<%BANNER%>

Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2010-08-31.

Permanent Link: http://ufdc.ufl.edu/UFE0022591/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2010-08-31.
Physical Description: Book
Language: english
Creator: Bloom, Sarah
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: 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

Statement of Responsibility: by Sarah Bloom.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Iwata, Brian.
Electronic Access: INACCESSIBLE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022591:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022591/00001

Material Information

Title: Record for a UF thesis. Title & abstract won't display until thesis is accessible after 2010-08-31.
Physical Description: Book
Language: english
Creator: Bloom, Sarah
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: 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

Statement of Responsibility: by Sarah Bloom.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Iwata, Brian.
Electronic Access: INACCESSIBLE UNTIL 2010-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022591:00001


This item has the following downloads:


Full Text





ASSESSMENT OF PREFERENCE FOR OLFACTORY STIMULI IN INDIVIDUALS WITH
PRADER-WILLI SYNDROME





















By

SARAH E. BLOOM


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008



































2008 Sarah E. Bloom


































To my dogs, Enzo and Crosby, for whom olfactory stimuli are very important.









ACKNOWLEDGMENTS

I thank Erin Camp, Natalie Rolider, Javier Virues-Ortega, Jennifer Hammond, and Leah

Koehler for help with data collection. I thank Dr. Brian Iwata for his guidance and tireless

assistance. I also thank Dr. Timothy Hackenberg, Dr. Hazel Jones, Dr. Neil Rowland and Dr.

Timothy Vollmer.










TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ..............................................................................................................4

LIST OF TABLES ..................... ............................ ........7

LIST OF FIGURES ................................... ............ .. .............................8

ABSTRACT ............................ .......................................................9

CHAPTER

1 INTRODU CTION ............... ........................ ............................ ...............

Prader-W illi Syndrom e ................... ................ .................................. ...............
Identification of Reinforcers in Individuals with Developmental Disabilities.....................13
Olfactory Stimuli .......................... ........................ 15
Goals of Current Research.................................................................. ......... 17

2 EXPERIMENT 1 ....................................... .. ....................18

Purpose .............................................................................18
Method ...................................... ........................................... .........18
Subjects and Setting .................................................18
Response M easurement and Reliability ............................... ............... 19
Preference Assessm ent .................. ............................. ....... ........ ... ................ 19
Results.............. ................... ............................. .............................. ........ 21
Discussion .............. ................... ............................................................... 23

3 EXPERIM EN T 2 ............... ...................... ............................... .. .............31

Purpose .............................................................................31
Method ...................................... ........................................... .........31
Subjects and Setting ............... .......... ...................... 31
Task Selection and Initial Training ........................................ ................. 32
Response M easurement and Reliability ............................... ...............32
Single-Schedule Reinforcer Assessment.................. .................................................33
Concurrent-Schedule Reinforcer Assessment .........................................................34
Results.............. ................... ............................. .............................. ........ 34
D iscu ssio n ................... .......................................................................................... . 3 5

4 G EN ER A L D ISCU SSIO N ............................................................................................... 41

APPENDIX: STIMULI USED IN PREFERENCE ASSESSMENTS DURING
EXPERIMENT 1 AND PREFERENCE RANKINGS FOR EACH STIMULUS.................46









L IST O F R E F E R E N C E S ....................................................................................................55

BIO GR A PH ICA L SK ETCH ................................................................................................ ........ 59





















































6









LIST OF TABLES

Table page

2-1. Experiment 1 demographic information................................... ......... 27

3-1. Reinforcer assessment type and tasks..................... ........ ...............38

A-1. Group 1: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................46

A-2. Group 2: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................48

A-3. Group 3: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................49

A-4. Group 4: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................50

A-5. Group 5: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings ........................................................5 1

A-6. Group 6: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................52

A-7. Group 7: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................52

A-8. Group 8: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................53

A-9. Group 9: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................53

A-10. Group 10: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................54

A-11. Group 11: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings .......................................................54









LIST OF FIGURES


Figure page

2-1 Percentage of trials on which each stimulus was selected during each of the paired-
stim ulus preference assessm ents.............................................................. ..............28

2-2 Mean preference rankings of stimulus classes for all PWS subjects from each of the
paired-stimulus preference assessments ................................ ............... 29

2-3 Results obtained for all non-PWS subjects from each of the paired-stimulus
preference assessments ................................................. ........ 30

3-1 Results of the single-schedule reinforcer assessment for Bettina, Taneisha, Peter,
D rake, Z eus, and A decline. ............................................................39

3-2 Results of the concurrent-schedule reinforcer assessment for Bettina, Annette, Patsy,
and Taneisha. ................................................ .........40









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

ASSESSMENT OF PREFERENCE FOR OLFACTORY STIMULI IN INDIVIDUALS WITH
PRADER-WILLI SYNDROME

By

Sarah E. Bloom

August 2008

Chair: Brian Iwata
Major: Psychology

We examined the reinforcing effects of olfactory stimuli in individuals with Prader-Willi

Syndrome (PWS). In Study 1, 30 individuals (20 diagnosed with PWS, 10 diagnosed with a

developmental disability other than PWS) were exposed to a series of three paired-stimulus

preference assessments involving edible, olfactory, leisure stimuli. In each assessment, 7 or 8

items representing 2 stimulus groups were included. Stimulus class hierarchies were generated

based on the results of these preference assessments. Nineteen of the 20 PWS subjects had edible

stimuli at the top of their preference hierarchy (the remaining subject showed mixed preference

for edible and leisure stimuli). Eight subjects demonstrated a stronger preference for leisure

stimuli than for olfactory stimuli, and 10 subjects demonstrated a stronger preference for

olfactory stimuli than for leisure stimuli. The remaining 2 subjects showed mixed preference for

leisure and olfactory stimuli. Preferences were similar for subjects without PWS. A subset of the

subjects from Study 1 was included in Study 2, in which preferences were verified in either

concurrent or single-schedule reinforcer assessments. Four subjects for whom olfactory stimuli

were the least preferred were exposed to a single-schedule reinforcer assessment in which

performance of a task resulted in access to olfactory stimuli. Increases in responding, relative to

baseline, were observed in 3 subjects but not the fourth. The fourth subject's responding only









increased when responses were reinforced with edible stimuli. Two subjects for whom olfactory

stimuli were ranked below edible but above leisure, or for whom preference for olfactory stimuli

and leisure stimuli was mixed, also were exposed to a single-schedule reinforcer assessment and

then were exposed to a concurrent-schedule reinforcer assessment in which performance of one

task resulted in access to olfactory stimuli and performance of another (identical) task resulted in

access to leisure stimuli. Two additional subjects only were exposed to the concurrent-schedule

reinforcer assessment. Increases in responding during the single-schedule reinforcer assessment

were observed. All four subjects allocated responses for both choices (olfactory and leisure) at

approximately equivalent rates.These findings suggest that olfactory stimuli may be a viable

class of reinforcers for some individuals with PWS or for others for whom edible reinforcers are

contraindicated.









CHAPTER 1
INTRODUCTION

Prader-Willi Syndrome

Prader-Willi Syndrome (PWS) is a genetic disorder with an estimated incidence in the

United States ranging from 1 in 16,000 (Burd, Vesely, Martsolf, & Kerbeshian, 1990) to 1 in

25,000 (Butler, 1996). Physical features associated by PWS include short stature, small hands

and feet, poor muscle tone, and obesity (Butler, Meaney, Opitz, & Reynolds, 1987). In addition,

some degree of cognitive impairment and behavior problems, including self-injury, sleep-

abnormalities, and disruptive behavior, have been observed (Clarke, Boer, & Webb, 1995).

Individuals with PWS are typically obese due to excessive eating or hyperphagia (Greenswag,

1987). Attempts to control this obesity often include restricted diets, thus eliminating food items

from the pool of stimuli that may be used as reinforcers. It may be possible that olfactory stimuli,

which share sensory properties with food, may effectively be used as reinforcers with this

population. The general purpose of the current study is to examine the relative efficacy of

olfactory stimuli to function as reinforcers.

Prader-Willi Syndrome (PWS) results most frequently from paternal deletion of band

15ql1-13 on chromosome 15 (Olander, Stamberg, Steinberg, & Wulfsberg, 2000). One reported

potential contributor to obesity is reduced resting energy expenditure, possibly as a result of

hypotonia (Butler, Theodoro, Bittel, & Donnelly, 2007). A more direct contributor to obesity is

excessive consumption of food (hyperphagia) and is a serious medical concern (Gunay-Aygon,

Cassidy, & Nicholls, 1997). When typical food sources are restricted, individuals with PWS have

been observed eating garbage or other inedible items (Russell & Oliver, 2003). Overeating was

documented in 98% of individuals with PWS in a sample of 91 children and adults with PWS

(Dykens & Shah, 2003). Morbid obesity has been documented with related coronary disease









and/or non-insulin dependent diabetes mellitus in individuals with PWS (Lamb & Johnson,

1987). Although infants with PWS are often diagnosed with feeding problems (Gunay-Aygun,

Schwartz, Heeger, O'Riordan, & Cassiday, 2001) eating increases exponentially and eventually

qualifies as hyperphagia, which worsens across the life-span of individuals with PWS (Akefeldt

& Gillberg, 1999). Hyperphagia can lead to fatal gastric necrosis following severe gastric

distention (Wharton, Wang, Graeme-Cook, Briggs, & Cole, 1997). In a survey documenting

cause of death in individuals with PWS, choking on food was implicated in 7.9% of the 152

individuals (Stevenson et al., 2007).

Problems related to food consumption are implicated in a variety of challenging behaviors

observed in individuals with PWS. Excessive food consumption and food-seeking/food-stealing

behaviors are commonly targeted in behavioral interventions (Maglieri, DeLeon, Rodriguez-

Catter, & Sevin, 2000; Page, Finney, Parrish, & Iwata, 1983; Page, Stanley, Richman, Deal, &

Iwata, 1983). Maglieri et al. paired punishment (a verbal reprimand) with warning stimuli,

which resulted in a decrease in consumption of prohibited foods (2000). Reinforcement for not

stealing food has also been examined (Page, Finney et al., 1983), as have interventions that

combine differential reinforcement and punishment (Page, Stanley et al., 1983).

Because individuals with PWS eat to the point of dangerous excess, they often are placed

on extremely restricted dietary regimens (Goldberg, Garrett, Van Piper, & Warzac, 2002).

Unlike diet programs that merely prescribe limited caloric intake, dietary management in

individuals with PWS includes restriction of opportunities to consume food. The nature of these

restrictions can be rather extreme, including not permitting the person with PWS to be alone in

the community in the presence of food and eliminating access to food in the home environment.

Families of children with PWS are counseled to put padlocks on pantries and refrigerators and to









restrict access to all potential food sources in order to keep calories consumed to approximately

1200 per day (Dykens & Shah, 2003). Group home placements for individuals with PWS include

similar restrictive precautions. The quantity of food consumed by individuals with PWS is

restricted and carefully monitored. Finally, although there is ample evidence that edible items are

highly preferred by individuals with PWS, food is not recommended for use as a reinforcer in

training or treatment programs due to its potential to undermine restricted-calorie diets. This

presents challenges in establishing and maintaining compliance with other therapeutic

interventions, such as exercise, menu planning, and academic or work performance because an

entire yet presumably potent class of reinforcers is prohibited. Classes of stimuli other than food

would be less problematic, if they can function as reinforcers.

Identification of Reinforcers in Individuals with Developmental Disabilities

Inclusion of effective reinforcers in training programs is essential to the acquisition and

maintenance of adaptive behavior. Three general approaches used to identify reinforcers in

individuals with developmental disabilities include the interview, direct observation, and

systematic assessment of preference. The interview involves simply asking individuals or their

care providers what stimuli are preferred. Although easy to administer, interviews have been

shown to be unreliable or inaccurate in identifying effective reinforcers (Reid, Everson, & Green,

1999). Direct observation typically attempts to identify how individuals spend their free time.

Data are recorded on duration of involvement in various activities and are used as the basis for

selecting some activities as reinforcers for engaging in other (target) activities. This method is

more objective than the interview but potentially very time consuming. Also, individuals may

have very limited access to potential reinforcers in their environments, thus restricting the pool

of stimuli included in the assessment. Due to limitations with both the interview and direct-

observation methods, most researchers have adopted more systematic procedures in which an









array of stimuli (in various formats) is systematically presented to the subject, and data are

collected on some measure of preference (either approach or selection).

Pace, Ivancic, Edwards, Iwata, and Page (1985) presented a two-stage process for

identifying reinforcers: an initial assessment of preference followed by a test for reinforcer

efficacy. They first presented a series of stimuli one at a time (single-stimulus presentation) and

measured approach behavior. Sixteen stimuli, 2 from each of 8 classes (visual, auditory,

olfactory, edible, tactile, thermal, kinesthetic, and social) were presented ten times each. They

then used stimuli approached on high vs. low percentages of assessment trials as consequences

for performance on learning tasks and observed that stimuli identified as more highly preferred

during the preference assessment functioned as more effective reinforcers.

Subsequent research on reinforcer identification has adopted the two-stage process

illustrated by Pace et al. (1985) and has included a number of procedural variations (see Ivancic,

2000, for a review). For example, stimulus presentation during preference assessments has

included the paired-stimulus variation (Fisher et al., 1992), in which a selection is made from 2

stimuli presented during each trial, as well as multiple-stimulus formats (DeLeon & Iwata, 1996;

(Windsor, Piche, & Locke, 1994), in which selections are made from larger arrays of stimuli.

Another variation has involved the mode of stimulus presentation. Most preference assessments

involve presentation of the actual stimulus to be used as a reinforcer (i.e., an edible or leisure

item). However, pictorial representations of stimuli also have been used occasionally (Hanley,

Iwata, & Lindberg, 1999; Northrup, George, Jones, Broussard, & Vollmer, 1996). Finally,

responding has been measured as approach behavior during stimulus presentation in discrete

trials, as in each of the above examples, or free-operant engagement when all stimuli are









available throughout the session (Hanley, Iwata, Lindberg, & Conners, 2003; Roane, Vollmer,

Ringdahl, & Marcus, 1998).

Tests for reinforcement effects (the second stage of reinforcer identification) also have

varied along a number of dimensions. The two most commonly used procedures are those in

which the effects of a particular stimulus are evaluated by presenting it following a target

response (single schedule; Pace et al., 1985) and those in which the effects of two competing

stimuli are evaluated by presenting one stimulus for each of two concurrently available target

responses (concurrent schedule; Fisher et al., 1992).

One finding in research on preference assessment, which is relevant to the current studies

although perhaps not surprising, is that some classes of stimuli seem to be preferred over others.

In an early example, Rincover and Newsom (1985) found that "sensory" reinforcers (objects that

provided sensory stimulation) maintained responding better than did food reinforcers in

participants who were diagnosed with autism. By contrast, results of subsequent research with

participants having developmental disabilities other than autism have shown a strong preference

for edible items. For example, DeLeon, Iwata, and Roscoe (1997) first identified highly

preferred edible items in an assessment that included only edible items and then identified highly

preferred leisure items in an assessment that included only leisure items. They then combined the

top-ranked stimuli from the edible-only and leisure-only assessments in a third preference

assessment and observed that the edible items routinely outranked the leisure items, a finding

that was replicated by Bojak and Carr (1999).

Olfactory Stimuli

Although not often the focus of research in behavior analysis, olfactory stimuli are an

important part of the lives of humans and other animals. Responses to olfactory stimuli are broad

and dramatic in range, from strong preference for particular stimuli to acute aversions to others.









Olfactory stimuli have an important role across the lifespan. For example, neonates have

consistent preferences for natural breast odors and select unwashed breasts over washed breasts

(Varendi, 1997). Thus, olfactory stimuli may have central roles in the regulation of diverse and

critical behavioral repertoires.

Olfactory stimuli have been said to evoke pleasant memories (Jellinek, 2004; Mohr, 2001)

and have been implicated in animal and insect mate selection (Bhutta, 2007; Buck & Axel, 1991;

Mombaerts, 2004). Humans may even engage in behavior to access olfactory stimuli associated

with their partners during periods of separation (McBurney, 2006). Food-related olfactory

stimuli have been implicated in operant paradigms involving dopamine reinforcement of food-

seeking behavior (Wise, 2006) and appetitive behavior in general (Aime, 2007; Yeomans, 2006).

In classical conditioning paradigms, exposure to olfactory stimuli has increased neural

responsiveness to conditioned odors (Davis, 2004). In an interesting study conducted with school

children, classical conditioning with an ambient odor and academic success increased academic

performance on other tasks when children were exposed to the same odor (Chu, 2008).

Olfactory stimuli share sensory characteristics with food, and both the gustatory and

olfactory systems are stimulated when exposed to edible and/or olfactory stimuli (Kettenmann,

2005; Verhagen & Engelen, 2006). This relationship suggests that olfactory stimuli may

represent a stimulus class especially effective for use with individuals with PWS. Previous

research on preference has included some olfactory stimuli such as coffee grounds and hibiscus

flower (Fisher et al., 1992; Pace et al., 1985), but the general efficacy of olfactory stimuli as

reinforcers has been directly assessed only recently in a study on individuals with autism (Wilder

et al., 2008).









Goals of Current Research

Though it may be assumed that individuals with PWS would show a general preference for

edible stimuli over others, it is unknown how olfactory stimuli might fare if pitted against edible

and/or leisure stimuli. It is possible that olfactory stimuli may effectively compete with leisure

stimuli. Regardless of relative preference for edible, leisure, or olfactory stimuli, it is possible

that olfactory stimuli may function as reinforcers, especially in populations that value food

highly, like those diagnosed with PWS. Given that little is known about preference for olfactory

stimuli or about the efficacy of olfactory stimuli as reinforcers, further research about the

potential role of olfactory stimuli in behavioral acquisition is warranted.

The present study examined preferences in individuals with Prader-Willi Syndrome. The

purposes were to determine (a) relative preference for edible, leisure, and olfactory stimuli, and

(b) whether olfactory stimuli would function as reinforcers for this population. An additional

purpose was to examine relative preference for edible, leisure, and olfactory stimuli in

individuals who did not have Prader-Willi syndrome









CHAPTER 2
EXPERIMENT 1

Purpose

The purpose of this study was to examine preference for edible, leisure, and olfactory

stimuli in individuals with PWS. We replicated previous studies that compared preference for

edible versus leisure stimuli (Bojak & Carr, 1999; DeLeon et al., 1997) and extended that work

by including a separate assessment of olfactory stimuli using a wider range of stimuli than has

been examined previously. We also compared preference for the three classes of stimuli in

individuals who have PWS with those who do not.

Method

Subjects and Setting

Twenty individuals with PWS and 10 individuals with a developmental disability other

than PWS served as subjects. All of the PWS subjects were on restricted diets as part of their

medical and behavior programs. Both children and adults were included, and their level of

functioning ranged from severe to mild mental retardation. Subject demographic information is

listed in Table 2-1. Subjects were selected for inclusion based on their PWS or non-PWS status,

availability, absence of food refusal or feeding problems, and absence of problem behavior likely

to interfere with sessions. Sessions were conducted at a day activity program for individuals with

developmental disabilities or at a school for children with exceptional needs. Subjects run at the

day activity program included 5 individuals with a developmental disability other than PWS

(Juan, Wanda, Mitchell, Jeffrey, and Mikey), and 17 subjects with PWS (Patsy, Trina, Annette,

Bettina, Penelope, Misty, Jewel, Jacob, Taneisha, Jamal, Dorian, Drake, Zeus, Peter, Adeline,

Barnaby, and Melissa). Subjects run at the school for children with exceptional needs included 5

individuals with a developmental disability other than PWS (Geoff, Montana, Cindy, Ernie, and









Leyla), and 3 subjects with PWS (Janet, Katarina, and Manuel). PWS subjects experienced

calorie-restricted diets and controlled access to food in both settings. Foods consumed as part of

this project were not included in the calorie plans for any subject. Sessions were conducted two

to three times per day, four to five days per week.

Response Measurement and Reliability

Observers recorded the occurrence of approach responses to one of two items presented

during preference assessments. A percentage selection score was given to each stimulus by

dividing the number of trials on which a stimulus was approached by the number of trials on

which it was present and multiplying by 100%

An independent observer collected data during a mean of 73.7% of all preference

assessments (range, 26.4% to 100%). Observer's records were compared on a trial-by-trial basis,

and an agreement was scored if the two observers marked the same item as selected by the

subject on a given trial. Interobserver agreement was calculated by dividing the number of

agreements by the total number of trials and multiplying by 100%. The mean agreement score

across all participants was 98.8% (range, 96.0% to 100%) during the preference assessments.

Preference Assessment

Three paired-stimulus preference assessments (Fisher et al., 1992) were conducted across

several days. The purpose of the assessments was to identify relative preference for stimulus

classes. Paired- stimulus preference assessments are better suited to determine relative preference

rather than absolute preference because a preference hierarchy is established based on a series of

selections in which each stimulus is pitted against every other stimulus (Fisher et al., 1992). Each

assessment involved the presentation of two classes of stimuli. One assessment included edible

and leisure stimuli, another included edible and olfactory stimuli, and another included leisure

and olfactory stimuli. The order of the preference assessments was not consistent across subjects.









Each stimulus class consisted of 7 or 8 stimuli (an equal number of stimuli from each class was

included in each assessment). Edible stimuli and olfactory stimuli were paired in flavor. For

example, if a mint-flavored candy was included in the edible stimulus array, a mint fragrance

was included in the olfactory array. Leisure stimuli included activities that were reported as

preferred by the subjects' care provider and were age-appropriate or appropriate to the level of

functioning of each subject. Leisure stimuli were not consistent across subjects.

Prior to each assessment, subjects were allowed access to each of the items. Access

consisted of tasting each edible stimulus and interacting with each leisure and olfactory stimulus

for 30-s. During each trial, two items were presented, and the subject was allowed to select one

item. Approach responses were defined as the subject reaching for, touching, or picking up an

item within 5 s of a verbal prompt to "Pick one." If the subject attempted to select both of the

items, the items were removed and re-presented. Selection of an edible stimulus resulted in

delivery of that stimulus; selection of an olfactory stimulus or a leisure stimulus resulted in 30-s

access to the chosen stimulus. Olfactory stimuli were delivered by allowing the subject to hold to

their nose a glass spice jar with a sealed top with small holes or a film canister with a cotton ball

soaked with aromatic/essential oil or with scented-candle shavings inside. Subjects did not

always hold the olfactory stimulus to their nose for the entire 30-s access period. Some subjects

held it to their nose and then put it back down on the table and waited for the access period to

end. Other subjects alternated between holding it to their nose and holding it about a foot away

from their nose for the 30-s access period. A few subjects held it to their nose for the entire 30-s

access period.









Trials were arranged by assigning each stimulus a number and then using those numbers to

generate a random sequence of number pairs. The number pairs were used to order trials. No

single stimulus was presented consecutively.

Results

Subjects with PWS almost exclusively preferred edible stimuli to both leisure and

olfactory stimuli. Beyond their shared preference for edible stimuli, three general patterns of

responding were observed: (a) preference for leisure over olfactory stimuli, (b) preference for

olfactory over leisure stimuli, and (c) mixed preference for olfactory and leisure stimuli. Figure

2-1 shows results of the preference assessments for Peter, Annette, and Taneisha, 3 subjects

whose responding illustrated each of these patterns. Data are expressed as the percentage of trials

on which each stimulus was selected. Peter showed exclusive preference for edible stimuli in that

he selected every edible stimulus more often than every leisure stimulus (top left) and every

olfactory stimulus (middle left). In the assessment of leisure vs. olfactory stimuli (bottom left),

he showed a strong preference for leisure stimuli: The four stimuli he selected most frequently

(slinky, bear, bumble ball and workbook) were leisure items, and the only leisure stimulus

selected less frequently than olfactory stimuli was the puzzle, which was chosen less than any

other stimulus.

Annette's results are shown in the center panels. Like Peter, she preferred all edible stimuli

over all leisure stimuli (top center) and all olfactory stimuli (middle center) Unlike Peter,

however, Annette showed a strong preference for leisure stimuli over olfactory stimuli (bottom

center) in that every olfactory stimulus outranked every leisure stimulus.

Taneisha's results are shown in the right panels. Her preference for edible stimuli over

leisure stimuli (top right) was evident but not exclusive because there were two exceptions. Her

highest ranked stimulus was puzzle, a leisure stimulus, and workbook, another leisure stimulus,









outranked 3 edible stimuli. She showed exclusive preference for edible stimuli over olfactory

stimuli (middle right), although the bottom ranked edible stimulus (cinnamon altoid) tied the top

ranked olfactory stimulus (honey nut). Taneisha's preference for olfactory versus leisure stimuli

(bottom right) was mixed. The top 2 stimuli were leisure stimuli (puzzle and workbook), but a

leisure stimulus (beads) and an olfactory stimulus (strawberry) tied for the third spot. The

remainder of her selections showed a slight preference for olfactory stimuli.

Based on results (percentage selection scores) obtained from the three preference

assessments, a numerical rank was assigned to each stimulus in each assessment. The item

selected most often was given a rank of 1, the item selected second most often was given a rank

of 2, and so on. If two or more items were selected equally often, all were given the same rank

obtained by adding the ranks for which there were a tie and dividing by the number of tied items.

A mean rank for each stimulus class was obtained by adding the individual rankings for stimuli

in that class and dividing by the number of stimuli of that class that were included in the

assessment (either 7 or 8). The individual ranks for each stimulus in each assessment are listed in

Appendix A.

Figure 2-2 shows results obtained for the 20 PWS subjects, expressed as the mean

preference rank for each stimulus class in each assessment. Nineteen of the 20 PWS subjects

preferred edible stimuli to leisure stimuli (left panel). The exception, Dorian, showed mixed

preference for edible and leisure stimuli. All 20 subjects preferred edible stimuli to olfactory

stimuli (middle panel). Less consistent preferences across subjects were observed in the leisure

vs. olfactory assessment (right panel): Ten subjects (Janet, Katarina, Patsy, Trina, Annette,

Bettina, Penelope, Misty, Jewel, and Jacob) preferred olfactory stimuli, 2 subjects (Taneisha and









Jamal) had mixed preference for olfactory and leisure stimuli, and 8 subjects (Dorian, Manuel,

Drake, Zeus, Peter, Adeline, Barnaby, and Melissa) preferred leisure stimuli.

Figure 2-3 shows mean preference rankings for the non-PWS subjects, which reflected

preferences similar to those obtained with the PWS subjects. One of the 10 non-PWS subjects

(Geoff) preferred leisure stimuli over edible stimuli (left panel); the remaining 9 subjects

(Montana, Cindy, Ernie, Juan, Leyla, Wanda, Mitchell, Jeffrey, and Mikey) preferred edible

stimuli. Seven of the 10 subjects (Montana, Cindy, Geoff, Ernie, Wanda, Jeffrey, and Mikey)

preferred edible stimuli to olfactory stimuli (middle panel); 1 subject (Juan) preferred olfactory

stimuli; the remaining 2 subjects (Leyla and Mitchell) showed mixed preference for edible and

olfactory stimuli. Finally, 6 subjects (Montana, Cindy, Juan, Leyla, Wanda, and Mitchell)

preferred olfactory stimuli over leisure stimuli (right panel); the remaining 4 subjects (Geoff,

Ernie, Jeffrey, and Mikey) preferred leisure stimuli to olfactory stimuli.

Discussion

Results of this study showed that both the PWS and non-PWS subjects showed a

preference for edible stimuli. In 39 out of 40 comparisons, PWS subjects preferred edible stimuli

over leisure stimuli (19 of 20 assessments) and over olfactory stimuli (20 out of 20 assessments).

Preference for food by non-PWS subjects also was generally consistent. : 9 of 10 non-PWS

subjects preferred edible stimuli to leisure stimuli, and 9 of 10 subjects preferred edible stimuli

to olfactory stimuli. Two of the 10 subjects showed mixed preference for edible and leisure

stimuli. Finally, when edible stimuli were excluded as selections in the leisure vs. olfactory

assessment, preferences by both PWS and non-PWS subjects were mixed in about equal

proportion.

The most noticeable similarity in the preferences of subjects with and without PWS is the

outcome of the assessments that included edible stimuli. Preference for edible stimuli in the PWS









subjects was very clear. It is unknown whether this difference is related to the clinical features of

PWS (i.e. food is the most valuable commodity and is always preferred regardless of what other

stimuli are available) or to other situational variables. For example, all of the subjects with PWS

participated in a treatment program in which access to food was severely restricted. Subjects had

strict caloric regimes, and all foods consumed conformed to a calorie plan and met other

nutritional needs. Therefore, certain high-calorie foods like snack foods were fairly novel. Very

small servings of snack foods compromised many of the selections in the preference

assessments. Deprivation from food in general and from these specific items in particular may

have served as an establishing operation that increased the value of edible items and could have

influenced preference assessment outcomes. Previous research has shown that deprivation from

and satiation to specific stimuli influences choice during preference assessments (Gottschalk,

Libby, & Graff, 2000; McAdam et al., 2005). The diets of the non-PWS subjects were not

restricted, so deprivation from food items would not have been a relevant establishing operation.

It would not have been possible to eliminate this potential establishing operation in our PWS

subjects because it is clinically undesirable to provide unrestricted access to food to individuals

with PWS.

The outcomes of the preference assessments in this study were consistent with previous

research suggesting that edible stimuli represent a highly preferred stimulus class (Bojak & Carr,

1999; DeLeon, Iwata, & Roscoe, 1997). In the study conducted by DeLeon et al., some particular

leisure item(s) outranked some edible item(s) for 5 of 14 subjects, but this variability was not

enough to alter overall preference for edible stimuli as a class. Exclusive edible preference was

observed in the Bojack and Carr study. Our data were only slightly different and suggest that

leisure stimuli may be preferred over edible stimuli for some individuals (see data for Geoff).









Our data also document preference for olfactory stimuli over edible stimuli or equal preference

for olfactory stimuli and edible stimuli (see data for Juan, Leyla and Mitchell).

Strong preference for a stimulus class like edible items, as reflected in mean rankings, does

not necessarily apply to all stimuli in a class because of the influence of outliers. Differences in

rankings within each stimulus class were observed that are not readily apparent when examining

mean ranks. In some cases, preference for a stimulus class was influenced by a single outlying

stimulus. Differences between mean ranks across each of the stimulus classes included in a given

assessment varied greatly (from 0.14 to 8.0). It should be noted that some individuals who

preferred edible stimuli to leisure stimuli as general stimulus classes preferred individual leisure

stimuli over all other stimuli included in the assessment. This information may be relevant when

designing treatment programs for individuals with PWS. For example, in the edible vs. leisure

assessment, Taneisha's highest ranked item was puzzle. This leisure item might be used as a

reinforcer in an intervention targeting food stealing, because her preference assessment outcome

data suggested that puzzles effectively competed with edible stimuli.

Also of note, olfactory stimuli were preferred to leisure stimuli by approximately half of all

subjects. Olfactory stimuli were even preferred to edible stimuli for 1 non-PWS subjects (Juan).

Preference for olfactory and edible stimuli was mixed for 2 additional non-PWS subjects. Results

of the leisure versus olfactory assessment are perhaps the most interesting from the standpoint of

reinforcer identification for individuals with PWS. Given that 10 of the 20 PWS subjects

preferred olfactory over leisure stimuli and an additional 2 subjects showed mixed preference for

olfactory and leisure stimuli, olfactory stimuli may function as reinforcers for adaptive

performances. It is possible that olfactory stimuli may be reinforcers even for subjects who

showed a lack of preference for olfactory stimuli relative to leisure stimuli because olfactory









stimuli may have some reinforcing value in contexts where no choice is provided. For example, a

typical training context involves a single response and a single consequence, rather than a choice

between multiple reinforcement options. In this type of setting, it is possible that olfactory

stimuli may have reinforcing value. Overall, the results of Experiment 1 suggest that further

examination of olfactory stimuli, specifically, the ability of olfactory stimuli to reinforce

adaptive behavior, is warranted.











Table 2-1. Experiment 1 demographic information
Name Gender Age Diagnoses
Janet Female 19 PWS, Moderate MR
Katarina Female 15 PWS, Mild MR
Patsy Female 55 PWS, Moderate MR
Trina Female 30 PWS, MR level not specified
Annette Female 37 PWS, Mild MR
Bettina Female 35 PWS, Mild MR
Penelope Female 37 PWS, Mild MR
Misty Female 32 PWS, Mild MR
Jewel Female 31 PWS, Moderate MR
Jacob Male 22 PWS, Mild MR
Taneisha Female 30 PWS, Mild MR
Jamal Male 38 PWS, Mild MR
Dorian Male 22 PWS, Mild MR
Manuel Male 19 PWS, MR level not specified, ADHD
Drake Male 30 PWS, Mild MR
Zeus Male 31 PWS, Moderate MR
Peter Male 45 PWS, Mild MR
Adeline Female 32 PWS, Moderate MR
Barnaby Male 21 PWS, Moderate MR
Melissa Female 46 PWS, Mild MR
Geoff Male 13 Autism
Montana Male 12 Autism
Cindy Female 20 Moderate MR
Ernie Male 3 PDD-NOS, Speech and Language Delay
Juan Male 37 MR level not specified, Seizure Disorder
Leyla Female 17 Mild MR
Wanda Female 49 Moderate MR
Mitchell Male 24 Mild MR
Jeffrey Male 29 Moderate MR
Mikey Male 38 Klinefelter's Syndrome, Mild MR
Subjects are in the left column. Gender is in the next column. Age is in the next column. Diagnosis and/or level of
mental retardation (MR) are in the right column.

































6(0 60- 60

40 40- 40

20 I 20. 20

0 LABSES- C S3 t' m ;W L t PS 5i 3 ag 3 L 5

I^^ jII I l 1 r h r |^ I



VSS n
100- Laesure 10-0 100
S o factory visr
80g- 80 80-

60 60- 60-



20204 2020
U. SSOlt. dE C UL,












Figure 2-1. Percentage of trials on which each stimulus was selected during each of the paired-
stimulus preference assessments, edible vs. leisure (top), edible vs. olfactory
(middle), and olfactory vs. leisure (bottom), for Peter (left), Annette (middle), and
Taneisha (right).
















Janet-

Kaarina-

Palsy-

Trina-

Annette-

Bettina-

Penelope

Misty-

Jewel-

Jacob-

Tancisha-

Jamal-

Dorian-

Manuel-

Drake-

Zeus-

Peter-

Adeline-

Barnaby-

Melissa-


E Edible
SLeisure Janet-

Kalarina-

Palsy-

Trina

SAnnette-

_____Bettina

Penelope-

Misty

__________V___ Jewel-

------- Jacob-

Taneisha-



Dorian-

__ __Manuel-

------- Drake-

---_ Zeus-

Peter-

Adeline-

Bamaby

-- Melissa-
5 10 I1


EI Edible
SOlfactory Janct-

2"-
____ Katarina-

zzzzzJ ^ ^ Pasy-

Trina-

Annette-

Bettina-

Penelopeb

Misty-

SJewd

IJaob-

____-1 Tancisha-

jamal-

Dorian-

------- Manuel-

Drake-

__- Zeus-
------ Peter-

Adeline-

Barnaby-

NMelki a-



Mean Rank


Figure 2-2. Mean preference rankings of stimulus classes for all PWS subjects from each of the

paired-stimulus preference assessments: edible vs. leisure (left), edible vs. olfactory
(middle), and olfactory vs. leisure (right).


ES Leisure
S s Olfactory
mmmm



































S Ir IS
AXWAVWAVWA1~PC

.WAW^CT~rC~a

VIAWAW VV'AV
IA^^A'.II




mm


cBSScm


5 SSSESSSS Y















Geoff


Montana


Cindy


Ernie


Juan


Leyla


Wanda


Mitchell


Jeffrey


Mikey


Q Edible
SS Leisure
I -u Geoff
II

Montana-
~-

Cindy


~Ernie


-I Juan


~-
1 1 Leyla


I Wanda


Mitchell
I-



111 Jeffrey


Mikey-

3 1 155


E1 Edible
Olfactory


































1 5 lb 15


Geoff-


Montana-


Cindy-


Ernie-


Juan


Leyla-


Wanda


Mitchell


Jeffrey


Mikey

0 i --- b --- f


Mean Rank



Figure 2-3. Results obtained for all non-PWS subjects from each of the paired-stimulus
preference assessments: edible vs. leisure (left), edible vs. olfactory (middle), and
olfactory vs. leisure (right).


, Leisure
SOlftactory









CHAPTER 3
EXPERIMENT 2

Purpose

The purpose of Experiment 2 was to determine whether preferences observed in

Experiment 1 were predictive of reinforcement effects when olfactory stimuli were delivered as

consequences. Because there were differences in PWS subjects' preference for olfactory vs.

leisure stimuli in Experiment 1, two different tests were used in Experiment 2. One test involved

the delivery of olfactory stimuli for responding under a single schedule of reinforcement to

determine whether they could function as reinforcers. If Experiment 1 results showed that

olfactory stimuli were more preferred than leisure stimuli (or equally preferred), reinforcement

effects were evaluated under a concurrent schedule in which subjects could respond for leisure or

olfactory reinforcers. This test was designed to determine whether olfactory stimuli could

compete with leisure stimuli as reinforcers.

Method

Subjects and Setting

Eight individuals with PWS from Experiment 1 served as subjects in Experiment 2. A

subset of subjects for whom olfactory stimuli were more preferred than leisure stimuli (Bettina),

for whom preference for olfactory stimuli and leisure stimuli was mixed (Taneisha), or for whom

olfactory stimuli were less preferred than leisure stimuli (Zeus, Peter, Drake, Adeline),

underwent a single-schedule assessment to determine if olfactory items had any reinforcing

value. Two of those subjects (Bettina and Taneisha) and an additional subset of subjects for

whom olfactory stimuli were more preferred than leisure stimuli (Annette and Patsy) underwent

a concurrent-schedule assessment to verify their preference for olfactory items. Subjects were

included based on whether or not they had PWS and their preferences from Experiment 1.









Sessions were conducted at a day activity program for individuals with developmental

disabilities or at a school for children with exceptional needs. All sessions were 10 min in

duration and were conducted two to three times per day, four to five days per week.

Task Selection and Initial Training

A task was selected for each subject to serve as the basis for evaluating reinforcement

effects. Tasks differed somewhat across subjects (see Table 3-1) but had the following

characteristics: (a) they were vocational or pre-vocational in nature, (b) they involved discrete

responses that were easily observed, (c) the responses were free operant in the sense that they

could occur repeatedly throughout a session, and (d) the responses could be performed by the

subject already or, if not, could be learned quickly. If the subject was unable to perform the task

at the beginning of the experiment, training was provided. Training consisted of the experimenter

initially modeling the response and then asking the subject to engage in the response. Feedback

was provided during training if the task was performed incorrectly. If a response was not easily

trained in less than 10-min, a new response was selected. During training, performance of the

target behavior resulted in access to one of four moderately preferred (middle ranked and not

used during the experiment) leisure or olfactory items identified during the preference

assessment on a fixed-ratio (FR) 1 schedule.

Response Measurement and Reliability

Target responses consisted of simple vocational or pre-vocational tasks (e.g., placing cards

in envelopes). Observers recorded the frequency of target responses and the delivery of

reinforcers using handheld computers. An independent observer collected data during a mean of

47% (range, 20%-70.5%) of all reinforcer assessment sessions. To assess reliability, session time

was divided into 10-s intervals, and observers' records were compared on an interval-by-interval

basis. Proportional agreement was calculated by dividing the smaller number of responses scored









in each interval by the larger number of responses, summing these fractions, and multiplying by

100%. The mean agreement score across all participants was 98% (range, 86.7%-100%) during

the reinforcer assessments. Response rates were corrected for reinforcer access time.

Single-Schedule Reinforcer Assessment

The assessment consisted of presentation of a task identical to that used during training. A

single task was placed on a table in front of the subject. No other materials were present. Prior to

each session, the subject was prompted to perform the task once and was exposed to the

contingency in place for task completion during that condition. The subject was free to do

nothing or engage in the task during all sessions.

Baseline. No programmed contingencies were in place for task completion.

Olfactory reinforcement. The olfactory stimuli available contingent upon task

performance was placed directly behind the task, visible to the subject. Task performance

resulted in 30-s access to 1 of 3 olfactory items that had been identified as preferred during

Experiment 1 (fixed-ratio 1 schedule [FR-1]).

Leisure and edible reinforcement. If responding in the olfactory reinforcement condition

did not increase relative to baseline, a leisure reinforcement condition was conducted. If

responding still did not increase in the leisure reinforcement condition, an edible reinforcement

condition was conducted. The purpose of these additional conditions was to determine whether

any reinforcement effect could be obtained, given that olfactory stimuli had been ineffective. The

stimuli (leisure or edible) available contingent upon task performance was placed directly behind

the task, visible to the subject. In the leisure-reinforcement condition, task performance resulted

in 30-s access to 1 of 3 leisure items that had been identified as preferred during Experiment 1

(FR-1). In the edible-reinforcement condition, task performance resulted in delivery of 1 piece of

1 of 3 items that had been identified as preferred during Experiment 1 (FR-1).









Concurrent-Schedule Reinforcer Assessment

The assessment consisted of concurrent presentation of two tasks, identical to those used

during training. One task was placed on the right side a rectangular table and the other task was

placed on the left side of the table. Task positions alternated across sessions, and a task did not

stay on a particular side for more than 2 consecutive sessions. Task materials were covered with

or placed on different colored construction paper to serve as discriminative stimuli for

responding associated with each task. Also, the stimuli available contingent upon each task was

placed directly behind each task, visible to the subject. Performance of one task produced access

to one of the top three olfactory stimuli (identified in the olfactory vs. leisure assessment during

Experiment 1); performance of the other task produced access to one of the top three leisure

stimuli (identified in the olfactory vs. leisure assessment during Experiment 1). Stimulus

presentation consisted of placing all 3 stimuli within arm's reach of the subject, and then

allowing the subject to select 1 stimulus. Prior to each session, the participant was prompted to

perform each task once and received 30-s access to the associated reinforcer. During each

session, performance of each task resulted in 30-s access (FR-1) to the stimulus associated with

that task. At the start of each session, the therapist told the subject that they could do either task

or do nothing at all.

Results

Figure 3-1 shows results of the single-schedule reinforcer assessments. Bettina did not

respond during the first baseline condition but responded during five out of six sessions during

the first olfactory reinforcement condition. Her responding quickly fell to zero during the second

baseline condition, and she responded steadily during the second olfactory reinforcement

condition. Taneisha, Peter,Drake, and Adeline never responded during baseline; Zeus responded

during only one of his five baseline sessions. Taneisha, Peter, Zeus, and Adeline showed









increases in responding during the olfactory-reinforcement condition. By contrast, Drake did

not emit any responses during the olfactory-reinforcement condition, nor did he respond in the

subsequent leisure-reinforcement condition. His responding increased only when edible items

were delivered as reinforcers, which raised the question of whether responding might maintain

under olfactory reinforcement. Drake's responding ceased entirely, however, when olfactory

reinforcement was reinstated, so his final condition was a re-exposure to the edible

reinforcement, which again resulted in an increase in responding.

Figure 3-1 shows results of the concurrent-schedule reinforcer assessments conducted with

Bettina, Annette, Patsy, and Taneisha. All 4 subjects alternated between the two response (and

reinforcer) options and, in fact, responded at nearly identical rates for olfactory and leisure

stimuli. Patsy's responding was most variable across sessions but still remained in within a rather

narrow range (0.3 to 0.8 responses per min), and the difference in responding towards one option

over another within a session never exceed 0.2 responses per min for any subject.

Discussion

Results from Peter's, Zeus's, and Adeline's single-schedule reinforcer assessment showed

that, despite the low ranking of olfactory stimuli in their preference hierarchy (Experiment 1),

olfactory stimuli had some reinforcing value. These data were consistent with previous research

showing that stimuli not selected during paired-stimulus preference assessments may still

function as reinforcers when other options are unavailable (Roscoe et al., 1999). Drake's data

paint a very different picture. Olfactory stimuli failed to function as reinforcers during his single-

schedule reinforcer assessment. More surprising was the fact that leisure stimuli also failed to

function as reinforcers during the same assessment. This suggests that Drake's reinforcer pool

was very limited and consisted only of his most preferred stimulus class, edible stimuli. The









potency of edible stimuli to function as reinforcers for Drake, or for other subjects with PWS, is

not surprising.

Results of Bettina and Taneisha's single-schedule reinforcer assessment and results of the

concurrent-schedule reinforcer assessment Bettina, Annette, Patsy, and Taneisha provided some

measure of verification for preference patterns observed in Experiment 1 in that olfactory stimuli

(a) functioned as reinforcers, and (b) competed with leisure stimuli. The subjects included in the

concurrent-schedule reinforcer assessment showed two distinct profiles in Experiment 1: Bettina,

Annette, and Patsy preferred olfactory stimuli over leisure stimuli, whereas Taneisha's

preference for olfactory and leisure stimuli was mixed. Both patterns suggested the possibility

that olfactory reinforcers might compete with leisure reinforcers under a concurrent schedule

and, in fact, their responding during Experiment 2 was virtually identical for both consequences.

It was interesting to note that none of the subjects who preferred olfactory over leisure stimuli in

Experiment 1 did so in Experiment 2.

One possible explanation for similarity observed across subjects in the concurrent-schedule

reinforcer assessment in Experiment 2 is related to the nature of the olfactory sensory system.

Repeated exposure to olfactory stimuli may result in reduced sensitivity of the olfactory

receptors and neurons by approximately 50% within 2.5 s tolO s of exposure (Wang, Walker,

Sardi, Fraser, & Jacob, 2002). Thus, following the first response or two that is reinforced with

olfactory stimuli, subjects may allocate responding towards leisure items due to the reduced

efficacy of the olfactory receptors following exposure to olfactory stimuli. Our olfactory

reinforcement access period was 30 s, much longer than the exposure periods noted by Wang et

al.









Additional support for this explanation can be found in research on the detection of an

olfactory stimulus. Short and intense exposure to odorants seems to be more important than

duration of exposure for optimization of olfactory detection (Zhao, Dalton, Yang, & Scherer,

2006). Perhaps this is because of the decrease in sensitivity of olfactory neurons following

olfactory stimulus exposure noted above. Thus, a refractory period between olfactory-stimulus

exposures may facilitate more effective responses to olfactory stimuli. If so, the equal response

allocation towards both the olfactory and leisure options may be a by-product of the periodic

reduction in the ability of the olfactory receptors to respond to olfactory stimuli. Moving from

the olfactory option to the leisure option may have allowed a refractory period between

olfactory-stimulus exposures.

An alternative explanation may be that the pre-session exposure to the tasks and the

associated stimuli may have served as a prompt to alternate between responses during the

session. If so, it is possible that pre-session exposure exerted control during session and

responding was not influenced by contingent access to the putative reinforcers. This potential

concern is somewhat mediated by the data from Bettina and Taneisha's single-schedule

reinforcer assessment because neither subject engaged in the target response during baseline.

However, future studies could include a control task in order to simultaneously evaluate whether

or not olfactory stimuli function as reinforcers and examine their relative efficacy. Nevertheless,

a reinforcement effect was observed for olfactory stimuli with 7 of the 8 subjects in Experiment

2 suggesting the potential viability of olfactory reinforcement for use in skill-acquisition or

treatment programs.









Table 3-1. Reinforcer assessment type and tasks
Subject Assessment Schedule Task
Bettina Single, Concurrent Raising arm above shoulder
Taneisha Single, Concurrent Calculator button pressing
Peter Single Touching symbol on wall
Drake Single Putting a card in an envelope
Zeus Single Phone dialing
Adeline Single Touching head, shoulders, knees and toes
Patsy Concurrent Placing card in bin
Annette Concurrent Punching a hole in a card
Subjects are in the left column. Assessment types (single-schedule or concurrent-schedule) are listed in the middle
column. Tasks are listed in the right column.













BL Sr+ BL Sr+
S (olfactory) I (olfactory)


2 4



0.a


0 Sr+ Sr+ Sr+ Sr+
0 (leisure) (edible) (olfactory) (edible)

S15-

10-








4













5 10 15 2(
Sessions



Figure 3-1. Results of the single-schedule reinforcer assessment for Bettina, Taneisha, Peter,
Drake, Zeus, and Adeline..

















20






Leisure Task
Olfactory Task






2.01


1.5


1B0





0.0 ----------------------


2.0





1.0


0.5
Tastnd


2 4 6 8 2 4 6
Sessions


Figure 3-2. Results of the concurrent-schedule reinforcer assessment for Bettina, Annette, Patsy,

and Taneisha.









CHAPTER 4
GENERAL DISCUSSION

Two experiments examined preference for edible, leisure, and olfactory stimuli in

individuals with PWS. Experiment 1 consisted of a series of 3 preference assessments (edible vs.

leisure, leisure vs. olfactory, and olfactory vs. leisure) administered to individuals with and

without PWS. The PWS and non-PWS subjects exhibited almost exclusive preference for edible

stimuli over leisure and olfactory stimuli. PWS and non-PWS subjects both showed mixed

preferences favoring neither stimulus class in the preference assessment that did not include food

(olfactory vs. leisure stimuli).

Experiment 2 validated the preference assessment outcomes from Experiment 1 with a

subset of subjects. Four subjects from Experiment 1 who preferred olfactory stimuli over leisure

stimuli (or for whom preference for olfactory stimuli and leisure stimuli was mixed) were

exposed to concurrent-schedule reinforcer assessments in Experiment 2. All 4 subjects allocated

responses toward both olfactory and leisure reinforcers. This outcome demonstrated that

olfactory stimuli competed with leisure stimuli. Four additional subjects for whom olfactory

stimuli were least preferred were exposed to a single-schedule reinforcer assessment in which

only olfactory reinforcers were available. Threee of the 4 subjects showed increases in

responding relative to baseline in the olfactory reinforcement condition, whereas the fourth

subject did not respond for olfactory reinforcement.

Taken together, results of these experiments suggest that olfactory stimuli have the

potential for use as reinforcers in skill acquisition and treatment programs for individuals with

PWS and for other individuals with developmental disabilities who are on restricted diets. The

preference-assessment data suggest that olfactory stimuli may be equally preferred to leisure

stimuli (some PWS subjects) or perhaps even preferred to food (some non-PWS subjects). The









reinforcer-test data suggest that preference for olfactory stimuli is likely to be reflected in

performance on tasks for which olfactory stimuli are delivered as consequences. Also of note,

access to olfactory stimuli was a reinforcer for individuals who previously showed no preference

for olfactory stimuli. This finding was consistent with previous research showing that paired-

stimulus preference assessments are best suited for predicting relative but not absolute reinforcer

value (Roscoe et al., 1999).

One finding from Experiment 1 that emphasizes the importance of individualized

assessment rather than generalizations based on aggregated data was the fact that single leisure

stimuli outranked edible stimuli in individuals with PWS. This has important implications for the

development of treatment programs directed at reducing food-related problem behavior. If a

preference assessment suggests that a specific leisure stimulus may be more preferred than edible

stimuli, that stimulus could be the reinforcer in a differential reinforcement of other behavior

(DRO) procedure for refraining from inappropriate food consumption or food stealing.

Several findings of the current studies were somewhat inconsistent with what has been

reported in previous research on preference assessment. First was the fact that, unlike data

reported by DeLeon et al. (1997) and Bojak and Carr (1999), the non-PWS subjects in our study

did not show an exclusive preference for edible stimuli, although preference for edible stimuli

was strong. It would be difficult to determine the basis for the difference. However, the results

of the current study suggest that there may be some exceptions to the general preference for

edible stimuli that has been previously reported.

Another somewhat unusual finding was that, although Bettina, Annette, and Patsy showed

a clear preference for leisure over olfactory stimuli in Experiment 1, their rates of responding

were almost identical for leisure versus olfactory reinforcement in Experiment 2. Previous









research (Fisher et al., 1992, Roscoe et al., 1999) has shown that stimuli higher ranked in paired-

stimulus preference assessments are usually preferred under concurrent reinforcer schedules.

The absence of preference for olfactory over leisure stimuli under the concurrent schedule may

have been due to several factors.

First, response allocation has not always favored higher ranked items from preference

assessments. One subject in both the Fisher et al. (1992) and Roscoe et al. (1999) studies

responded equally for high- and low-ranked reinforcers; thus, it is possible that some subjects

simply respond so as to maximize reinforcer variability. Previous research has shown that some

subjects' performance improves when exposed to varied reinforcer presentation relative to

constant reinforcer presentation (Egel, 1981).

Second, it may be possible that a refractory period is necessary between olfactory stimulus

exposure periods for the maximum effect of the olfactory stimulus to be experienced (Wang et

al, 2002). If that is the case, alternation between the task that resulted in olfactory reinforcement

and the task that resulted in leisure reinforcement would be better understood. This would

suggest a potential limitation to the use of olfactory stimuli as reinforcers. More specifically,

olfactory stimuli may be more sensitive to a habituation effect than other classes of stimuli.

Third, specific top-ranked leisure stimuli may have been better able to compete with

olfactory stimuli than other members of the leisure class. Although Bettina, Annette, and Patsy

generally preferred olfactory over leisure stimuli, preference for specific leisure stimuli and

specific olfactory stimuli was more mixed. For example, Bettina preferred olfactory stimuli over

leisure stimuli as a general class. However, her top 3 stimuli in the olfactory versus leisure

assessment included 2 leisure stimuli. Thus, when the top 3 olfactory stimuli were pitted against

the top 3 leisure stimuli in a concurrent-schedule reinforcer assessment, 1 of the top 3 leisure









items outranked all olfactory stimuli, and a second leisure item outranked 2 of the top 3 olfactory

stimuli. Thus, mixed responding in the concurrent-schedule reinforcer assessment is not

surprising. Future research might gradually increase the schedule requirements for the leisure

stimuli in the concurrent schedule reinforcer assessment to determine at what point preference

might shift towards the olfactory reinforcement option.

Several limitations of this study should be noted. All of the olfactory stimuli included in

the preference assessments were food flavors. In order to minimize differences due to preference

for specific flavors, an attempt was made to make the olfactory stimuli somewhat similar to the

edible stimuli. It was unclear, however, whether the olfactory essences would be as preferred as

the samples of the actual foods (and the real odors associated with foods). Also, exclusive use of

food odors may have restricted the range of olfactory stimuli included in Experiment 1. Edible

and olfactory stimuli have a well-established pairing history due to the physiology of the oral and

nasal orifices. It is possible that preference for olfactory stimuli over leisure stimuli may be due

to the exclusive inclusion of food-based olfactory stimuli in this experiment. Expanding the

range of olfactory stimuli to include non food-based olfactory stimuli might disentangle the

edible/olfactory pairing history and provide a more convincing demonstration of preference for

olfactory stimuli unrelated to preference for edible stimuli. Anecdotal evidence suggests that

non-food olfactory stimuli may also be preferred. Walking through the perfume aisle in a

department store clearly demonstrates that humans are willing to engage in responses in order to

access floral, woodsy, musky, and other non-food fragrances. Future research might examine a

broader range of olfactory stimuli.

Another limitation was the nature of responses in Experiment 2. Although we selected

responses designed to approximate vocational tasks, they were simple and repetitive. Therefore,









it is unknown if olfactory reinforcement is sufficient to establish or maintain more effortful

responses or response chains.

Finally, Experiment 2 focused on effects for several PWS subjects only. Therefore, it does

not permit any general conclusions about olfactory reinforcement. There was no reinforcer test

for the non-PWS subjects because their inclusion in Experiment 1 was as a comparison sample.

Given their preference (some observed preference for olfactory stimuli), it would have been

interesting to see the extent to which olfactory stimuli served as reinforcers for non-PWS

subjects. Future research could include reinforcer assessments to validate the results of

preference assessments with individuals with disabilities other than PWS.

In summary, the present studies provide information about a potential class of reinforcers

that typically has not been examined in operant research. Olfactory stimuli do not pose the same

types of problems for individuals on restricted diets as edible stimuli pose. Thus, they may have

special benefits for individuals with PWS. Results showed that olfactory stimuli have properties

that compete with those of leisure stimuli (and sometimes edible stimuli) in the context of

preference assessments and that olfactory stimuli can function as reinforcers in some individuals.

Both of these findings support an expansion of the "reinforcer pool" to include olfactory stimuli

and further studies examining their varied properties as well as their influence on operant

behavior.










APPENDIX A
STIMULI USED IN PREFERENCE ASSESSMENTS DURING EXPERIMENT 1 AND
PREFERENCE RANKINGS FOR EACH STIMULUS

Table A-1. Group 1: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings
Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Mikey, Misty, Peter, 1(Orange Slice), 2 (Honey 1 (Beads), 2 1(Apple), 2(Orange),
Jeffrey, Penelope, Nut), 3 (Coconut), 4 (Workbook), 3 (Puzzle), 3 (Strawberry), 4
Mitchell, Bettina, (Strawberry Creme Saver), 4 (Bear), 5 (Magazines), (Mint), 5(Coconut),6
Annette, Zeus, 5 (Mint Tic Tac), 6 (Apple 6 (Bumbleball),7 (Cinnamon), 7
Taneisha, Drake Cane), 7 (Cinnamon Altoid) (Slinky) (Honey Nut)


Mikey .-.'-PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 1 3 5.5 7 8.5 8.5 13 6.64
Leisure 3 10.5 5.5 10.5 14 3 12 8.36
Edible vs. 1 2 3.5 3.5 6.5 6.5 5 4
Olfactory 8.5 13 11 11 11 14 8.5 11
Leisure vs. 1 2 6.5 6.5 11.5 3 4.5 5
Olfactory 4.5 11.5 11.5 11.5 11.5 11.5 8 10


Misty (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 2 6.5 6.5 1 6.5 3.5 3.5 4.21
Leisure 14 6.5 9 11 11 11 13 10.79
Edible vs. 1 4 3 2 7 5 6 4
Olfactory 14 13 11.5 11.5 9.5 9.5 8 11
Leisure vs. 13 2 1 11 14 9 12 8.86
Olfactory 9 9 6.5 5 4 6.5 3 6.14

Peter (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 4.5 2.5 6.5 2.5 6.5 1 4.5 4
Leisure 12.5 11 12.5 9.5 14 8 9.5 11
Edible vs. 1 6.5 6.5 3.5 5 3.5 2 4
Olfactory 11 11 13.5 8.5 13.5 11 8.5 11
Leisure vs. 6.5 4 14 2 6.5 3 1 5.29
Olfactory 10.5 10.5 10.5 6.5 10.5 13 6.5 9.71











Table A-1. Continued
Jeffrey .. ,-PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 1 3 8.5 3 6 3 5 4.21
Leisure 14 11 7 12.5 12.5 8.5 10 10.79
Edible vs. 1.5 1.5 7.5 4 5 6 3 4.07
Olfactory 7.5 9 10.5 13 13 10.5 13 10.93
Leisure vs. 5.5 3 7.5 1 3 3 5.5 4.07
Olfactory 9.5 7.5 9.5 12 14 12 12 10.93

Penelope (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 2 4 8 1 8 4 6 4.71
Leisure 11 8 4 13 11 14 11 10.29
Edible vs. 2 4 5 1 10 3 6 4.43
Olfactory 8.5 11.5 11.5 14 13 8.5 7 10.57
Leisure vs. 10.5 2.5 1 10.5 8 12.5 12.5 8.21
Olfactory 5 5 7 14 9 2.5 5 6.79

Mitchell (Non-PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 14 1 4 8 2.5 11 2.5 6.14
Leisure 11 11 13 6.5 9 5 6.5 8.86
Edible vs. 14 2 5 11 3 13 1 7.0
Olfactory 9.5 8 12 6.5 9.5 6.5 4 8.0
Leisure vs. 13 12 14 5.5 9.5 11 9.5 10.64
Olfactory 5.5 5.5 8 2 5.5 2 2 4.36

Bettina (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 6.5 2 12.5 3.5 5 8 3.5 5.86
Leisure 14 6.5 9.5 1 9.5 12.5 11 9.14
Edible vs. 5 1 13 2 5 7 3 5.14
Olfactory 9.5 14 9.5 11.5 11.5 8 5 9.86
Leisure vs. 14 9.5 4 1 11 13 12 9.21
Olfactory 4 2 7.5 7.5 9.5 4 6 5.79

Annette (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 6.5 3 5 1 6.5 3 3 4.0
Leisure 12 12 8 9 12 12 10 10.71
Edible vs. 1 2.5 6.5 4 6.5 2.5 5 4.0
Olfactory 12.5 10.5 8 14 9 12.5 10.5 11
Leisure vs. 10 14 10 12.5 10 8 12.5 11.0
Olfactory 4.5 4.5 2.5 7 2.5 6 1 4










Table A-1. Continued
Zeus (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 1 4 7 2 6 3 5 4.0
Leisure 8 10.5 8 14 10.5 8 13 10.29
Edible vs. 1 4 6 2 6 3 6 4.0
Olfactory 13.5 8.5 10 11 12 13.5 8.5 11.0
Leisure vs. 2 7.5 1 5.5 10 3.5 3.5 4.71
Olfactory 10 13 5.5 12 10 14 7.5 10.29

Taneisha (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7

Edible vs. 5 2 9 3.5 7.5 3.5 7.5 5.43
Leisure 13 6 1 13 13 10 11 9.57
Edible vs. 2.5 4 5 2.5 6 1 7.5 4.07
Olfactory 9.5 12.5 11 9.5 12.5 14 7.5 10.93
Leisure vs. 3.5 1.5 1.5 13.5 13.5 7.5 11 7.43
Olfactory 5 7.5 3.5 12 9.5 9.5 6 7.57

Drake (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 1 3.5 3.5 2 8.5 5 6.5 4.29
Leisure 14 11 6.5 8.5 11 13 11 10.71
Edible vs. 2 3 4 1 7 5.5 5.5 4.0
Olfactory 11 8.5 13 11 11 14 8.5 11.0
Leisure vs. 12 6 4.5 1 2.5 13 2.5 5.93
Olfactory 8 10.5 10.5 14 8 8 4.5 9.07


Table A-2. Group 2: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings
Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Patsy, 1(Teddy Graham), 2 (Banana 1 (Beads), 2 (Workbook), 3 1(Apple), 2 (Cookie), 3
Wanda, Nut Muffin), 3 (Honey Dew), (Puzzle), 4 (Bear), 5 (Strawberry), 4 (Mint), 5
Dorian 4 (Pineapple Jelly Bean), 5 (Magazines), 6 (Bumble Ball), 7 (Pineapple), 6
(Strawberry Cream), 6 (Mint (Slinky), 8 (Kaleidoscope) (Cinnamon), 7 (Honey
Tic Tac), 7 (Apple Cane), 8 Nut), 8 (Banana Nut)
(Cinnamon Altoid)

Patsy (P WS)
Assessment Stimuli Mean
1 2 3 4 5 6 7 8
Edible vs. 6 1 3 6 2 8 6 4 4.5
Leisure 16 10.5 15 12.5 10.5 14 12.5 9 12.5
Edible vs. 5 2.5 4 8 1 6 7 2.5 4.5
Olfactory 15.5 15.5 11 11 13.5 9 13.5 11 12.5
Leisure vs. 16 5 15 8.5 14 8.5 12.5 8.5 11.0
Olfactory 5 12.5 8.5 5 11 2 1 3 6.0










Table A-2. Continued
Wanda .-.'-PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7 8
Edible vs. 3.5 1.5 9 3.5 1.5 6.5 5 6.5 4.63
Leisure 16 12 13 8 11 10 13.5 13.5 12.13
Edible vs. 3 1.5 8 4 1.5 5.5 5.5 7 4.5
Olfactory 12.5 15.5 10.5 12.5 10.5 9 14 15.5 12.5
Leisure vs. 15 12 10.5 13.5 7.5 10.5 13.5 16 12.31
Olfactory 5.5 5.5 1 3 3 3 7.5 9 4.69


Dorian (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7 8
Edible vs. 13 2 16 3.5 1 15 11 8.5 8.75
Leisure 5.5 3.5 7 8.5 11 11 5.5 14 8.25
Edible vs. 8.5 2.5 11.5 2.5 1 16 8.5 5 6.94
Olfactory 8.5 5 5 13.5 13.5 11.5 15 8.5 10.06
Leisure vs. 5 1.5 7 7 1.5 3.5 3.5 16 5.63
Olfactory 14 9.5 7 15 12.5 9.5 12.5 11 11.38


Table A-3.


Group 3: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings


Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Juan 1(Teddy Graham), 2 1 (Beads), 2 1(Apple), 2 (Cookie), 3
(Banana Nut Muffin), 3 (Workbook), 3 (Puzzle), (Strawberry), 4 (Mint), 5
(Honey Dew), 4 4 (Bear), 5 (Magazines), (Pineapple), 6
(Pineapple Jelly Bean), 5 6 (Bumble Ball), 7 (Cinnamon), 7 (Honey
(Strawberry Cream), 6 (Slinky), 8 Nut), 8 (Banana Nut)
(Mint Tic Tac), 7 (Apple (Kaleidoscope)
Cane), 8 (Cinnamon
Altoid)
Juan ., o-PWS)
Assessment Stimuli Mean

1 2 3 4 5 6 7 8
Edible vs. 10.5 8.5 6 6 3.5 2 1 3.5 5.13
Leisure 14 15 16 10.5 12 8.5 6 13 11.88
Edible vs. 15 16 11 12 14 4.5 13 4.5 11.25
Olfactory 8.5 1 6.5 3 2 6.5 8.5 10 5.75
Leisure vs. 16 15 13.5 12 11 2 10 13.5 11.63
Olfactory 3.5 8.5 8.5 7 5 1 3.5 6 5.38











Table A-4. Group 4: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings
Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Melissa, Jacob, Barnaby, 1 (Mint Tic Tac), 2 1 (Beads), 2 1(Vanilla), 2 (Orange), 3
Trina, Adeline, Jewel, (Orange Tictac), 3 (Workbook), 3 (Puzzle), (Strawberry), 4 (Mint), 5
Jamal (Almond), 4 (Hot 4 (Bear), 5 (Magazines), (Coconut), 6
Tamale), 5 (Coconut 6 (Bumbleball), 7 (Cars) (Cinnamon), 7 (Almond)
Jelly Bean), 6 (Vanilla
Wafer), 7 (Strawberry
Twizzler)

Melissa (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 11 5.5 1.5 3.5 1.5 5.5 3.5 4.57
Leisure 8 12.5 7 9.5 9.5 12.5 14 10.43
Edible vs. 7.5 5.5 3 2 1 5.5 4 4.07
Olfactory 11 9.5 13.5 13.5 9.5 11 13.5 11.64
Leisure vs. 1 2 3.5 6 6 6 14 5.5
Olfactory 10.5 12 9.5 13 10.5 3.5 9.5 9.79

Jacob (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 6 7.5 1.5 4.5 3 1.5 4.5 4.07
Leisure 14 7.5 11 12.5 10 12.5 9 10.93
Edible vs. 7.5 4.5 2 3 4.5 1 6 4.07
Olfactory 9.5 13 14 11.5 11.5 7.5 9.5 10.93
Leisure vs. 14 10 12 6.5 6.5 11 13 10.43
Olfactory 3 8.5 8.5 1 4 2 5 4.57

Barnaby (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 7 6 5 3 2 1 4 4.0
Leisure 12 13 9 14 11 9 9 11.0
Edible vs. 7 6 4 4 2 1 4 4.0
Olfactory 8.5 10 8.5 12 14 12 12 11.0
Leisure vs. 9 5.5 1.5 11 3.5 1.5 5.5 5.36
Olfactory 3.5 9 9 7 14 12 13 9.64

Trina (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 7.5 11.5 4.5 1 2.5 7.5 2.5 5.29
Leisure 4.5 11.5 7.5 7.5 11.5 14 11.5 9.71
Edible vs. 10 7.5 2.5 2.5 1 4 5 4.64
Olfactory 10 14 7.5 13 10 6 12 10.36
Leisure vs. 5.5 11 7 11 5.5 11 11 8.86
Olfactory 1 2.5 2.5 11 4 11 11 6.14











Table A-4. Continued
Adeline (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 8 6 3 1.5 4.5 9 4.5 5.21
Leisure 7 10.5 1.5 12 10.5 13 14 9.79
Edible vs. 5 5 5 1 2.5 7 2.5 4.0
Olfactory 11 11 9 14 13 8 11 11.0
Leisure vs. 2 2 2 9.5 5 11.5 13.5 6.5
Olfactory 11.5 8 6 13.5 4 7 9.5 8.5

Jewel (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 2.5 4.5 7 1 2.5 4.5 6 4.0
Leisure 11 14 10 8.5 13 12 8.5 11.0
Edible vs. 2 4 7 1 3 6 5 4.0
Olfactory 8 13.5 9 11 11 11 13.5 11.0
Leisure vs. 12.5 10 7 8 14 10 12.5 10.57
Olfactory 3.5 5.5 1 3.5 2 5.5 10 4.43

Jamal (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 12 8 3 6 2 6 6 6.14
Leisure 4 13.5 1 10 10 13.5 10 8.86
Edible vs. 4 4 1 7.5 2 4 7.5 4.29
Olfactory 11.5 11.5 11.5 14 7.5 7.5 11.5 10.71
Leisure vs. 2 3 1 12 13.5 14 13.5 8.43
Olfactory 10 10 6 6 4 10 6 7.43




Table A-5. Group 5: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings
Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Manuel 1 (Mint Tic Tac), 2 (Orange Tic Tac), 1 (Castle), 2 (Leap Pad), 3 1(Vanilla), 2 (Orange), 3
3 (Almond), 4 (Hot Tamale), 5 (Football Game), 4 (Strawberry), 4 (Mint), 5
(Coconut Jelly Bean), 6 (Vanilla (Coloring Book), 5 (Books), (Coconut), 6
Wafer), 7 (Strawberry Twizzler) 6 (Blue Car), 7 (Legos) (Cinnamon), 7 (Almond)

Manuel (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 4 10 8.5 1.5 3 5.5 1.5 4.86
Leisure 7 13 14 11 12 5.5 8.5 10.14
Edible vs. 3 6 5 3 3 1 7 4.0
Olfactory 8.5 11 11 8.5 11 13 14 11.0
Leisure vs. 3 6 8.75 8.75 8.75 1.5 1.5 5.46
Olfactory 4.5 4.5 8.75 8.75 8.75 8.75 8.75 7.53












Table A-6. Group 6: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings

Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli

Katarina 1 (Mint Tic Tac), 2 (Orange 1 (Caterpillar), 2 (Coloring 1(Vanilla), 2 (Orange), 3
Tictac), 3 (Almond), 4 (Hot Book), 3 (Magazine), 4 (Strawberry), 4 (Mint), 5
Tamale), 5 (Coconut Jelly (Memory Match), 5 (Dog (Coconut), 6 (Cinnamon), 7
Bean), 6 (Vanilla Wafer), 7 Puppet), 6 (Lincoln Logs), (Almond)
(Strawberry Twizzler) 7 (Mr. Potato Head)

Katarina (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 6.5 6.5 3.5 1.5 1.5 3.5 5 4.0
Leisure 8 9 10.5 10.5 12 13 14 11.0
Edible vs. 7 6 1 3 2 4 5 4.0
Olfactory 11 12 13.5 13.5 9 9 9 11.0
Leisure vs. 11 3 11 11 11 11 11 9.86
Olfactory 4.5 7 11 6 1.5 1.5 4.5 5.14




Table A-7. Group 7: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings
Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Cindy 1 (Mint Tic Tac), 2 (Orange Tic 1 (Leap Pad), 2 (Coloring 1(Vanilla), 2 (Orange), 3
Tac), 3 (Almond), 4 (Hot Tamale), Book), 3 (Magazine), 4 (Race (Strawberry), 4 (Mint), 5
5 (Coconut Jelly Bean), 6 (Vanilla Cars), 5 (Dog Puppet), 6 (Coconut), 6
Wafer), 7 (Strawberry Twizzler) (Lincoln Logs), 7 (Mr. Potato (Cinnamon), 7 (Almond)
Head)

Cindy ..,i-PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 3 6.5 1.5 6.5 4.5 1.5 4.5 4.0
Leisure 11.5 11.5 11.5 9 8 11.5 14 11.0
Edible vs. 5 6.5 1 2.5 4 2.5 6.5 4.0
Olfactory 9 11 9 14 12.5 12.5 9 11.0
Leisure vs. 14 10 13 5 5 10 12 9.86
Olfactory 5 7 1.5 8 10 3 1.5 5.14










Table A-8. Group 8: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings
Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Montana 1 (Mint Tic Tac), 2 (Orange Tic 1 (Leap Pad), 2 (Coloring 1(Vanilla), 2 (Orange), 3
Tac), 3 (Almond), 4 (Hot Tamale), 5 Book), 3 (Reading Book), 4 (Strawberry), 4 (Mint), 5
(Coconut Jelly Bean), 6 (Vanilla (Race Cars), 5 (Legos), 6 (Coconut), 6
Wafer), 7 (Strawberry Twizzler) (Football Game), 7 (Castle) (Cinnamon), 7 (Almond)

Montana .-.'-PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 5.5 5.5 7 2.5 1 4 2.5 4.0
Leisure 14 12 13 8.5 10.5 10.5 8.5 11.0
Edible vs. 3 4.5 6.5 1.5 1.5 6.5 4.5 4.0
Olfactory 11 8.5 13 14 8.5 11 11 11.0
Leisure vs. 14 13 10 1.5 6 1.5 12 8.29
Olfactory 10 8 3.5 6 6 3.5 10 6.71


Table A-9. Group 9: Stimuli used in preference assessments during experiment 1 and individual
stimulus rankings
Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Ernie 1 (Mint Tic Tac), 2 (Orange Tic 1 (Caterpillar), 2 (Play Bus), 1(Vanilla), 2 (Orange), 3
Tac), 3 (Almond), 4 (Hot Tamale), 5 3 (Books), 4 (Race Cars), 5 (Strawberry), 4 (Mint), 5
(Coconut Jelly Bean), 6 (Vanilla (Legos), 6 (Dog Puppet), 7 (Coconut), 6
Wafer), 7 (Strawberry Twizzler) (Leap Pad) (Cinnamon), 7 (Almond)

Ernie .-.'-PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 6 13 8 5 2 3 1 5.43
Leisure 10 11.5 4 4 8 11.5 8 8.14
Edible vs. 5.5 9 3 5.5 1.5 4 1.5 4.29
Olfactory 12 12 12 12 12 7.5 7.5 10.71
Leisure vs. 5 2.5 8 2.5 1 6 6 4.43
Olfactory 12 12 12 9 12 6 2 9.29











Table A-10. Group 10: Stimuli used in preference assessments during experiment 1 and
individual stimulus rankings
Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Janet, 1 (Mint Tic Tac), 2 (Orange Tic 1 (Football Game), 2 1(Vanilla), 2 (Orange), 3
Geoff Tac), 3 (Almond), 4 (Hot Tamale), 5 (Coloring Book), 3 (Books), (Strawberry), 4 (Mint), 5
(Coconut Jelly Bean), 6 (Vanilla 4 (Race Cars), 5 (Legos), 6 (Coconut), 6
Wafer), 7 (Strawberry Twizzler) (Castle), 7 (Leap Pad) (Cinnamon), 7 (Almond)

Janet (PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 1.5 7 7 1.5 3.5 7 3.5 4.43
Leisure 11 5 13.5 13.5 9 11 11 10.57
Edible vs. 2 3.5 7 1 3.5 5.5 5.5 4.0
Olfactory 13 10 10 8 13 10 13 11.0
Leisure vs. 12 1 12 12 5 10 6 8.29
Olfactory 7.5 3.5 3.5 2 9 7.5 12 6.43


Geoff, ..I-PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7
Edible vs. 6.5 14 6.5 12.5 8.5 3.5 10.5 8.86
Leisure 1 3.5 12.5 10.5 8.5 2 5 6.14
Edible vs. 2 3.5 7 5 3.5 7 1 4.14
Olfactory 10 10 10 14 12.5 7 12.5 10.86
Leisure vs. 4 1 8 2 4 6 4 4.14
Olfactory 10.5 4 12.5 8 8 10.5 12.5 9.43




Table A- 1. Group 11: Stimuli used in preference assessments during experiment 1 and
individual stimulus rankings
Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli
Leyla 1 (Mint Tic Tac), 2 (Orange Tic 1 (Leap Pad), 2 (Books), 3 1(Vanilla), 2 (Orange), 3
Tac), 3 (Almond), 4 (Hot Tamale), (Magazines), 4 (Cars), 5 (Strawberry), 4 (Mint), 5
5 (Coconut Jelly Bean), 6 (Vanilla (Lincoln Logs), 6 (Dog (Coconut), 6
Wafer), 7 (Strawberry Twizzler) Puppet), 7 (Mr. Potato Head) (Cinnamon), 7 (Almond)

Leyla ..i-PWS)
Assessment Stimuli Mean
1 2 3 4 5 6 7


Edible vs.
Leisure
Edible vs.
Olfactory
Leisure vs.
Olfactory


2.5 6.5 1 6.5
8 10.5 10.5 13.'
14 11 1 3
5 11 4 7
1 10.5 9 13
6 4 4 7.5


5
4
7
11
10.5
4


4.93
10.07
7.0
8.0
10.0
5.0









LIST OF REFERENCES


Aime, P. (2007). Fasting increases and satiation decreases olfactory detection for a neutral odor
in rats. Behavioural Brain Research, 179(2), 258-264.

Akefeldt, A., & Gillberg, C. (1999). Behavior and personality characteristics of children and
young adults with Prader-Willi syndrome: A controlled study. Journal of the American
Academy of Child and Adolescent Psychiatry, 38, 761-769.

Bhutta, M. F. (2007). Sex and the nose: human pheromonal responses. Journal of the Royal
Society of Medicine, 100(6), 268-274.

Bojak, S. L., & Carr, J. E. (1999). On the displacement of leisure items by food during multiple-
stimulus preference assessments. Journal ofApplied Behavior Analysis, 32, 515-518.

Buck, L. (2000). Smell and taste: The chemical senses. In E. r. Kandel, J. H. Schwartz & T. M.
Jessell (Eds.), Principles of Neural Science (4th ed., pp. 625-645). New York: McGraw-
Hill.

Buck, L., & Axel, R. (1991). A novel multigene family may encode odorant receptors: A
molecular basis for odor recognition. Cell, 65, 175-187.

Burd, L., Vesely, B., Martsolf, J., & Kerbeshian, J. (1990). Prevalence study of Prader-Willi
syndrome in North Dakota. American Journal of Medical Genetics, 37, 97-99.

Butler, M. G. (1996). Molecular diagnosis of Prader-Willi syndrome: Comparison of cytogenetic
and molecular genetic data including parent of origin dependent methylation DNA
patterns. American Journal of Medical Genetics, 61, 188-190.

Butler, M. G., Meaney, F. J., Opitz, J. M., & Reynolds, J. F. (1987). An anthropometric study of
38 individuals with Prader-Labhart-Willi syndrome. American Journal of Medical
Genetics, 26, 445-455.

Butler, M. G., Theodoro, M. F., Bittel, D. C., & Donnelly, J. E. (2007). Energy expenditure and
physical activity in Prader-Willi syndrome: Comparison with obese subjects. American
Journal of Medical Genetics, 143A, 449-459.

Chu, S. (2008). Olfactory conditioning of positive performance in humans. Chemical Senses, 33,
65-71.

Clarke, D. J., Boer, H., & Webb, T. (1995). Genetic and behavioral aspects of Prader-Willi
syndrome: A review with a translation. Mental Handicap Research, 8, 38-53.

Davis, R. L. (2004). Olfactory learning. Neuron, 44(1), 31-48.

DeLeon, I. G., & Iwata, B. A. (1996). Evaluation of a multiple-stimulus presentation format for
assessing reinforcer preferences. Journal ofApplied 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 ofApplied Behavior Analysis, 30, 475-484.

Djordjevic, J. (2008). Olfaction in patients with mild cognitive impairment and Alzheimer's
disease. Neurobiology ofAginig. 29(5), 693-706.

Eibenstein, A. (2005). Olfactory screening test in mild cognitive impairment. Neurological
Sciences, 26(3), 156-160.

Egel, A.L. (1981). Reinforcer variation: Implication for motivating developmentally disabled
children. Journal ofApplied Behavior Analysis, 14, 345-350.

Fisher, W., Piazza, C. C., Bowman, L. G., Hagopian, L. P., Owens, J. C., & Slevin, I. (1992). A
comparison of two approaches for identifying reinforcers for persons with severe and
profound disabilities. Journal ofApplied Behavior Analysis, 25, 491-498.

Goldberg, D. L., Garrett, C. L., Van Piper, C., & Warzac, W. J. (2002). Coping with Prader-Willi
syndrome. Journal of the American Dietetic Association, 102, 537-542.

Gottschalk, J. M., Libby, M. E., & Graff, R. B. (2000). The effects of establishing operations on
preference assessment outcomes. Journal ofApplied Behavior Analysis, 33, 85-88.

Gunay-Aygon, M., Cassidy, S. B., & Nicholls, R. D. (1997). Prader-Willi and other syndromes
associated with obesity and mental retardation. Behavior Genetics, 27, 307-324.

Hanley, G. P., Iwata, B. A., Lindberg, J. S., & Conners, J. (2003). Response-restriction analysis:
I. Assessment of activity preferences. Journal ofApplied Behavior Analysis, 36, 47-58.

Hawkes, C. (2006). Olfaction in neurodegenerative disorder. Taste and Smell: an Update, 63,
133-151.

Ivancic, M. T. (2000). Stimulus preference and reinforcer assessment applications. In J. A. J. E.
Carr (Ed.), Handbook of applied behavior analysis (pp. 19-3 8). Reno: Context Press.

Jellinek, J. S. (2004). Proust remembered: Has Proust's account of odor-cued autobiographical
memory recall really been investigated? Chemical Senses, 29(5), 455-458.

Kettenmann, B. (2005). Odor and taste interaction on brain responses in humans. Chemical
Senses, 30, I1234-i235.

Lamb, A. S., & Johnson, W. M. (1987). Premature coronary artery atherosclerosis in a patient
with Prader-Willi syndrome. American Journal of Medical Genetics, 28, 873-880.

Maglieri, K. A., DeLeon, I. G., Rodriguez-Catter, V., & Sevin, B. M. (2000). Treatment of
covert food stealing in an individual with Prader-Willi syndrome. Journal ofApplied
Behavior Analysis, 33, 615-618.









McAdam, D. B., Klatt, K. P., Koffamus, M., Dicesare, A., Solberg, K., Welch, C., et al. (2005).
The effects of establishing operations on preferences for tangible items. Journal of
Applied Behavior Analysis, 38, 107-110.

McBurney, D. H. (2006). Olfactory comfort: Smelling a partner's clothing during periods of
separation. Journal ofApplied Social Psychology, 36(9), 2325-2335.

Mohr, C. (2001). Associations to smell are more pleasant than to sound. Journal of Clinical and
Experimental Neuropsychology, 23(4), 484-489.

Mombaerts, P. (2004). Love at first smell The 2004 Nobel Prize in Physiology or Medicine.
New England Journal of Medicine, 351(25), 2579-2580.

Morgan, C. D. (2002). Olfactory event-related potentials in Alzheimer's disease. Journal of the
International Neuropsychological Society, 8(6), 753-763.

Murphy, C. (1999). Loss of olfactory function in dementing disease. Physiology & Behavior,
66(2), 177-182.

Northrup, J., George, T., Jones, K., Broussard, C., & Vollmer, T. R. (1996). A comparison of
reinforcer assessment methods: The utility of a verbal and a pictorial choice procedure.
Journal ofApplied Behavior Analysis, 29, 201-212.

Olander, E., Stamberg, J., Steinberg, L., & Wulfsberg, E. A. (2000). Third Prader-Willi
syndrome phenotype due to maternal uniparental disomy 15 with mosaic trisomy 15.
American Journal of Medical Genetics, 93, 215-218.

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
Applied Behavior Analysis, 18, 249-255.

Page, T. J., Finney, J. W., Parrish, J. M., & Iwata, B. A. (1983). Assessment and reduction of
food stealing in Prader-Willi children. Applied Research in Mental Retardation, 4, 219-
228.

Page, T. J., Stanley, A. E., Richman, G. S., Deal, R. M., & Iwata, B. A. (1983). Reduction of
food theft and long-term maintenance of weight loss in a Prader-Willi adult. Journal of
Behavior Therapy and Experimental Psychiatry(l4), 261-268.

Reid, D. H., Everson, J. M., & Green, C. W. (1999). A systematic evaluation of preferences
identified through person-centered planning for people with profound multiple
disabilities. Journal ofApplied Behavior Analysis, 32, 467-477.

Richardson, B. E. (2004). Altered olfactory acuity in the morbidly obese. Obesity Surgery, 14(7),
967-969.









Rincover, A., & Newsom, C. D. (1985). The relative motivational properties of sensory and
edible reinforcers in teaching autistic children. Journal ofApplied 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 ofApplied Behavior Analysis, 31, 605-620.

Roscoe, E. M., Iwata, B. A., & Kahng, S. (1999). Relative versus absolute reinforcement effects:
Implications for preference assessments. Journal ofApplied Behavior Analysis, 32, 479-
493.

Seiden, A. M. (2001). The diagnosis of a conductive olfactory loss. Laryngoscope, 111(1), 9-14.

Stevenson, D. A., Heinemann, J., Angulo, M., Butler, M. G., Loker, J., Rupe, N., et al. (2007).
Deaths due to choking in Prader-Willi syndrome. American Journal of Medical Genetics,
143, 484-487.

Verhagen, J. V., & Engelen, L. (2006). Neurocognitive bases of human multimodal food
perception: Sensory integration. Neuroscience and Biobehavioral Reviews, 30, 613-650.

Wang, L., Walker, V. E., Sardi, H., Fraser, C., & Jacob, T. J. C. (2002). The correlation between
physiological and psychological responses to odour stimulation in human subjects.
Clinical Neurophysiology, 113, 542-551.

Wharton, R. H., Wang, T., Graeme-Cook, F., Briggs, S., & Cole, R. E. (1997). Acute idiopathic
gastric dilation with gastric necrosis in individuals with Prader-Willi syndrome.
American Journal of Medical Genetics, 73, 437-441.

Wilder, D. A., Schadler, J., Higbee, T. S., Haymes, L. K., Bajagic, V., & Register, M. (2008).
Identification of olfactory stimuli as reinforcers in individuals with autism: A preliminary
investigation. Behavioral Interventions, 23, 97-103.

Windsor, J., Piche, L. M., & Locke, P. A. (1994). Preference testing: A comparison of two
presentation methods. Research in Developmental Disabilities, 15, 439-455.

Wise, R. A. (2006). Role of brain dopamine in food reward and reinforcement. Philosophical
Transactions of the Royal Society B-Biological Sciences, 361(1471), 1149-1158.

Yeomans, M. R. (2006). Olfactory influences on appetite and satiety in humans. Physiology &
Behavior, 89(1), 10-14.

Zhao, K., Dalton, P., Yang, G. C., & Scherer, P. W. (2006). Numerical modeling of turbulent and
laminar airflow and odorant transport during sniffing in the human and rat nose.
Chemical Senses, 31, 107-118.









BIOGRAPHICAL SKETCH

Sarah Bloom earned an A.A. Degree from Simon's Rock College of Bard in 1996 and a

B.A. degree at the University of Washington in 1998. She was initially exposed to behavior

analysis as a long-term substitute teacher at the Sussex Consortium, part of the Delaware Autism

Program. Sarah began graduate training in the area of Behavior Analysis in the Psychology

Department at the University of Florida in 2003 under the supervision of Dr. Brian Iwata. She

earned her Masters Degree and became a Board Certified Behavior Analyst while pursing her

doctoral degree at UF. She has conducted research projects on olfactory stimuli as reinforcers in

individuals with Prader-Willi Syndrome, assessment and treatment of rumination and vomiting, a

trial-based approach to functional analysis of problem behavior, a work-place health and

wellness program, and acquisition of verbal behavior in children who engage in escape-

maintained problem behavior. She has presented her research at multiple state, national, and

international conferences. She has served as the site coordinator of Dr. Iwata's laboratory at a

local school for children with exceptional needs and has provided behavioral services at a local

branch of the ARC, a residential and day program for individuals with developmental disabilities

including Prader-Willi Syndrome. She has served as a teaching assistant and instructor of record

for undergraduate courses in applied behavior analysis. This fall Sarah will be an assistant

professor in the Special Education and Rehabilitation Department at Utah State University.





PAGE 1

1 ASSESSMENT OF PREFERENCE FOR OLFACT ORY STIMULI IN INDIVIDUALS WITH PRADER-WILLI SYNDROME By SARAH E. BLOOM A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Sarah E. Bloom

PAGE 3

3 To my dogs, Enzo and Crosby, for whom olfactory stimuli are very important.

PAGE 4

4 ACKNOWLEDGMENTS I thank Erin Ca mp, Natalie Rolider, Javier Virues-Ortega, Jennifer Hammond, and Leah Koehler for help with data collection. I tha nk Dr. Brian Iwata for hi s guidance and tireless assistance. I also thank Dr. Timothy Hackenbe rg, Dr. Hazel Jones, Dr. Neil Rowland and Dr. Timothy Vollmer.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF TABLES ...........................................................................................................................7LIST OF FIGURES .........................................................................................................................8ABSTRACT ...................................................................................................................... ...............9 CHAP TER 1 INTRODUCTION .................................................................................................................. 11Prader-Willi Syndrome ......................................................................................................... ..11Identification of Reinforcers in Indivi duals with Developm ental Disabilities ....................... 13Olfactory Stimuli ............................................................................................................. .......15Goals of Current Research ......................................................................................................172 EXPERIMENT 1 .................................................................................................................. ..18Purpose ....................................................................................................................... ............18Method ........................................................................................................................ ............18Subjects and Setting ........................................................................................................18Response Measurement and Reliability .......................................................................... 19Preference Assessment .................................................................................................... 19Results .....................................................................................................................................21Discussion .................................................................................................................... ...........233 EXPERIMENT 2 .................................................................................................................. ..31Purpose ....................................................................................................................... ............31Method ........................................................................................................................ ............31Subjects and Setting ........................................................................................................31Task Selection and Initial Training .................................................................................32Response Measurement and Reliability .......................................................................... 32Single-Schedule Reinforcer Assessment .........................................................................33Concurrent-Schedule Reinforcer Assessment ................................................................. 34Results .....................................................................................................................................34Discussion .................................................................................................................... ...........354 GENERAL DISCUSSION ..................................................................................................... 41APPENDIX: STIMULI USED IN PREFER ENCE ASSESSMENTS DURING EXPERIMENT 1 AND PREFERENCE RANKI NGS FOR EACH STIMULUS ................. 46

PAGE 6

6 LIST OF REFERENCES ...............................................................................................................55BIOGRAPHICAL SKETCH .........................................................................................................59

PAGE 7

7 LIST OF TABLES Table page 2-1. Experiment 1 demograhic information .................................................................................. 273-1. Reinforcer assessm ent type and tasks .................................................................................... 38A-1. Group 1: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..46A-2. Group 2: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..48A-3. Group 3: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..49A-4. Group 4: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..50A-5. Group 5: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..51A-6. Group 6: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..52A-7. Group 7: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..52A-8. Group 8: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..53A-9. Group 9: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..53A-10. Group 10: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..54A-11. Group 11: Stimuli used in preference asse ssments during experiment 1 and individual stimulus rankings ............................................................................................................. ..54

PAGE 8

8 LIST OF FIGURES Figure page 2-1 Percentage of trials on which each stimul us was selected du ring each of the pairedstimulus preference assessments ........................................................................................ 282-2 Mean preference rankings of stimulus classes for all PW S subjects from each of the paired-stimulus preference assessments ............................................................................292-3 Results obtained for all non-PWS subjects from each of the paired-stimulus preference assessments ...................................................................................................... 303-1 Results of the single-schedule reinforcer assessment for Bettina, Taneisha, Peter, Drake, Zeus, and Adeline. ................................................................................................393-2 Results of the concurrent -schedule reinforcer assessment for Bettina, Annette, Patsy, and Taneisha. .....................................................................................................................40

PAGE 9

9 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ASSESSMENT OF PREFERENCE FOR OLFACT ORY STIMULI IN INDIVIDUALS WITH PRADER-WILLI SYNDROME By Sarah E. Bloom August 2008 Chair: Brian Iwata Major: Psychology We examined the reinforcing e ffects of olfactory stimuli in individuals with Prader-Willi Syndrome (PWS). In Study 1, 30 individuals (20 diagnosed with PWS, 10 diagnosed with a developmental disability other than PWS) were exposed to a series of three paired-stimulus preference assessments involving ed ible, olfactory, leisure stimuli. In each assessment, 7 or 8 items representing 2 stimulus groups were included. Stimulus class hierarchies were generated based on the results of these preference assessment s. Nineteen of the 20 PWS subjects had edible stimuli at the top of their preference hierarchy (the remaining subject sh owed mixed preference for edible and leisure stimuli). Eight subjects demonstrated a stronger preference for leisure stimuli than for olfactory stimuli, and 10 s ubjects demonstrated a stronger preference for olfactory stimuli than for leisure stimuli. The re maining 2 subjects showed mixed preference for leisure and olfactory stimuli. Preferences were si milar for subjects without PWS. A subset of the subjects from Study 1 was included in Study 2, in which preferences were verified in either concurrent or single-schedule re inforcer assessments. Four subj ects for whom olfactory stimuli were the least preferred were exposed to a si ngle-schedule reinforcer assessment in which performance of a task resulted in access to olfactory stimuli. Increases in responding, relative to baseline, were observed in 3 subjects but not the fourth. The fourth subjects responding only

PAGE 10

10 increased when responses were re inforced with edible stimuli. Two subjects for whom olfactory stimuli were ranked below edible but above leisure, or for whom preference for olfactory stimuli and leisure stimuli was mixed, also were exposed to a single-schedule reinforcer assessment and then were exposed to a concurrent-schedule rein forcer assessment in which performance of one task resulted in access to olfactory stimuli and perfo rmance of another (identic al) task resulted in access to leisure stimuli. Two a dditional subjects only were exposed to the concurrent-schedule reinforcer assessment. Increases in responding du ring the single-schedule reinforcer assessment were observed. All four subjects allocated responses for both choi ces (olfactory and leisure) at approximately equivalent rates.These findings su ggest that olfactory stimuli may be a viable class of reinforcers for some individuals with PW S or for others for whom edible reinforcers are contraindicated.

PAGE 11

11 CHAPTER 1 INTRODUCTION Prader-Willi Syndrome Prader-W illi Syndrome (PWS) is a genetic diso rder with an estimated incidence in the United States ranging from 1 in 16,000 (Burd, Vesely, Martsolf, & Kerbes hian, 1990) to 1 in 25,000 (Butler, 1996). Physical feat ures associated by PWS include short stature, small hands and feet, poor muscle tone, and obesity (But ler, Meaney, Opitz, & Reynolds, 1987). In addition, some degree of cognitive impairment and beha vior problems, including self-injury, sleepabnormalities, and disruptive behavior, have been observed (Clarke, Boer, & Webb, 1995). Individuals with PWS are typically obese due to excessive eating or hyperphagia (Greenswag, 1987). Attempts to control this obe sity often include restricted diets, thus eliminating food items from the pool of stimuli that may be used as reinfo rcers. It may be possible that olfactory stimuli, which share sensory properties with food, may effectively be us ed as reinforcers with this population. The general purpose of the current study is to exam ine the relative efficacy of olfactory stimuli to function as reinforcers. Prader-Willi Syndrome (PWS) results most frequently from paternal deletion of band 15q11-13 on chromosome 15 (Olander, Stamberg, Steinberg, & Wulfsberg, 2000). One reported potential contributor to obesity is reduced resting energy expend iture, possibly as a result of hypotonia (Butler, Theodoro, Bittel & Donnelly, 2007). A more direct contributor to obesity is excessive consumption of food (hyperphagia) and is a serious medical concern (Gunay-Aygon, Cassidy, & Nicholls, 1997). When typical food sources are restricted, individuals with PWS have been observed eating garbage or other inedible items (Russell & Oliver, 2003). Overeating was documented in 98% of individuals with PWS in a sample of 91 children and adults with PWS (Dykens & Shah, 2003). Morbid obesity has been documented with related coronary disease

PAGE 12

12 and/or non-insulin dependent diabetes mellitus in individuals with PWS (Lamb & Johnson, 1987). Although infants with PWS are often diagnosed with feeding problems (Gunay-Aygun, Schwartz, Heeger, O'Riordan, & Cassiday, 2001) eating increases exponen tially and eventually qualifies as hyperphagia, which worsens across the life-span of individuals with PWS (Akefeldt & Gillberg, 1999). Hyperphagia can lead to fatal gastric necrosis following severe gastric distention (Wharton, Wang, Graem e-Cook, Briggs, & Cole, 1997). In a survey documenting cause of death in individuals with PWS, choking on food was implicated in 7.9% of the 152 individuals (Stevenson et al., 2007). Problems related to food consumption are implic ated in a variety of challenging behaviors observed in individuals with PWS. Excessive food consumption and food-seeking/food-stealing behaviors are commonly targeted in behavioral interventions (Magli eri, DeLeon, RodriguezCatter, & Sevin, 2000; Page, Finney, Parrish, & Iwata, 1983; Page, Stanley, Richman, Deal, & Iwata, 1983). Maglieri et al. pa ired punishment (a verbal reprimand) with warning stimuli, which resulted in a decrease in consumption of prohibited foods (2000). Reinforcement for not stealing food has also been examined (Page, Finney et al., 1983), as have interventions that combine differential reinforcement and puni shment (Page, Stanley et al., 1983). Because individuals with PWS eat to the point of dangerous excess, they often are placed on extremely restricted dietary regimens (Goldberg, Garrett, Van Piper, & Warzac, 2002). Unlike diet programs that merely prescribe li mited caloric intake, di etary management in individuals with PWS includes re striction of opportuni ties to consume food. The nature of these restrictions can be rather extreme, including no t permitting the person with PWS to be alone in the community in the presence of food and elimin ating access to food in the home environment. Families of children with PWS are counseled to put padlocks on pantries and refrigerators and to

PAGE 13

13 restrict access to all potential food sources in orde r to keep calories consumed to approximately 1200 per day (Dykens & Shah, 2003). Group home placements for individuals with PWS include similar restrictive precautions. The quantity of food consumed by individuals with PWS is restricted and carefully monitore d. Finally, although there is ample evidence that edible items are highly preferred by individuals wi th PWS, food is not recommende d for use as a reinforcer in training or treatment programs due to its potentia l to undermine restricted-calorie diets. This presents challenges in establ ishing and maintaining compliance with other therapeutic interventions, such as exercise, menu planni ng, and academic or work performance because an entire yet presumably potent class of reinforcers is prohibited. Classes of stimuli other than food would be less problematic, if they can function as reinforcers. Identification of Reinforcers in Indivi duals with Develop mental Disabilities Inclusion of effective reinforcers in training programs is essential to the acquisition and maintenance of adaptive behavior Three general approaches used to identify reinforcers in individuals with developmental disabilities include the interv iew, direct observation, and systematic assessment of preferen ce. The interview involves simp ly asking individuals or their care providers what stimuli are preferred. Although easy to administer, interviews have been shown to be unreliable or inaccurate in identifying effective reinforcers (Reid, Everson, & Green, 1999). Direct observation typically attempts to identify how individuals spend their free time. Data are recorded on duration of involvement in various activities and are used as the basis for selecting some activities as reinforcers for engaging in other (target) activ ities. This method is more objective than the interview but potentially very time consuming. Also, individuals may have very limited access to potential reinforcers in their environments, thus restricting the pool of stimuli included in the assessment. Due to limitations with both the interview and directobservation methods, most research ers have adopted more systema tic procedures in which an

PAGE 14

14 array of stimuli (in various formats) is system atically presented to the subject, and data are collected on some measure of preferen ce (either approach or selection). Pace, Ivancic, Edwards, Iwata, and Page (1985) presented a two-stage process for identifying reinforcers: an initial assessment of preference followed by a test for reinforcer efficacy. They first presented a series of stimu li one at a time (single-stimulus presentation) and measured approach behavior. Sixteen stimuli, 2 from each of 8 classes (visual, auditory, olfactory, edible, tactile, therma l, kinesthetic, and social) were presented ten times each. They then used stimuli approached on high vs. low per centages of assessment trials as consequences for performance on learning tasks and observed that stimuli identified as more highly preferred during the preference assessment functione d as more effective reinforcers. Subsequent research on reinforcer identif ication has adopted the two-stage process illustrated by Pace et al. (1985) and has included a number of procedural variations (see Ivancic, 2000, for a review). For example, stimulus pr esentation during preference assessments has included the paired-stimulus variat ion (Fisher et al., 1992), in whic h a selection is made from 2 stimuli presented during each trial, as well as multiple-stimulus formats (DeLeon & Iwata, 1996; (Windsor, Piche, & Locke, 1994), in which selections are made from larger arrays of stimuli. Another variation has involved the mode of stimulus presentation. Most preference assessments involve presentation of the actual stimulus to be used as a reinforcer (i.e., an edible or leisure item). However, pictorial representations of st imuli also have been used occasionally (Hanley, Iwata, & Lindberg, 1999; Northrup, George, Jones, Broussard, & Vollmer, 1996). Finally, responding has been measured as approach behavior during stimulus presentation in discrete trials, as in each of the above examples, or free-operant engagement when all stimuli are

PAGE 15

15 available throughout the session (Hanley, Iwata, Lindberg, & Conners, 2003; Roane, Vollmer, Ringdahl, & Marcus, 1998). Tests for reinforcement effects (the second stag e of reinforcer identification) also have varied along a number of dimensions. The two most commonly used procedures are those in which the effects of a particular stimulus ar e evaluated by presenting it following a target response (single schedule; Pace et al., 1985) and those in which the effects of two competing stimuli are evaluated by presenting one stimulus for each of two concurrently available target responses (concurrent schedul e; Fisher et al., 1992). One finding in research on preference assessmen t, which is relevant to the current studies although perhaps not surprising, is that some classes of stimuli seem to be preferred over others. In an early example, Rincover and Newsom (1985) found that sensory re inforcers (objects that provided sensory stimulation) maintained res ponding better than did food reinforcers in participants who were diagnosed with autism. By contrast, results of subsequent research with participants having developmental disabilities ot her than autism have shown a strong preference for edible items. For example, DeLeon, Iwata, and Roscoe (1997) first identified highly preferred edible items in an assessment that included only edible items and then identified highly preferred leisure items in an assessment that incl uded only leisure items. They then combined the top-ranked stimuli from the edible-only and leisure-only assessments in a third preference assessment and observed that the edible items routinely outranked the leisure items, a finding that was replicated by Bojak and Carr (1999). Olfactory Stimuli Although not often the f ocus of research in be havior analysis, olfactory stimuli are an important part of the lives of hu mans and other animals. Responses to olfactory stimuli are broad and dramatic in range, from strong preference for particular stimuli to acute aversions to others.

PAGE 16

16 Olfactory stimuli have an important role across the lifespan. For example, neonates have consistent preferences for natural breast odors and select unwashed breasts over washed breasts (Varendi, 1997). Thus, olfactory st imuli may have central roles in the regulation of diverse and critical behavioral repertoires. Olfactory stimuli have been said to evoke pleasant memories (Jellinek, 2004; Mohr, 2001) and have been implicated in animal and insect mate selection (Bhutta, 2007; Buck & Axel, 1991; Mombaerts, 2004). Humans may even engage in be havior to access olfactor y stimuli associated with their partners during periods of sepa ration (McBurney, 2006). Food-related olfactory stimuli have been implicated in operant pa radigms involving dopamine reinforcement of foodseeking behavior (Wise, 2006) and appetitive be havior in general (Aime, 2007; Yeomans, 2006). In classical conditioning paradigms, exposur e to olfactory stimuli has increased neural responsiveness to conditioned odors (Davis, 2004). In an interesting study conducted with school children, classical conditioning w ith an ambient odor and acade mic success increased academic performance on other tasks when children were exposed to the same odor (Chu, 2008). Olfactory stimuli share sensor y characteristics with food, and both the gustatory and olfactory systems are stimulated when exposed to edible and/or olfactory stimuli (Kettenmann, 2005; Verhagen & Engelen, 2006). This relationship suggests that olfactory stimuli may represent a stimulus class especi ally effective for use with in dividuals with PWS. Previous research on preference has included some olfactor y stimuli such as coffee grounds and hibiscus flower (Fisher et al., 1992; Pace et al., 1985), but the general efficacy of olfactory stimuli as reinforcers has been directly a ssessed only recently in a study on individuals with autism (Wilder et al., 2008).

PAGE 17

17 Goals of Current Research Though it may be assumed that individuals with PW S would show a general preference for edible stimuli over others, it is unknown how olfact ory stimuli might fare if pitted against edible and/or leisure stimuli. It is po ssible that olfactory stimuli may effectively compete with leisure stimuli. Regardless of relative preference for edib le, leisure, or olfactory stimuli, it is possible that olfactory stimuli may function as reinforc ers, especially in populations that value food highly, like those diagnosed with PWS. Given that little is known about preference for olfactory stimuli or about the efficacy of olfactory stimuli as reinforcer s, further research about the potential role of olfactory stimuli in behavioral acquisi tion is warranted. The present study examined preferences in i ndividuals with Prader-Willi Syndrome. The purposes were to determine (a) re lative preference for edible, leisur e, and olfactory stimuli, and (b) whether olfactory stimuli would function as reinforcers for this population. An additional purpose was to examine relative preference for edible, leisure, and olfactory stimuli in individuals who did not have Prader-Willi syndrome

PAGE 18

18 CHAPTER 2 EXPERIMENT 1 Purpose The purpose of this study was to exam ine pr eference for edible, leisure, and olfactory stimuli in individuals with PWS. We replicated previous studies that compared preference for edible versus leisure stimuli (Bojak & Carr, 199 9; DeLeon et al., 1997) and extended that work by including a separate assessment of olfactory stimuli using a wider range of stimuli than has been examined previously. We also compared preference for the three classes of stimuli in individuals who have PW S with those who do not. Method Subjects and Setting Twenty individuals with PW S and 10 individuals with a developmental disability other than PWS served as subjects. All of the PWS s ubjects were on restricted diets as part of their medical and behavior programs. Both children an d adults were included, and their level of functioning ranged from severe to mild mental retardation. Subject demographic information is listed in Table 2-1. Subjects were selected fo r inclusion based on their PWS or non-PWS status, availability, absence of food refu sal or feeding problems, and abse nce of problem behavior likely to interfere with sessions. Sessi ons were conducted at a day activity program for individuals with developmental disabilities or at a school for children with exceptiona l needs. Subjects run at the day activity program included 5 individuals with a developmental disability other than PWS (Juan, Wanda, Mitchell, Jeffrey, and Mikey), and 17 subjects with PWS (Patsy, Trina, Annette, Bettina, Penelope, Misty, Jewel, Jacob, Taneisha, Jamal, Dorian, Drake, Zeus, Peter, Adeline, Barnaby, and Melissa). Subjects ru n at the school for children w ith exceptional needs included 5 individuals with a developmenta l disability other than PWS (Geoff, Montana, Cindy, Ernie, and

PAGE 19

19 Leyla), and 3 subjects with PWS (Janet, Kata rina, and Manuel). PWS subjects experienced calorie-restricted diets and contro lled access to food in both settings. Foods consumed as part of this project were not included in the calorie plans for any subject Sessions were conducted two to three times per day, four to five days per week. Response Measurement and Reliability Observers recorded the occurrence of approach responses to one of two item s presented during preference assessments. A percentage se lection score was given to each stimulus by dividing the number of trials on which a stimulus was approached by the number of trials on which it was present and multiplying by 100% An independent observer collected data dur ing a mean of 73.7% of all preference assessments (range, 26.4% to 100%). Observers reco rds were compared on a trial-by-trial basis, and an agreement was scored if the two observers marked the same item as selected by the subject on a given trial. Interobserver agreement was calcu lated by dividing the number of agreements by the total number of trials and multiplying by 100%. The mean agreement score across all participants was 98.8% (range, 96.0% to 100%) during the preference assessments. Preference Assessment Three paired-stim ulus preference assessments (Fisher et al., 1992) were conducted across several days. The purpose of the assessments was to identify relative preference for stimulus classes. Pairedstimulus prefer ence assessments are better suited to determine relative preference rather than absolute preference because a preferen ce hierarchy is established based on a series of selections in which each stimulus is pitted agains t every other stimulus (Fisher et al., 1992). Each assessment involved the presentation of two classe s of stimuli. One assessment included edible and leisure stimuli, another included edible and olfactory stimuli, and another included leisure and olfactory stimuli. The order of the preference assessments was not consistent across subjects.

PAGE 20

20 Each stimulus class consisted of 7 or 8 stimuli (an equal number of stimuli from each class was included in each assessment). Edible stimuli and olfactory stimuli were paired in flavor. For example, if a mint-flavored candy was included in the edible stimulus array, a mint fragrance was included in the olfactory a rray. Leisure stimuli included activities that were reported as preferred by the subjects care provider and were age-appropriate or approp riate to the level of functioning of each subject. Leisure stimu li were not consistent across subjects. Prior to each assessment, subjects were allowed access to each of the items. Access consisted of tasting each edible stimulus and inte racting with each leisure and olfactory stimulus for 30-s. During each trial, two items were pres ented, and the subject was allowed to select one item. Approach responses were defined as the subject reaching for, touching, or picking up an item within 5 s of a verbal prompt to Pick one. If the subject attempted to select both of the items, the items were re moved and re-presented. Selection of an edible stimulus resulted in delivery of that stimulus; selection of an olfactory stimulus or a leisure stimulus resulted in 30-s access to the chosen stimulus. Olf actory stimuli were delivered by allowing the subject to hold to their nose a glass spice jar with a sealed top with small holes or a film canister with a cotton ball soaked with aromatic/essential oil or with scented-candle shav ings inside. Subjects did not always hold the olfactory stimulus to their nose for the entire 30 -s access period. Some subjects held it to their nose and then put it back down on the table and waited for the access period to end. Other subjects alternated be tween holding it to their nose a nd holding it about a foot away from their nose for the 30-s access period. A few s ubjects held it to their nose for the entire 30-s access period.

PAGE 21

21 Trials were arranged by assigning each stimul us a number and then using those numbers to generate a random sequence of number pairs. The number pairs were used to order trials. No single stimulus was presented consecutively. Results Subjects with PW S almost exclusively pref erred edible stimuli to both leisure and olfactory stimuli. Beyond their shared preference for edible stimuli, three general patterns of responding were observed: (a) preference for leis ure over olfactory stimuli, (b) preference for olfactory over leisure stimuli, and (c) mixed pref erence for olfactory and le isure stimuli. Figure 2-1 shows results of the prefer ence assessments for Peter, Annette, and Taneisha, 3 subjects whose responding illustrated each of these patterns. Data are expresse d as the percentage of trials on which each stimulus was select ed. Peter showed exclusive preference for edible stimuli in that he selected every edible stimulus more often than every leisur e stimulus (top left) and every olfactory stimulus (middle left). In the assessment of leisure vs. olfactory stimuli (bottom left), he showed a strong preference for leisure stimuli: The four stimuli he selected most frequently (slinky, bear, bumble ball and workbook) were leisure items, and the only leisure stimulus selected less frequently than olfactory stimuli was the puzzle, which was chosen less than any other stimulus. Annettes results are shown in the center panels Like Peter, she preferred all edible stimuli over all leisure stimuli (top center) and all ol factory stimuli (middle center) Unlike Peter, however, Annette showed a strong preference for leisure stimuli over olf actory stimuli (bottom center) in that every olfa ctory stimulus outranked every leisure stimulus. Taneishas results are shown in the right pa nels. Her preference for edible stimuli over leisure stimuli (top right) was evident but not ex clusive because there were two exceptions. Her highest ranked stimulus was puzzle, a leisure stimulus, and workbook, another leisure stimulus,

PAGE 22

22 outranked 3 edible stimuli. She showed exclusive preference fo r edible stimuli over olfactory stimuli (middle right), although the bottom ranked edible stimulus (cinnamon altoid) tied the top ranked olfactory stimulus (honey nut). Taneishas preference for olfactory versus leisure stimuli (bottom right) was mixed. The top 2 stimuli we re leisure stimuli (pu zzle and workbook), but a leisure stimulus (beads) and an olfactory stimulus (strawberry) tied for the third spot. The remainder of her selections showed a slig ht preference for olfactory stimuli. Based on results (percentage selection sc ores) obtained from the three preference assessments, a numerical rank was assigned to each stimulus in each assessment. The item selected most often was given a rank of 1, the item selected second most often was given a rank of 2, and so on. If two or more items were sel ected equally often, all were given the same rank obtained by adding the ranks for which there were a tie and dividing by the number of tied items. A mean rank for each stimulus class was obtained by adding the individual rankings for stimuli in that class and dividing by th e number of stimuli of that cl ass that were included in the assessment (either 7 or 8). The individual ranks for each stimulus in each assessment are listed in Appendix A. Figure 2-2 shows results obtained for the 20 PWS subjects, expressed as the mean preference rank for each stimulus class in each assessment. Nineteen of the 20 PWS subjects preferred edible stimuli to leisure stimuli (l eft panel). The exception, Dorian, showed mixed preference for edible and leisure stimuli. All 20 s ubjects preferred edible stimuli to olfactory stimuli (middle panel). Less consistent preferen ces across subjects were observed in the leisure vs. olfactory assessment (right panel): Ten subj ects (Janet, Katarina, Patsy, Trina, Annette, Bettina, Penelope, Misty, Jewel, and Jacob) preferred olfactory stimuli, 2 subjects (Taneisha and

PAGE 23

23 Jamal) had mixed preference for olfactory and le isure stimuli, and 8 subjects (Dorian, Manuel, Drake, Zeus, Peter, Adeline, Barnaby, and Melissa) preferred leisure stimuli. Figure 2-3 shows mean preference rankings for the non-PWS subjects, which reflected preferences similar to those obt ained with the PWS subjects. One of the 10 non-PWS subjects (Geoff) preferred leisure stimuli over edible s timuli (left panel); the remaining 9 subjects (Montana, Cindy, Ernie, Juan, Le yla, Wanda, Mitchell, Jeffrey, and Mikey) preferred edible stimuli. Seven of the 10 subjects (Montana, Cindy, Geoff, Ernie, Wanda Jeffrey, and Mikey) preferred edible stimuli to olfactory stimuli (mid dle panel); 1 subject (Jua n) preferred olfactory stimuli; the remaining 2 subjects (Leyla and Mitc hell) showed mixed preference for edible and olfactory stimuli. Finally, 6 subjects (Montan a, Cindy, Juan, Leyla, Wanda, and Mitchell) preferred olfactory stimuli over leisure stimuli (r ight panel); the remaining 4 subjects (Geoff, Ernie, Jeffrey, and Mikey) preferred leis ure stimuli to olfactory stimuli. Discussion Results of this study showed that bot h the PWS and non-PWS subjects showed a preference f or edible stimuli. In 39 out of 40 co mparisons, PWS subjects pr eferred edible stimuli over leisure stimuli (19 of 20 asse ssments) and over olfactory stim uli (20 out of 20 assessments). Preference for food by non-PWS subjects also was generally consistent. : 9 of 10 non-PWS subjects preferred edible stimuli to leisure stim uli, and 9 of 10 subjects preferred edible stimuli to olfactory stimuli. Two of the 10 subjects showed mixed preference for edible and leisure stimuli. Finally, when edible stimuli were excl uded as selections in the leisure vs. olfactory assessment, preferences by both PWS and nonPWS subjects were mixed in about equal proportion. The most noticeable similarity in the preferen ces of subjects with a nd without PWS is the outcome of the assessments that included edible stimuli. Preference for edible stimuli in the PWS

PAGE 24

24 subjects was very clear. It is unknown whether this di fference is related to th e clinical features of PWS (i.e. food is the most valuable commodity and is always preferred regardless of what other stimuli are available) or to other situational variables. For example, all of the subjects with PWS participated in a treatment program in which acc ess to food was severely restricted. Subjects had strict caloric regimes, and all foods consumed conformed to a calorie plan and met other nutritional needs. Therefore, cert ain high-calorie foods like snack f oods were fairly novel. Very small servings of snack foods compromised many of the selections in the preference assessments. Deprivation from food in general an d from these specific items in particular may have served as an establishing operation that increased the value of edible items and could have influenced preference assessment outcomes. Previ ous research has shown that deprivation from and satiation to specific stimuli influences choice during preference a ssessments (Gottschalk, Libby, & Graff, 2000; McAdam et al., 2005). Th e diets of the non-PWS subjects were not restricted, so deprivation from food items would not have been a relevant establishing operation. It would not have been possible to eliminate th is potential establishing operation in our PWS subjects because it is clinically undesirable to provide unrestricted access to food to individuals with PWS. The outcomes of the preference assessments in this study were consistent with previous research suggesting that edible stimuli represent a highly preferred stimulus class (Bojak & Carr, 1999; DeLeon, Iwata, & Roscoe, 1997). In the study conducted by DeLeon et al., some particular leisure item(s) outranked some edible item(s) for 5 of 14 subjects, but th is variability was not enough to alter overall preference fo r edible stimuli as a class. Exclusive edible preference was observed in the Bojack and Carr study. Our data were only slightly different and suggest that leisure stimuli may be preferred over edible stim uli for some individuals (see data for Geoff).

PAGE 25

25 Our data also document preference for olfactory s timuli over edible stimuli or equal preference for olfactory stimuli and edible stimuli (s ee data for Juan, Le yla and Mitchell). Strong preference for a stimulus class like edib le items, as reflected in mean rankings, does not necessarily apply to all stimuli in a class beca use of the influence of outliers. Differences in rankings within each stimulus class were observed that are not readily apparent when examining mean ranks. In some cases, preference for a s timulus class was influenced by a single outlying stimulus. Differences between mean ranks across each of the stimulus classes included in a given assessment varied greatly (from 0.14 to 8.0). It should be noted that some individuals who preferred edible stimuli to leisure stimuli as general stimulus classes preferred individual leisure stimuli over all other stimuli included in the assessment. This information may be relevant when designing treatment programs for individuals with PWS. For example, in the edible vs. leisure assessment, Taneishas highest ranked item was pu zzle. This leisure item might be used as a reinforcer in an intervention targeting food st ealing, because her preference assessment outcome data suggested that puzzles effectivel y competed with edible stimuli. Also of note, olfactory stimuli were preferred to leisure stimuli by approximately half of all subjects. Olfactory stimuli were even preferred to edible stimu li for 1 non-PWS subjects (Juan). Preference for olfactory and edible stimuli was mixed for 2 additional non-PWS subjects. Results of the leisure versus olfactory assessment are pe rhaps the most interesting from the standpoint of reinforcer identification for individuals with PWS. Given that 10 of the 20 PWS subjects preferred olfactory over leisure stimuli and an additional 2 subjects showed mixed preference for olfactory and leisure stimuli, olfactory stim uli may function as reinforcers for adaptive performances. It is possible that olfactory s timuli may be reinforcers even for subjects who showed a lack of preference for olfactory stimu li relative to leisure stimuli because olfactory

PAGE 26

26 stimuli may have some reinforcing value in cont exts where no choice is provided. For example, a typical training context involves a single response and a single cons equence, rather than a choice between multiple reinforcement options. In this type of setting, it is possible that olfactory stimuli may have reinforcing valu e. Overall, the results of Expe riment 1 suggest that further examination of olfactory stimuli, specifically, the ability of olfactor y stimuli to reinforce adaptive behavior, is warranted.

PAGE 27

27 Table 2-1. Experiment 1 demograhic information Name Gender Age Diagnoses Janet Female 19 PWS, Moderate MR Katarina Female 15 PWS, Mild MR Patsy Female 55 PWS, Moderate MR Trina Female 30 PWS, MR level not specified Annette Female 37 PWS, Mild MR Bettina Female 35 PWS, Mild MR Penelope Female 37 PWS, Mild MR Misty Female 32 PWS, Mild MR Jewel Female 31 PWS, Moderate MR Jacob Male 22 PWS, Mild MR Taneisha Female 30 PWS, Mild MR Jamal Male 38 PWS, Mild MR Dorian Male 22 PWS, Mild MR Manuel Male 19 PWS, MR level not specified, ADHD Drake Male 30 PWS, Mild MR Zeus Male 31 PWS, Moderate MR Peter Male 45 PWS, Mild MR Adeline Female 32 PWS, Moderate MR Barnaby Male 21 PWS, Moderate MR Melissa Female 46 PWS, Mild MR Geoff Male 13 Autism Montana Male 12 Autism Cindy Female 20 Moderate MR Ernie Male 3 PDD-NOS, Speech and Language Delay Juan Male 37 MR level not specified, Seizure Disorder Leyla Female 17 Mild MR Wanda Female 49 Moderate MR Mitchell Male 24 Mild MR Jeffrey Male 29 Moderate MR Mikey Male 38 Klinefelte rs Syndrome, Mild MR Subjects are in the left column. Gender is in the next colu mn. Age is in the next column. Diagnosis and/or level of mental retardation (MR) are in the right column.

PAGE 28

28 Figure 2-1. Percentage of trials on which each stimulus was selected during each of the pairedstimulus preference assessments, edible vs. leisure (top), edible vs. olfactory (middle), and olfactory vs. leisure (bottom) for Peter (left), Annette (middle), and Taneisha (right).

PAGE 29

29 Figure 2-2. Mean preference rankings of stimulus classes for all PWS subjects from each of the paired-stimulus preference assessments: edible vs. leisure (left), edible vs. olfactory (middle), and olfactory vs. leisure (right).

PAGE 30

30 Figure 2-3. Results obtained for all non-PWS subjects from each of the paired-stimulus preference assessments: edible vs. leisure (left), edible vs. olfactory (middle), and olfactory vs. leisure (right).

PAGE 31

31 CHAPTER 3 EXPERIMENT 2 Purpose The purpose of Experiment 2 was to dete rm ine whether preferences observed in Experiment 1 were predictive of reinforcement ef fects when olfactory stimuli were delivered as consequences. Because there were differences in PWS subjects preference for olfactory vs. leisure stimuli in Experiment 1, tw o different tests were used in Experiment 2. One test involved the delivery of olfactory stimuli for responding under a single schedule of reinforcement to determine whether they could function as reinforcers. If Experiment 1 results showed that olfactory stimuli were more preferred than leisur e stimuli (or equally preferred), reinforcement effects were evaluated under a concurrent schedule in which subj ects could respond for leisure or olfactory reinforcers. This te st was designed to determine wh ether olfactory stimuli could compete with leisure stimuli as reinforcers. Method Subjects and Setting Eight individuals with P WS from Experiment 1 served as subjects in Experiment 2. A subset of subjects for whom olfactory stimuli were more preferred than leisure stimuli (Bettina), for whom preference for olfactory stimuli and leis ure stimuli was mixed (Taneisha), or for whom olfactory stimuli were less preferred than leisure stimuli (Zeus, Peter, Drake, Adeline), underwent a single-schedule assessment to determ ine if olfactory items had any reinforcing value. Two of those subjects (Bettina and Taneisha) and an additional subset of subjects for whom olfactory stimuli were more preferred th an leisure stimuli (Annette and Patsy) underwent a concurrent-schedule assessment to verify thei r preference for olfactory items. Subjects were included based on whether or not they had PW S and their preferences from Experiment 1.

PAGE 32

32 Sessions were conducted at a day activity pr ogram for individuals with developmental disabilities or at a school for children with ex ceptional needs. All sess ions were 10 min in duration and were conducted two to three ti mes per day, four to five days per week. Task Selection and Initial Training A task was selected for each subject to se rve as th e basis for evaluating reinforcement effects. Tasks differed somewhat across s ubjects (see Table 3-1) but had the following characteristics: (a) they were vo cational or pre-vocational in natu re, (b) they involved discrete responses that were easily obser ved, (c) the responses were free ope rant in the sense that they could occur repeatedly throughout a session, and (d) the responses could be performed by the subject already or, if not, could be learned quickly If the subject was unable to perform the task at the beginning of the experiment, training was provided. Training consisted of the experimenter initially modeling the response and then asking the subject to engage in the response. Feedback was provided during training if the task was perf ormed incorrectly. If a response was not easily trained in less than 10-min, a new response was selected. During training, performance of the target behavior resulted in access to one of f our moderately preferred (middle ranked and not used during the experiment) leisure or olfactory items id entified during the preference assessment on a fixed-ratio (FR) 1 schedule. Response Measurement and Reliability Target responses consisted of sim ple vocatio nal or pre-vocational ta sks (e.g., placing cards in envelopes). Observers recorded the freque ncy of target responses and the delivery of reinforcers using handheld computers. An independent observer collected data during a mean of 47% (range, 20%-70.5%) of all rein forcer assessment sessions. To assess reliability, session time was divided into 10-s intervals, and observers records were compared on an interval-by-interval basis. Proportional agreement wa s calculated by dividing the smalle r number of responses scored

PAGE 33

33 in each interval by the larger number of responses, summing these fractions, and multiplying by 100%. The mean agreement score across all part icipants was 98% (range, 86.7%-100%) during the reinforcer assessments. Response rates we re corrected for reinforcer access time. Single-Schedule Reinforcer Assessment The assessm ent consisted of presentation of a ta sk identical to that used during training. A single task was placed on a table in front of the subject. No other materials were present. Prior to each session, the subject was prompted to pe rform the task once and was exposed to the contingency in place for task completion during that condition. The subject was free to do nothing or engage in the task during all sessions. Baseline. No programmed contingencies were in place for task completion. Olfactory reinforcement. The olfactory stimuli avai lable contingent upon task performance was placed directly behind the tas k, visible to the subjec t. Task performance resulted in 30-s access to 1 of 3 olfactory it ems that had been identified as preferred during Experiment 1 (fixed-ratio 1 schedule [FR-1]). Leisure and edible reinforcement. If responding in the olfactory reinforcement condition did not increase relative to baseline, a leisure reinforcement condition was conducted. If responding still did not increase in the leisure reinforcement condi tion, an edible reinforcement condition was conducted. The purpose of these addi tional conditions was to determine whether any reinforcement effect could be obtained, given that olfactory stimuli had been ineffective. The stimuli (leisure or edible) available contingent upon task performance was placed directly behind the task, visible to the subject. In the leisure-reinforcement cond ition, task performance resulted in 30-s access to 1 of 3 leisure items that had be en identified as preferred during Experiment 1 (FR-1). In the edible-reinforcemen t condition, task performance resu lted in delivery of 1 piece of 1 of 3 items that had been identified as preferred during Experiment 1 (FR-1).

PAGE 34

34 Concurrent-Schedule Reinforcer Assessment The assessm ent consisted of concurrent presenta tion of two tasks, identical to those used during training. One task was placed on the right side a rectangular table and the other task was placed on the left side of the table. Task positio ns alternated across sessions, and a task did not stay on a particular side for more than 2 consecu tive sessions. Task materials were covered with or placed on different colore d construction paper to serve as discriminative stimuli for responding associated with each task. Also, the stimuli available conti ngent upon each task was placed directly behind each task, visible to the subject. Performance of one task produced access to one of the top three olfactor y stimuli (identified in the olf actory vs. leisure assessment during Experiment 1); performance of the other task pr oduced access to one of the top three leisure stimuli (identified in the olfactory vs. leis ure assessment during Experiment 1). Stimulus presentation consisted of placing all 3 stimuli within arms reach of the subject, and then allowing the subject to select 1 stimulus. Prior to each session, the participant was prompted to perform each task once and received 30-s access to the associated reinforcer. During each session, performance of each task resulted in 30-s access (FR-1) to the stimulus associated with that task. At the start of each se ssion, the therapist told the subject that they could do either task or do nothing at all. Results Figure 3-1 shows results of the single-schedu le r einforcer assessments. Bettina did not respond during the first baseline condition but re sponded during five out of six sessions during the first olfactory reinforcement condition. Her responding quickly fell to zero during the second baseline condition, and she responded steadily during the second olf actory reinforcement condition. Taneisha, Peter,Drake, and Adeline ne ver responded during baseline; Zeus responded during only one of his five baseline sessions. Taneisha, Pete r, Zeus, and Adeline showed

PAGE 35

35 increases in responding during the olfactory-rein forcement condition. By contrast, Drake did not emit any responses during the olfactory-reinforcement condition, nor did he respond in the subsequent leisure-reinforcement condition. His responding increased only when edible items were delivered as reinforcers, which raised th e question of whether responding might maintain under olfactory reinforcement. Drakes res ponding ceased entirely, however, when olfactory reinforcement was reinstated, so his final condition was a re-expos ure to the edible reinforcement, which again resulted in an increase in responding. Figure 3-1 shows results of the concurrent-sch edule reinforcer assessments conducted with Bettina, Annette, Patsy, and Taneisha. All 4 subjects alternated between the two response (and reinforcer) options and, in fact, responded at nearly identical rates for olfactory and leisure stimuli. Patsys responding was most variable across sessions but sti ll remained in within a rather narrow range (0.3 to 0.8 responses per min), and the difference in responding towards one option over another within a session never exceed 0.2 responses per min for any subject. Discussion Results from Peters, Z euss, and Adelines si ngle-schedule reinforcer assessment showed that, despite the low ranking of olfactory stimuli in their pref erence hierarchy (Experiment 1), olfactory stimuli had some reinforcing value. These data were consistent w ith previous research showing that stimuli not sele cted during paired-stimulus pref erence assessments may still function as reinforcers when other options are unavailable (Roscoe et al., 1999). Drakes data paint a very different picture. Ol factory stimuli failed to function as reinforcers during his singleschedule reinforcer assessment. More surprising was the fact that leisure stimuli also failed to function as reinforcers during the same assessment. This suggests that Drakes reinforcer pool was very limited and consisted only of his most preferred stimulus class, edible stimuli. The

PAGE 36

36 potency of edible stimuli to func tion as reinforcers for Drake, or for other subjects with PWS, is not surprising. Results of Bettina and Taneishas single-schedul e reinforcer assessment and results of the concurrent-schedule reinforcer assessment Bettin a, Annette, Patsy, and Taneisha provided some measure of verification for preference patterns observed in Experiment 1 in that olfactory stimuli (a) functioned as reinforcers, and (b) competed with leisure stimuli. The subjects included in the concurrent-schedule reinforcer asse ssment showed two distinct profil es in Experiment 1: Bettina, Annette, and Patsy preferred ol factory stimuli over leisure stimuli, whereas Taneishas preference for olfactory and leisure stimuli was mixed. Both patterns suggested the possibility that olfactory reinforcers might compete with leisure reinforcers under a concurrent schedule and, in fact, their responding during Experiment 2 was virtually identical for both consequences. It was interesting to note that none of the subj ects who preferred olfactor y over leisure stimuli in Experiment 1 did so in Experiment 2. One possible explanation for similarity observe d across subjects in the concurrent-schedule reinforcer assessment in Experiment 2 is related to the nature of the ol factory sensory system. Repeated exposure to olfactory stimuli may resu lt in reduced sensitivity of the olfactory receptors and neurons by approximately 50% within 2.5 s to10 s of exposure (Wang, Walker, Sardi, Fraser, & Jacob, 2002). Thus, following the first response or two that is reinforced with olfactory stimuli, subjects may allocate res ponding towards leisure item s due to the reduced efficacy of the olfactory receptors following e xposure to olfactory stimuli. Our olfactory reinforcement access period was 30 s, much longer than the exposure periods noted by Wang et al.

PAGE 37

37 Additional support for this expl anation can be found in resear ch on the detection of an olfactory stimulus. Short and intense exposure to odorants seems to be more important than duration of exposure for optimization of olfact ory detection (Zhao, Dalton, Yang, & Scherer, 2006). Perhaps this is because of the decrease in sensitivity of olfactory neurons following olfactory stimulus exposure noted above. Thus, a refractory period betwee n olfactory-stimulus exposures may facilitate more e ffective responses to olfactory s timuli. If so, the equal response allocation towards both the olfactory and leisur e options may be a by-product of the periodic reduction in the ability of the olfactory receptors to respond to olfactory stimuli. Moving from the olfactory option to the le isure option may have allowe d a refractory period between olfactory-stimulus exposures. An alternative explanation may be that th e pre-session exposure to the tasks and the associated stimuli may have served as a prom pt to alternate between responses during the session. If so, it is possible that pre-session exposure exerted cont rol during session and responding was not influenced by contingent access to the putative reinfo rcers. This potential concern is somewhat mediated by the data from Bettina and Taneishas single-schedule reinforcer assessment because neither subject en gaged in the target response during baseline. However, future studies could in clude a control task in order to simultaneously evaluate whether or not olfactory stimuli function as reinforcers and examine their relative efficacy. Nevertheless, a reinforcement effect was observed for olfactory stimuli with 7 of the 8 subjects in Experiment 2 suggesting the potential viability of olfactory reinforcement fo r use in skill-acquisition or treatment programs.

PAGE 38

38 Table 3-1. Reinforcer assessment type and tasks Subject Assessment Schedule Task Bettina Single, Concurrent Raising arm above shoulder Taneisha Single, Concurrent Calculator button pressing Peter Single Touching symbol on wall Drake Single Putting a card in an envelope Zeus Single Phone dialing Adeline Single Touching head, shoulders, knees and toes Patsy Concurrent Placing card in bin Annette Concurrent P unching a hole in a card Subjects are in the left column. Assessment types (single-sc hedule or concurrent-schedule) are listed in the middle column. Tasks are listed in the right column.

PAGE 39

39 Figure 3-1. Results of the single-schedule reinforcer assessment for Bettina, Taneisha, Peter, Drake, Zeus, and Adeline.

PAGE 40

40 Figure 3-2. Results of the concur rent-schedule reinforcer assessm ent for Bettina, Annette, Patsy, and Taneisha.

PAGE 41

41 CHAPTER 4 GENERAL DISCUSSION Two experim ents examined preference for ed ible, leisure, and olfactory stimuli in individuals with PWS. Experiment 1 consisted of a series of 3 preference assessments (edible vs. leisure, leisure vs. olfactory, a nd olfactory vs. leisure) administered to individuals with and without PWS. The PWS and non-PWS subjects exhibited almost exclusive preference for edible stimuli over leisure and olfactory stimuli. PWS and non-PWS subjects both showed mixed preferences favoring neither stimul us class in the preference asse ssment that did not include food (olfactory vs. leisure stimuli). Experiment 2 validated the preference assessment outcomes from Experiment 1 with a subset of subjects. Four subjects from Experiment 1 who preferre d olfactory stimuli over leisure stimuli (or for whom preference for olfactory stimuli and leisure stimuli was mixed) were exposed to concurrent-schedule reinforcer assess ments in Experiment 2. All 4 subjects allocated responses toward both olfactory and leisure re inforcers. This outcome demonstrated that olfactory stimuli competed with leisure stimuli. Four additional subjects for whom olfactory stimuli were least preferred were exposed to a single-schedule reinforcer assessment in which only olfactory reinforcers were available. Thre ee of the 4 subjects showed increases in responding relative to baseline in the olfactory reinforcemen t condition, whereas the fourth subject did not respond for olfactory reinforcement. Taken together, results of these experiment s suggest that olfactory stimuli have the potential for use as reinforcers in skill acquisi tion and treatment programs for individuals with PWS and for other individuals with developmental disabilities who are on restricted diets. The preference-assessment data suggest that olfactor y stimuli may be equally preferred to leisure stimuli (some PWS subjects) or perhaps even pr eferred to food (some non-PWS subjects). The

PAGE 42

42 reinforcer-test data suggest that preference for olfactory stimuli is likely to be reflected in performance on tasks for which olfactory stimuli are delivered as consequences. Also of note, access to olfactory stimuli was a reinforcer for in dividuals who previously showed no preference for olfactory stimuli. This finding was consistent with previous research showing that pairedstimulus preference assessments are best suited for predicting relative but not absolute reinforcer value (Roscoe et al., 1999). One finding from Experiment 1 that emphasizes the importance of individualized assessment rather than generaliza tions based on aggregated data wa s the fact that single leisure stimuli outranked edible stimuli in individuals with PWS. This has important implications for the development of treatment programs directed at reducing food-related problem behavior. If a preference assessment suggests that a specific leisure stimulus may be more preferred than edible stimuli, that stimulus could be the reinforcer in a differential reinforcement of other behavior (DRO) procedure for refraining from inappropr iate food consumption or food stealing. Several findings of the current studies were somewhat incons istent with what has been reported in previous research on preference assessment. First was the fact that, unlike data reported by DeLeon et al. (1997) and Bojak a nd Carr (1999), the non-PWS subjects in our study did not show an exclusive preference for edible stimuli, although preference for edible stimuli was strong. It would be difficult to determine the basis for the difference. However, the results of the current study suggest that there may be some exceptions to the general preference for edible stimuli that has be en previously reported. Another somewhat unusual finding was that, although Bettina, Annette, and Patsy showed a clear preference for leisure over olfactory stim uli in Experiment 1, their rates of responding were almost identical for leisure versus olfact ory reinforcement in Experiment 2. Previous

PAGE 43

43 research (Fisher et al., 1992, Rosc oe et al., 1999) has shown that stimuli higher ranked in pairedstimulus preference assessments are usually pref erred under concurrent reinforcer schedules. The absence of preference for olfactory over le isure stimuli under the concurrent schedule may have been due to several factors. First, response allocation has not always favored higher ranked items from preference assessments. One subject in both the Fisher et al. (1992) and Roscoe et al. (1999) studies responded equally for highand low-ranked reinforcers; thus, it is possible that some subjects simply respond so as to maximize reinforcer vari ability. Previous research has shown that some subjects performance improves when exposed to varied reinforcer presentation relative to constant reinforcer pres entation (Egel, 1981). Second, it may be possible that a refractory peri od is necessary between olfactory stimulus exposure periods for the maximum effect of the olfactory stimulus to be experienced (Wang et al, 2002). If that is the case, alte rnation between the task that resulted in olfactory reinforcement and the task that resulted in leisure reinforcement would be better understood. This would suggest a potential limitati on to the use of olfactory stimuli as reinforcers. More specifically, olfactory stimuli may be more sens itive to a habituation effect than other classes of stimuli. Third, specific top-ranked leisure stimuli may have been better able to compete with olfactory stimuli than other members of the le isure class. Although Bettina, Annette, and Patsy generally preferred olfactory over leisure stimuli, preference fo r specific leisure stimuli and specific olfactory stimuli was more mixed. For example, Bettina preferred olfactory stimuli over leisure stimuli as a general class. However, he r top 3 stimuli in the ol factory versus leisure assessment included 2 leisure stimuli. Thus, when the top 3 olfactory stimuli were pitted against the top 3 leisure stimuli in a concurrent-schedul e reinforcer assessment, 1 of the top 3 leisure

PAGE 44

44 items outranked all olfactory stimuli, and a second leisure item outranked 2 of the top 3 olfactory stimuli. Thus, mixed responding in the concur rent-schedule reinforcer assessment is not surprising. Future research might gradually incr ease the schedule requirements for the leisure stimuli in the concurrent schedule reinforcer a ssessment to determine at what point preference might shift towards the olf actory reinforcement option. Several limitations of this study should be noted. All of the olfactor y stimuli included in the preference assessments were food flavors. In order to minimize differences due to preference for specific flavors, an attempt was made to make the olfactory stimuli somewhat similar to the edible stimuli. It was unclear, however, whether the olfactory esse nces would be as preferred as the samples of the actual foods (and the real odors associated with foods). Also, exclusive use of food odors may have restricted the range of olf actory stimuli included in Experiment 1. Edible and olfactory stimuli have a well-established pairi ng history due to the physiology of the oral and nasal orifices. It is possible that preference for olfactory stimuli over leis ure stimuli may be due to the exclusive inclusion of food-based olfactory stimuli in this experiment. Expanding the range of olfactory stimuli to include non food-based olfactory stimuli might disentangle the edible/olfactory pairing histor y and provide a more convincing demonstration of preference for olfactory stimuli unrelated to preference for edib le stimuli. Anecdotal evidence suggests that non-food olfactory stimuli may also be prefer red. Walking through the perfume aisle in a department store clearly demonstrates that humans are willing to engage in responses in order to access floral, woodsy, musky, and other non-food frag rances. Future research might examine a broader range of ol factory stimuli. Another limitation was the nature of respons es in Experiment 2. Although we selected responses designed to approximate vocational tasks, they were simple and repetitive. Therefore,

PAGE 45

45 it is unknown if olfactory reinforcement is suffici ent to establish or maintain more effortful responses or response chains. Finally, Experiment 2 focused on effects for se veral PWS subjects only. Therefore, it does not permit any general conclusions about olfactory reinforcement. There was no reinforcer test for the non-PWS subjects because their inclusion in Experiment 1 was as a comparison sample. Given their preference (some observed preference for olfactory stimuli), it would have been interesting to see the extent to which olfactory stimuli served as reinforcers for non-PWS subjects. Future research could include reinfo rcer assessments to validate the results of preference assessments with individuals with disabilities other than PWS. In summary, the present studies provide inform ation about a potential class of reinforcers that typically has not been examined in operant research. Olfactory stimuli do not pose the same types of problems for individuals on restricted di ets as edible stimuli pose. Thus, they may have special benefits for individuals with PWS. Result s showed that olfactory stimuli have properties that compete with those of leisure stimuli (and sometimes edible stimuli) in the context of preference assessments and that ol factory stimuli can function as reinforcers in some individuals. Both of these findings support an expansion of th e reinforcer pool to include olfactory stimuli and further studies examining their varied prop erties as well as thei r influence on operant behavior.

PAGE 46

46 APPENDIX A STIMULI USED IN PREFERENCE ASSESSMENTS DURING EXPERIMENT 1 AND PREFE RENCE RANKINGS FOR EACH STIMULUS Table A-1. Group 1: Stimuli used in preference assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Mikey, Misty, Peter, Jeffrey, Penelope, Mitchell, Bettina, Annette, Zeus, Taneisha, Drake 1(Orange Slice), 2 (Honey Nut), 3 (Coconut), 4 (Strawberry Crme Saver), 5 (Mint Tic Tac), 6 (Apple Cane), 7 (Cinnamon Altoid) 1 (Beads), 2 (Workbook), 3 (Puzzle), 4 (Bear), 5 (Magazines), 6 (Bumbleball),7 (Slinky) 1(Apple), 2(Orange), 3 (Strawberry), 4 (Mint), 5(Coconut),6 (Cinnamon), 7 (Honey Nut) Mikey (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 1 3 3 10.5 5.5 5.5 7 10.5 8.5 14 8.5 3 13 12 6.64 8.36 Edible vs. Olfactory 1 8.5 2 13 3.5 11 3.5 11 6.5 11 6.5 14 5 8.5 4 11 Leisure vs. Olfactory 1 4.5 2 11.5 6.5 11.5 6.5 11.5 11.5 11.5 3 11.5 4.5 8 5 10 Misty (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 2 14 6.5 6.5 6.5 9 1 11 6.5 11 3.5 11 3.5 13 4.21 10.79 Edible vs. Olfactory 1 14 4 13 3 11.5 2 11.5 7 9.5 5 9.5 6 8 4 11 Leisure vs. Olfactory 13 9 2 9 1 6.5 11 5 14 4 9 6.5 12 3 8.86 6.14 Peter (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 4.5 12.5 2.5 11 6.5 12.5 2.5 9.5 6.5 14 1 8 4.5 9.5 4 11 Edible vs. Olfactory 1 11 6.5 11 6.5 13.5 3.5 8.5 5 13.5 3.5 11 2 8.5 4 11 Leisure vs. Olfactory 6.5 10.5 4 10.5 14 10.5 2 6.5 6.5 10.5 3 13 1 6.5 5.29 9.71

PAGE 47

47 Table A-1. Continued Jeffrey (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 1 14 3 11 8.5 7 3 12.5 6 12.5 3 8.5 5 10 4.21 10.79 Edible vs. Olfactory 1.5 7.5 1.5 9 7.5 10.5 4 13 5 13 6 10.5 3 13 4.07 10.93 Leisure vs. Olfactory 5.5 9.5 3 7.5 7.5 9.5 1 12 3 14 3 12 5.5 12 4.07 10.93 Penelope (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 2 11 4 8 8 4 1 13 8 11 4 14 6 11 4.71 10.29 Edible vs. Olfactory 2 8.5 4 11.5 5 11.5 1 14 10 13 3 8.5 6 7 4.43 10.57 Leisure vs. Olfactory 10.5 5 2.5 5 1 7 10.5 14 8 9 12.5 2.5 12.5 5 8.21 6.79 Mitchell (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 14 11 1 11 4 13 8 6.5 2.5 9 11 5 2.5 6.5 6.14 8.86 Edible vs. Olfactory 14 9.5 2 8 5 12 11 6.5 3 9.5 13 6.5 1 4 7.0 8.0 Leisure vs. Olfactory 13 5.5 12 5.5 14 8 5.5 2 9.5 5.5 11 2 9.5 2 10.64 4.36 Bettina (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 6.5 14 2 6.5 12.5 9.5 3.5 1 5 9.5 8 12.5 3.5 11 5.86 9.14 Edible vs. Olfactory 5 9.5 1 14 13 9.5 2 11.5 5 11.5 7 8 3 5 5.14 9.86 Leisure vs. Olfactory 14 4 9.5 2 4 7.5 1 7.5 11 9.5 13 4 12 6 9.21 5.79 Annette (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 6.5 12 3 12 5 8 1 9 6.5 12 3 12 3 10 4.0 10.71 Edible vs. Olfactory 1 12.5 2.5 10.5 6.5 8 4 14 6.5 9 2.5 12.5 5 10.5 4.0 11 Leisure vs. Olfactory 10 4.5 14 4.5 10 2.5 12.5 7 10 2.5 8 6 12.5 1 11.0 4

PAGE 48

48 Table A-1. Continued Zeus (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 1 8 4 10.5 7 8 2 14 6 10.5 3 8 5 13 4.0 10.29 Edible vs. Olfactory 1 13.5 4 8.5 6 10 2 11 6 12 3 13.5 6 8.5 4.0 11.0 Leisure vs. Olfactory 2 10 7.5 13 1 5.5 5.5 12 10 10 3.5 14 3.5 7.5 4.71 10.29 Taneisha (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 5 13 2 6 9 1 3.5 13 7.5 13 3.5 10 7.5 11 5.43 9.57 Edible vs. Olfactory 2.5 9.5 4 12.5 5 11 2.5 9.5 6 12.5 1 14 7.5 7.5 4.07 10.93 Leisure vs. Olfactory 3.5 5 1.5 7.5 1.5 3.5 13.5 12 13.5 9.5 7.5 9.5 11 6 7.43 7.57 Drake (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 1 14 3.5 11 3.5 6.5 2 8.5 8.5 11 5 13 6.5 11 4.29 10.71 Edible vs. Olfactory 2 11 3 8.5 4 13 1 11 7 11 5.5 14 5.5 8.5 4.0 11.0 Leisure vs. Olfactory 12 8 6 10.5 4.5 10.5 1 14 2.5 8 13 8 2.5 4.5 5.93 9.07 Table A-2. Group 2: Stimuli used in preference assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Patsy, Wanda, Dorian 1(Teddy Graham), 2 (Banana Nut Muffin), 3 (Honey Dew), 4 (Pineapple Jelly Bean), 5 (Strawberry Cream), 6 (Mint Tic Tac), 7 (Apple Cane), 8 (Cinnamon Altoid) 1 (Beads), 2 (Workbook), 3 (Puzzle), 4 (Bear), 5 (Magazines), 6 (Bumble Ball), 7 (Slinky), 8 (Kaleidoscope) 1(Apple), 2 (Cookie), 3 (Strawberry), 4 (Mint), 5 (Pineapple), 6 (Cinnamon), 7 (Honey Nut), 8 (Banana Nut) Patsy (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 8 Edible vs. Leisure 6 16 1 10.5 3 15 6 12.5 2 10.5 8 14 6 12.5 4 9 4.5 12.5 Edible vs. Olfactory 5 15.5 2.5 15.5 4 11 8 11 1 13.5 6 9 7 13.5 2.5 11 4.5 12.5 Leisure vs. Olfactory 16 5 5 12.5 15 8.5 8.5 5 14 11 8.5 2 12.5 1 8.5 3 11.0 6.0

PAGE 49

49 Table A-2. Continued Wanda (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 8 Edible vs. Leisure 3.5 16 1.5 12 9 13 3.5 8 1.5 11 6.5 10 5 13.5 6.5 13.5 4.63 12.13 Edible vs. Olfactory 3 12.5 1.5 15.5 8 10.5 4 12.5 1.5 10.5 5.5 9 5.5 14 7 15.5 4.5 12.5 Leisure vs. Olfactory 15 5.5 12 5.5 10.5 1 13.5 3 7.5 3 10.5 3 13.5 7.5 16 9 12.31 4.69 Dorian (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 8 Edible vs. Leisure 13 5.5 2 3.5 16 7 3.5 8.5 1 11 15 11 11 5.5 8.5 14 8.75 8.25 Edible vs. Olfactory 8.5 8.5 2.5 5 11.5 5 2.5 13.5 1 13.5 16 11.5 8.5 15 5 8.5 6.94 10.06 Leisure vs. Olfactory 5 14 1.5 9.5 7 7 7 15 1.5 12.5 3.5 9.5 3.5 12.5 16 11 5.63 11.38 Table A-3. Group 3: Stimuli used in preference assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Juan 1(Teddy Graham), 2 (Banana Nut Muffin), 3 (Honey Dew), 4 (Pineapple Jelly Bean), 5 (Strawberry Cream), 6 (Mint Tic Tac), 7 (Apple Cane), 8 (Cinnamon Altoid) 1 (Beads), 2 (Workbook), 3 (Puzzle), 4 (Bear), 5 (Magazines), 6 (Bumble Ball), 7 (Slinky), 8 (Kaleidoscope) 1(Apple), 2 (Cookie), 3 (Strawberry), 4 (Mint), 5 (Pineapple), 6 (Cinnamon), 7 (Honey Nut), 8 (Banana Nut) Juan (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 8 Edible vs. Leisure 10.5 14 8.5 15 6 16 6 10.5 3.5 12 2 8.5 1 6 3.5 13 5.13 11.88 Edible vs. Olfactory 15 8.5 16 1 11 6.5 12 3 14 2 4.5 6.5 13 8.5 4.5 10 11.25 5.75 Leisure vs. Olfactory 16 3.5 15 8.5 13.5 8.5 12 7 11 5 2 1 10 3.5 13.5 6 11.63 5.38

PAGE 50

50 Table A-4. Group 4: Stimuli used in preference assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Melissa, Jacob, Barnaby, Trina, Adeline, Jewel, Jamal 1 (Mint Tic Tac), 2 (Orange Tictac), 3 (Almond), 4 (Hot Tamale), 5 (Coconut Jelly Bean), 6 (Vanilla Wafer), 7 (Strawberry Twizzler) 1 (Beads), 2 (Workbook), 3 (Puzzle), 4 (Bear), 5 (Magazines), 6 (Bumbleball), 7 (Cars) 1(Vanilla), 2 (Orange), 3 (Strawberry), 4 (Mint), 5 (Coconut), 6 (Cinnamon), 7 (Almond) Melissa (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 11 8 5.5 12.5 1.5 7 3.5 9.5 1.5 9.5 5.5 12.5 3.5 14 4.57 10.43 Edible vs. Olfactory 7.5 11 5.5 9.5 3 13.5 2 13.5 1 9.5 5.5 11 4 13.5 4.07 11.64 Leisure vs. Olfactory 1 10.5 2 12 3.5 9.5 6 13 6 10.5 6 3.5 14 9.5 5.5 9.79 Jacob (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 6 14 7.5 7.5 1.5 11 4.5 12.5 3 10 1.5 12.5 4.5 9 4.07 10.93 Edible vs. Olfactory 7.5 9.5 4.5 13 2 14 3 11.5 4.5 11.5 1 7.5 6 9.5 4.07 10.93 Leisure vs. Olfactory 14 3 10 8.5 12 8.5 6.5 1 6.5 4 11 2 13 5 10.43 4.57 Barnaby (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 7 12 6 13 5 9 3 14 2 11 1 9 4 9 4.0 11.0 Edible vs. Olfactory 7 8.5 6 10 4 8.5 4 12 2 14 1 12 4 12 4.0 11.0 Leisure vs. Olfactory 9 3.5 5.5 9 1.5 9 11 7 3.5 14 1.5 12 5.5 13 5.36 9.64 Trina (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 7.5 4.5 11.5 11.5 4.5 7.5 1 7.5 2.5 11.5 7.5 14 2.5 11.5 5.29 9.71 Edible vs. Olfactory 10 10 7.5 14 2.5 7.5 2.5 13 1 10 4 6 5 12 4.64 10.36 Leisure vs. Olfactory 5.5 1 11 2.5 7 2.5 11 11 5.5 4 11 11 11 11 8.86 6.14

PAGE 51

51 Table A-4. Continued Adeline (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 8 7 6 10.5 3 1.5 1.5 12 4.5 10.5 9 13 4.5 14 5.21 9.79 Edible vs. Olfactory 5 11 5 11 5 9 1 14 2.5 13 7 8 2.5 11 4.0 11.0 Leisure vs. Olfactory 2 11.5 2 8 2 6 9.5 13.5 5 4 11.5 7 13.5 9.5 6.5 8.5 Jewel (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 2.5 11 4.5 14 7 10 1 8.5 2.5 13 4.5 12 6 8.5 4.0 11.0 Edible vs. Olfactory 2 8 4 13.5 7 9 1 11 3 11 6 11 5 13.5 4.0 11.0 Leisure vs. Olfactory 12.5 3.5 10 5.5 7 1 8 3.5 14 2 10 5.5 12.5 10 10.57 4.43 Jamal (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 12 4 8 13.5 3 1 6 10 2 10 6 13.5 6 10 6.14 8.86 Edible vs. Olfactory 4 11.5 4 11.5 1 11.5 7.5 14 2 7.5 4 7.5 7.5 11.5 4.29 10.71 Leisure vs. Olfactory 2 10 3 10 1 6 12 6 13.5 4 14 10 13.5 6 8.43 7.43 Table A-5. Group 5: Stimuli used in preference assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Manuel 1 (Mint Tic Tac), 2 (Orange Tic Tac), 3 (Almond), 4 (Hot Tamale), 5 (Coconut Jelly Bean), 6 (Vanilla Wafer), 7 (Strawberry Twizzler) 1 (Castle), 2 (Leap Pad), 3 (Football Game), 4 (Coloring Book), 5 (Books), 6 (Blue Car), 7 (Legos) 1(Vanilla), 2 (Orange), 3 (Strawberry), 4 (Mint), 5 (Coconut), 6 (Cinnamon), 7 (Almond) Manuel (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 4 7 10 13 8.5 14 1.5 11 3 12 5.5 5.5 1.5 8.5 4.86 10.14 Edible vs. Olfactory 3 8.5 6 11 5 11 3 8.5 3 11 1 13 7 14 4.0 11.0 Leisure vs. Olfactory 3 4.5 6 4.5 8.75 8.75 8.75 8.75 8.75 8.75 1.5 8.75 1.5 8.75 5.46 7.53

PAGE 52

52 Table A-6. Group 6: Stimuli used in preference assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Katarina 1 (Mint Tic Tac), 2 (Orange Tictac), 3 (Almond), 4 (Hot Tamale), 5 (Coconut Jelly Bean), 6 (Vanilla Wafer), 7 (Strawberry Twizzler) 1 (Caterpillar), 2 (Coloring Book), 3 (Magazine), 4 (Memory Match), 5 (Dog Puppet), 6 (Lincoln Logs), 7 (Mr. Potato Head) 1(Vanilla), 2 (Orange), 3 (Strawberry), 4 (Mint), 5 (Coconut), 6 (Cinnamon), 7 (Almond) Katarina (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 6.5 8 6.5 9 3.5 10.5 1.5 10.5 1.5 12 3.5 13 5 14 4.0 11.0 Edible vs. Olfactory 7 11 6 12 1 13.5 3 13.5 2 9 4 9 5 9 4.0 11.0 Leisure vs. Olfactory 11 4.5 3 7 11 11 11 6 11 1.5 11 1.5 11 4.5 9.86 5.14 Table A-7. Group 7: Stimuli used in preference assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Cindy 1 (Mint Tic Tac), 2 (Orange Tic Tac), 3 (Almond), 4 (Hot Tamale), 5 (Coconut Jelly Bean), 6 (Vanilla Wafer), 7 (Strawberry Twizzler) 1 (Leap Pad), 2 (Coloring Book), 3 (Magazine), 4 (Race Cars), 5 (Dog Puppet), 6 (Lincoln Logs), 7 (Mr. Potato Head) 1(Vanilla), 2 (Orange), 3 (Strawberry), 4 (Mint), 5 (Coconut), 6 (Cinnamon), 7 (Almond) Cindy (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 3 11.5 6.5 11.5 1.5 11.5 6.5 9 4.5 8 1.5 11.5 4.5 14 4.0 11.0 Edible vs. Olfactory 5 9 6.5 11 1 9 2.5 14 4 12.5 2.5 12.5 6.5 9 4.0 11.0 Leisure vs. Olfactory 14 5 10 7 13 1.5 5 8 5 10 10 3 12 1.5 9.86 5.14

PAGE 53

53 Table A-8. Group 8: Stimuli used in preference assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Montana 1 (Mint Tic Tac), 2 (Orange Tic Tac), 3 (Almond), 4 (Hot Tamale), 5 (Coconut Jelly Bean), 6 (Vanilla Wafer), 7 (Strawberry Twizzler) 1 (Leap Pad), 2 (Coloring Book), 3 (Reading Book), 4 (Race Cars), 5 (Legos), 6 (Football Game), 7 (Castle) 1(Vanilla), 2 (Orange), 3 (Strawberry), 4 (Mint), 5 (Coconut), 6 (Cinnamon), 7 (Almond) Montana (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 5.5 14 5.5 12 7 13 2.5 8.5 1 10.5 4 10.5 2.5 8.5 4.0 11.0 Edible vs. Olfactory 3 11 4.5 8.5 6.5 13 1.5 14 1.5 8.5 6.5 11 4.5 11 4.0 11.0 Leisure vs. Olfactory 14 10 13 8 10 3.5 1.5 6 6 6 1.5 3.5 12 10 8.29 6.71 Table A-9. Group 9: Stimuli used in preference assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Ernie 1 (Mint Tic Tac), 2 (Orange Tic Tac), 3 (Almond), 4 (Hot Tamale), 5 (Coconut Jelly Bean), 6 (Vanilla Wafer), 7 (Strawberry Twizzler) 1 (Caterpillar), 2 (Play Bus), 3 (Books), 4 (Race Cars), 5 (Legos), 6 (Dog Puppet), 7 (Leap Pad) 1(Vanilla), 2 (Orange), 3 (Strawberry), 4 (Mint), 5 (Coconut), 6 (Cinnamon), 7 (Almond) Ernie (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 6 10 13 11.5 8 4 5 4 2 8 3 11.5 1 8 5.43 8.14 Edible vs. Olfactory 5.5 12 9 12 3 12 5.5 12 1.5 12 4 7.5 1.5 7.5 4.29 10.71 Leisure vs. Olfactory 5 12 2.5 12 8 12 2.5 9 1 12 6 6 6 2 4.43 9.29

PAGE 54

54 Table A-10. Group 10: Stimuli used in prefer ence assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Janet, Geoff 1 (Mint Tic Tac), 2 (Orange Tic Tac), 3 (Almond), 4 (Hot Tamale), 5 (Coconut Jelly Bean), 6 (Vanilla Wafer), 7 (Strawberry Twizzler) 1 (Football Game), 2 (Coloring Book), 3 (Books), 4 (Race Cars), 5 (Legos), 6 (Castle), 7 (Leap Pad) 1(Vanilla), 2 (Orange), 3 (Strawberry), 4 (Mint), 5 (Coconut), 6 (Cinnamon), 7 (Almond) Janet (PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 1.5 11 7 5 7 13.5 1.5 13.5 3.5 9 7 11 3.5 11 4.43 10.57 Edible vs. Olfactory 2 13 3.5 10 7 10 1 8 3.5 13 5.5 10 5.5 13 4.0 11.0 Leisure vs. Olfactory 12 7.5 1 3.5 12 3.5 12 2 5 9 10 7.5 6 12 8.29 6.43 Geoff (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 6.5 1 14 3.5 6.5 12.5 12.5 10.5 8.5 8.5 3.5 2 10.5 5 8.86 6.14 Edible vs. Olfactory 2 10 3.5 10 7 10 5 14 3.5 12.5 7 7 1 12.5 4.14 10.86 Leisure vs. Olfactory 4 10.5 1 4 8 12.5 2 8 4 8 6 10.5 4 12.5 4.14 9.43 Table A-11. Group 11: Stimuli used in prefer ence assessments during experiment 1 and individual stimulus rankings Subjects Edible Stimuli Leisure Stimuli Olfactory Stimuli Leyla 1 (Mint Tic Tac), 2 (Orange Tic Tac), 3 (Almond), 4 (Hot Tamale), 5 (Coconut Jelly Bean), 6 (Vanilla Wafer), 7 (Strawberry Twizzler) 1 (Leap Pad), 2 (Books), 3 (Magazines), 4 (Cars), 5 (Lincoln Logs), 6 (Dog Puppet), 7 (Mr. Potato Head) 1(Vanilla), 2 (Orange), 3 (Strawberry), 4 (Mint), 5 (Coconut), 6 (Cinnamon), 7 (Almond) Leyla (Non-PWS) Assessment Stimuli Mean 1 2 3 4 5 6 7 Edible vs. Leisure 2.5 8 6.5 10.5 1 10.5 6.5 13.5 2.5 10.5 5 4 10.5 13.5 4.93 10.07 Edible vs. Olfactory 14 5 11 11 1 4 3 7 2 11 7 11 11 7 7.0 8.0 Leisure vs. Olfactory 1 6 10.5 4 9 4 13 7.5 12 7.5 10.5 4 14 2 10.0 5.0

PAGE 55

55 LIST OF REFERENCES Ai me, P. (2007). Fasting increases and satiation decreases olfactory detection for a neutral odor in rats. Behavioural Brain Research, 179(2), 258-264. Akefeldt, A., & Gillberg, C. (1999). Behavior an d personality characteristics of children and young adults with Prader-Willi syndrome: A controlled study. Journal of the American Academy of Child and Adolescent Psychiatry, 38 761-769. Bhutta, M. F. (2007). Sex and the nose: human pheromonal responses. Journal of the Royal Society of Medicine, 100 (6), 268-274. Bojak, S. L., & Carr, J. E. (1999). On the displ acement of leisure item s by food during multiplestimulus preference assessments. Journal of Applied Behavior Analysis, 32 515-518. Buck, L. (2000). Smell and taste: The chemical sens es. In E. r. Kandel, J. H. Schwartz & T. M. Jessell (Eds.), Principles of Neural Science (4th ed., pp. 625-645). New York: McGrawHill. Buck, L., & Axel, R. (1991). A novel multigene family may encode odorant receptors: A molecular basis for odor recognition. Cell, 65 175-187. Burd, L., Vesely, B., Martsolf, J., & Kerbeshi an, J. (1990). Prevalence study of Prader-Willi syndrome in North Dakota. American Journal of Medical Genetics, 37 97-99. Butler, M. G. (1996). Molecular diagnosis of Prader-Willi syndrome: Comparison of cytogenetic and molecular genetic data including pare nt of origin dependent methylation DNA patterns. American Journal of Medical Genetics, 61 188-190. Butler, M. G., Meaney, F. J., Opitz, J. M., & Reynolds, J. F. (1987). An anthropometric study of 38 individuals with Prader-Labhart-Willi syndrome. American Journal of Medical Genetics, 26 445-455. Butler, M. G., Theodoro, M. F., Bittel, D. C., & Donnelly, J. E. (2007). Energy expenditure and physical activity in Prader-Willi syndro me: Comparison with obese subjects. American Journal of Medical Genetics, 143A 449-459. Chu, S. (2008). Olfactory conditioning of positive performance in humans. Chemical Senses, 33 65-71. Clarke, D. J., Boer, H., & Webb, T. (1995). Ge netic and behavioural aspects of Prader-Willi syndrome: A review with a translation. Mental Handicap Research, 8 38-53. Davis, R. L. (2004). Olfactory learning. Neuron, 44(1), 31-48. 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.

PAGE 56

56 DeLeon, I. G., Iwata, B. A., & Roscoe, E. M. (1997). Displacement of le isure reinforcers by food during preference assessments. Journal of Applied Behavior Analysis, 30 475-484. Djordjevic, J. (2008). Olfaction in patients with mild cognitive impairment and Alzheimer's disease. Neurobiology of Aging, 29 (5), 693-706. Eibenstein, A. (2005). Olfactory screeni ng test in mild cognitive impairment. Neurological Sciences, 26 (3), 156-160. Egel, A.L. (1981). Reinforcer variation: Impli cation for motivating developmentally disabled children. Journal of Applied Behavior Analysis, 14, 345-350. Fisher, W., Piazza, C. C., Bowman, L. G., Hagopi an, L. P., Owens, J. C. & Slevin, I. (1992). A comparison of two approaches for identifying reinforcers for persons with severe and profound disabilities. Journal of Applied Behavior Analysis, 25 491-498. Goldberg, D. L., Garrett, C. L., Van Piper, C ., & Warzac, W. J. (2002). Coping with Prader-Willi syndrome. Journal of the American Di etetic Association, 102 537-542. Gottschalk, J. M., Libby, M. E., & Graff, R. B. (2000). The effects of establishing operations on preference assessment outcomes. Journal of Applied Behavior Analysis, 33 85-88. Gunay-Aygon, M., Cassidy, S. B., & Nicholls, R. D. (1997). Prader-Willi and other syndromes associated with obesity and mental retardation. Behavior Genetics, 27 307-324. Hanley, G. P., Iwata, B. A., Lindberg, J. S., & Conners, J. (2003). Response-restriction analysis: I. Assessment of activity preferences. Journal of Applied Behavior Analysis, 36 47-58. Hawkes, C. (2006). Olfaction in neurodegenerative disorder. Taste and Smell: an Update, 63 133-151. Ivancic, M. T. (2000). Stimulus pr eference and reinforcer assessment applications. In J. A. J. E. Carr (Ed.), Handbook of applied behavior analysis (pp. 19-38). Reno: Context Press. Jellinek, J. S. (2004). Proust remembered: Ha s Proust's account of odor-cued autobiographical memory recall really been investigated? Chemical Senses, 29 (5), 455-458. Kettenmann, B. (2005). Odor and taste inte raction on brain responses in humans. Chemical Senses, 30 I234-i235. Lamb, A. S., & Johnson, W. M. (1987). Premature co ronary artery atherosclerosis in a patient with Prader-Willi syndrome. American Journal of Medical Genetics, 28 873-880. Maglieri, K. A., DeLeon, I. G., Rodriguez-Catter, V., & Sevin, B. M. (2000). Treatment of covert food stealing in an individual with Prader-Willi syndrome. Journal of Applied Behavior Analysis, 33 615-618.

PAGE 57

57 McAdam, D. B., Klatt, K. P., Koffarnus, M., Di cesare, A., Solberg, K., Welch, C., et al. (2005). The effects of establishing operation s on preferences for tangible items. Journal of Applied Behavior Analysis, 38 107-110. McBurney, D. H. (2006). Olfactory comfort: Sme lling a partner's clothing during periods of separation. Journal of Applied Social Psychology, 36 (9), 2325-2335. Mohr, C. (2001). Associations to sm ell are more pleasant than to sound. Journal of Clinical and Experimental Neuropsychology, 23(4), 484-489. Mombaerts, P. (2004). Love at first smell Th e 2004 Nobel Prize in Physiology or Medicine. New England Journal of Medicine, 351 (25), 2579-2580. Morgan, C. D. (2002). Olfactory event-rela ted potentials in Alzheimer's disease. Journal of the International Neuropsyc hological Society, 8 (6), 753-763. Murphy, C. (1999). Loss of olfactor y function in dementing disease. Physiology & Behavior, 66(2), 177-182. Northrup, J., George, T., Jones, K., Broussard, C., & Vollmer, T. R. (1996). A comparison of reinforcer assessment methods: The utility of a verbal and a pictoria l choice procedure. Journal of Applied Behavior Analysis, 29 201-212. Olander, E., Stamberg, J., Steinberg, L., & Wulfsberg, E. A. (2000). Third Prader-Willi syndrome phenotype due to maternal unipare ntal disomy 15 with mosaic trisomy 15. American Journal of Medical Genetics, 93 215-218. Pace, G. M., Ivancic, M. T., Edwards, G. L., Iw ata, B. A., & Page, T. J. (1985). Assessment of stimulus preference and reinforcer valu e with profoundly retarded individuals. Journal of Applied Behavior Analysis, 18 249-255. Page, T. J., Finney, J. W., Parrish, J. M., & Iw ata, B. A. (1983). Assessment and reduction of food stealing in Prader-Willi children. Applied Research in Mental Retardation, 4 219228. Page, T. J., Stanley, A. E., Richman, G. S., D eal, R. M., & Iwata, B. A. (1983). Reduction of food theft and long-term maintenance of weight loss in a Prader-Willi adult. Journal of Behavior Therapy and Experimental Psychiatry (14), 261-268. Reid, D. H., Everson, J. M., & Green, C. W. ( 1999). A systematic eval uation of preferences identified through person-centered pl anning for people with profound multiple disabilities. Journal of Applied Behavior Analysis, 32 467-477. Richardson, B. E. (2004). Altered olfactory acuity in the morbidly obese. Obesity Surgery, 14 (7), 967-969.

PAGE 58

58 Rincover, A., & Newsom, C. D. (1985). The re lative motivational prope rties of sensory and edible reinforcers 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, 31 605-620. Roscoe, E. M., Iwata, B. A., & Kahng, S. (1999). Rela tive versus absolute reinforcement effects: Implications for preference assessments. Journal of Applied Behavior Analysis, 32 479493. Seiden, A. M. (2001). The diagnosis of a conductive olfactory loss. Laryngoscope, 111 (1), 9-14. Stevenson, D. A., Heinemann, J., Angulo, M., Butler, M. G., Loker, J., Rupe, N., et al. (2007). Deaths due to choking in Prader-Willi syndrome. American Journal of Medical Genetics, 143, 484-487. Verhagen, J. V., & Engelen, L. (2006). Neurocognitive bases of human multimodal food perception: Sensory integration. Neuroscience and Biobehavioral Reviews, 30 613-650. Wang, L., Walker, V. E., Sardi, H., Fraser, C., & Jacob, T. J. C. (2002). The correlation between physiological and psychological responses to odour stimul ation in human subjects. Clinical Neurophysiology, 113 542-551. Wharton, R. H., Wang, T., Graeme-Cook, F., Briggs, S., & Cole, R. E. (1997). Acute idiopathic gastric dilation with gastric necrosis in individuals with Prader-Willi syndrome. American Journal of Medical Genetics, 73 437-441. Wilder, D. A., Schadler, J., Higbee, T. S., Haym es, L. K., Bajagic, V., & Register, M. (2008). Identification of olfactory stim uli as reinforcers in individuals with autism: A preliminary investigation. Behavioral Interventions, 23 97-103. Windsor, J., Piche, L. M., & Locke, P. A. (1994). Preference testing: A comparison of two presentation methods. Research in Developmental Disabilities, 15 439-455. Wise, R. A. (2006). Role of brain dopami ne in food reward and reinforcement. Philosophical Transactions of the Royal Society B-Biological Sciences, 361 (1471), 1149-1158. Yeomans, M. R. (2006). Olfactory influen ces on appetite and satiety in humans. Physiology & Behavior, 89 (1), 10-14. Zhao, K., Dalton, P., Yang, G. C., & Scherer, P. W. (2006). Numerical mode ling of turbulent and laminar airflow and odorant transport during sniffing in the human and rat nose. Chemical Senses, 31 107-118.

PAGE 59

59 BIOGRAPHICAL SKETCH Sarah Bloom earned an A.A. Degree from Simons Rock College of Bard in 1996 and a B.A. degree at the University of Washington in 1998. She was initially exposed to behavior analysis as a long-term substitute teacher at th e Sussex Consortium, part of the Delaware Autism Program. Sarah began graduate training in the area of Behavior Anal ysis in the Psychology Department at the University of Florida in 2003 under the supervision of Dr. Brian Iwata. She earned her Masters Degree and became a Board Ce rtified Behavior Analyst while pursing her doctoral degree at UF. She has conducted research pr ojects on olfactory stimuli as reinforcers in individuals with Prader-Willi Syndrome, assessme nt and treatment of rumination and vomiting, a trial-based approach to functional analysis of problem behavior, a work-place health and wellness program, and acquisition of verbal be havior in children who engage in escapemaintained problem behavior. She has presented her research at multiple state, national, and international conferences. She has served as the si te coordinator of Dr. Iwatas laboratory at a local school for children with ex ceptional needs and has provided be havioral services at a local branch of the ARC, a residentia l and day program for individuals with developmental disabilities including Prader-Willi Syndrome. She has served as a teaching assistant and instructor of record for undergraduate courses in applied behavior an alysis. This fall Sarah will be an assistant professor in the Special Education and Rehabil itation Department at Ut ah State University.