in teaching autistic children. Journal of Applied Behavior Analysis, 18 237 248. Roane, H. S., Vollmer, T. R., Ringdahl, J. E., & Marcus, B. A. (1998). Evaluation of a brief stimulus preference assessment. Journal of Applied Behavior Analysis, 3 1, 605 620. Rogers, S. J., & Ozonoff, S. (2005). Annotation: What do we know about sensory dysfunction in autism? A critical review of the empirical evidence. [Review]. Journal of Child Psychology and Psychiatry, 46 1255 1268. Roscoe, E. M., Iwata, B. A., & Kahng, S. (1999). Relative versus absolute reinforcement effects: Implications for preference assessments. Journal of Applied Behavior Analysis, 32 479 493. Vollmer, T. R., & Iwata, B. A. (1991). Establishing operations and reinforcement effects. Jour nal of Applied Behavior Analysis, 24 279 291. Whitman, T. L., Zakaras, M., & Chardos, S. (1971). Effects of reinforcement and guidance procedures on instruction following behavior of severely retarded children. Journal of Applied Behavior Analysis, 4 283 290.

PAGE 45

45 Wolf, M. M., Risley, T., & Mees, H. (1964). Application of operant conditioning procedures to the behavior problems of an autistic child. Behaviour Research and Therapy, 1 305 312.

PAGE 46

46 BIOGRAPHICAL SKETCH Tara Fahmie initially studied behavior analys is at the University of Florida (UF), where behavioral science from the Universit y of Kansas the subsequent year under the supervision of Dr. Gregory Hanley. Upon gradua tion, she entered the applied behavior analysis PhD program at UF under the supervision of Dr. Brian Iwata. While at UF, she participated in a variety of clinical services, including the provision of behavior services to adults with Prader Willi Syndrome a t the ARC of Alachua. Tara also was involved in several research projects on the assessment and treatment of behavior disorders, skill acquisition, and the prevention of problem behavior. She has been a teaching assistant and an instructor for introductory and advanced undergraduate courses in applied behavior analysis at UF. Following graduation, Tara intends to pursue an academic and research career in applied behavior analysis.