<%BANNER%>

Response Restriction and Contingent Access to Responding in Multi-Response Environments


PAGE 1

RESPONSE RESTRICTION AND CONTINGENT ACCESS TO RESPONSES IN MULTI-RESPONSE ENVIRONMENTS By JASON C. BOURRET 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 2005

PAGE 2

Copyright 2005 by Jason C. Bourret

PAGE 3

iii ACKNOWLEDGMENTS I would like to express my sincere gratitude to the members of my committee, Drs. Jesse Dallery, Brian Iwata, Mark Lewis, Terry Scott, and Timothy Vollmer, for their support, guidance, and mentorship. I would al so like to thank my wife, Eileen Roscoe, my mother, Patricia Machkovich, and my gr andparents, George and Lucille Machkovich, for their love, support, and encouragement in all my endeavors. I would also like the thank the collaborators of this study, Timothy Vollmer a nd Luanne Witherup, for all their contributions to the study.

PAGE 4

iv TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF FIGURES...........................................................................................................vi ABSTRACT......................................................................................................................vii CHAPTER 1 INTRODUCTION........................................................................................................1 Overview.......................................................................................................................1 Basic Research on Contingent Access to Responses as Reinforcement in MultiOperant Environments.............................................................................................7 Application of Response Restriction a nd the Response-Deprivation Hypothesis......10 Purpose.......................................................................................................................17 2 STUDY 1: AN EVALUATION OF A BRIEF RESPONSE-RESTRICTION PROCEDURE IN A NATURALLY OCCURING MULTI-RESPONSE ENVIRONMNET.......................................................................................................20 Method........................................................................................................................20 Participants and Setting.......................................................................................20 Response Recording and Reliability...................................................................21 Procedure.............................................................................................................21 Results and Discussion...............................................................................................22 3 STUDY 2: AN EVALUATION OF REPO NSE RESTRICTION AS A METHOD FOR INCREASING SOCIALLY DESIREABLE RESPONSEES IN A NATURALLY OCCURING MULTI-R ESPONSE ENVIRONMENT.....................34 Method........................................................................................................................34 Results and Discussion...............................................................................................35 4 STUDY 3: AN EVALUATION OF CONTIN GENT ACCESS TO RESPONSES AS A METHOD FOR INCREASING SOCIAL LY DESIREABLE RESPONSES IN A NATURALLY OCCURING MULTI-R ESPONSE ENVIRONMENT.....................43 Method........................................................................................................................43

PAGE 5

v Results and Discussion...............................................................................................45 5 GENERAL DISCUSSION.........................................................................................53 LIST OF REFERENCES...................................................................................................60 BIOGRAPHICAL SKETCH.............................................................................................63

PAGE 6

vi LIST OF FIGURES Figure page 1 Seconds of responding during each mi nute of each session during brief response restriction for Kyle...................................................................................................24 2 Seconds of responding during each mi nute of each session during brief response restriction for John...................................................................................................26 3 Seconds of responding during each mi nute of each session during brief response restriction for Sarah..................................................................................................28 4 Seconds of responding during each mi nute of each session during brief response restriction for Beth...................................................................................................30 5 Seconds of responding during the re sponse-restriction tr eatment for Kyle.............36 6 Seconds of responding during the re sponse-restriction tr eatment for John.............37 7 Seconds of responding during the re sponse-restriction tr eatment for Sarah............39 8 Seconds of responding during the re sponse-restriction tr eatment for Beth.............41 9 Seconds of responding during th e contingency treatment for Kyle.........................46 10 Seconds of responding during th e contingency treatment for John.........................47 11 Seconds of responding during th e contingency treatment for Sarah........................49 12 Seconds of responding during th e contingency treatment for Beth.........................51

PAGE 7

vii 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 RESPONSE RESTRICTION AND CONTINGENT ACCESS TO RESPONDING IN MULTI-RESPONSE ENVIRONMENTS By Jason C. Bourret December, 2005 Chair: Timothy Vollmer Major Department: Psychology In this investigation, the e ffects of response restricti on and contingent access to responses in multi-response environments were evaluated in three studies. The same four participants took part in all studies. Kyle was a typically developing 13-year-old male. John was an 8-year-old male diagnosed with autism. Sarah was a typically developing 14-year-old female. Beth was a typically developing 7-year-old female. For all participants, caregivers select ed reading and mathematics as responses in need of increase. In Study 1, participants were expos ed to a brief response -restriction analysis, consisting of a short series of response rest rictions. For all participants, reading was observed following the restrict ion of certain responses and, for all subjects other than Sarah, both reading and mathematics were obs erved during response-restriction phases. In Study 2, the efficacy of response restricti on as a treatment method for increasing the likelihood of socially desirable targeted res ponses in naturally occurring multi-response environments was examined. The utility of the brief restriction analyses conducted in

PAGE 8

viii Study 1 in identifying responses that compete with targeted behavior was evaluated by restricting access to th ese responses for greater periods of time and examining response reallocation. When restrict ion of these responses was in sufficient to produce reliable increases in targeted responding, the restricti on of additional responses was evaluated. Response restriction produced increases in targeted responding for 3 of 4 participants; however, these increases were only reliable and sustained for Kyle. In study 3, the effects of contingent access to response s on targeted behavior in multi-response environments were examined. For all participants, contingent access to multiple responses produced reliable increas es in targeted behavior. Im plications of the results for research and application are discussed.

PAGE 9

1 CHAPTER 1 INTRODUCTION Overview Classifying stimuli functionally is a hallmar k of behavior analysis. As stimulus operations are performed, their effects on beha vior are the defining features used in differentiating one type of stimulus from anot her (Skinner, 1935). Take, for example, the classification of a stimulus as a reinforcer. Any stimulus that increases the probability of a response being emitted when its presentation or removal follows the response is classified as a reinforcer (Skinner, 19 53). Thus, grain presented following the keypecking of a pigeon that results in an increased lik elihood of keypecking, a token presented following completion of a math problem that results in an increased likelihood of completing math problems, and a demand th at is removed following hitting a teacher which results in an increased likelihood of hitting the teacher would all be reinforcers. These would be classified as positive reinforc ers in the case of those that are presented and negative reinforcers in the case of those that are removed. The classification of stimuli, and reinfor cers in particular, based on their function, or the effects they have on behavior, has prove n fruitful in keeping research in behavior analysis conceptually systematic and engende ring effective technologies for application to socially significant behavior (Catania, 1998; Hanley, Iwata, & McCord, 2003). This method of identifying reinforcers, however, has not been without criticism (Postman, 1947; Timberlake & Allison, 1974). Postman po inted out the circularity of defining a reinforcer based solely on its effect on re sponding. This definition occurs after the

PAGE 10

2 operation has taken place and allows for no independent means of classifying the stimulus as a reinforcer. From a pragmatic standpoint this is tr oublesome. Without any independent characteristics of a stimulus iden tifying it as a reinforcer, we are left with identifying reinforcers through tr ial and error. For basic re searchers, the fact that a stimulus is not identified as a reinforcer until it has been demonstrated to have a reinforcing effect may be an annoyance in th at multiple stimuli may have to be tested until a reinforcer is identified. In the appl ication of behavior analysis to socially significant behavior, it is sometimes critical to quickly and accurately identify reinforcing stimuli and a failure to identify a reinforcer can result in the failu re of an important behavioral treatment. Meehl (1950) proposed a partial solution to difficulties arising from post-hoc identification of reinforcers. Meehl argued that reinfo rcers should be considered transituational. A reinforcer identified in one context could then be used as a reinforcer in another context. Thus, researchers or clin icians in need of a reinforcer may simply make use of a stimulus that had previously been demonstrated to be a reinforcer. The argument of transitua tionality depends upon the assumption that stimulus functions have a certain amount of generality (Sidman, 1960) This assumption is commonly, and often effectively, made by behavior analysts. The logic behind the use of preference assessments to identify reinforcers for use in other contexts, for example, relies upon the generality of the stimuli identified as re inforcers (Timberlake & Farmer-Dougan, 1991). This generality, however, may be limited ac ross varying levels of response effort, schedule parameters, motivating operations, or the availability of substitutable or

PAGE 11

3 competing reinforcers (Timberlake & A llison, 1974; Timberlake & Farmer-Dougan, 1991). The probability-differential hypothesis, introduced by Premack (1959), suggested an independent and a priori method of reinfor cer identification. Premack conceptualized reinforcement as involving contingent access to responses rather than stimuli. Rather than grain as the reinforcer for the keypecki ng of a pigeon, according to this account, the reinforcer is access to eating. The probabil ity-differential hypothesis, sometimes called the Premack principle, states that conti ngent access to a more probable response will reinforce a less probable response. The initia l probabilities are dete rmined in a pairedoperant baseline. In this baseline, free acces s is permitted to both responses. Data are collected on duration of each of the two responses in baseline in order to determine which response is more probable. A re inforcement effect is then ob served if access to the more probable response is restricted and provided contingent upon the less probable response. While there are conditions under which the Premack principle does not hold, it has proven to have broad generality and can be a heuristic that may help clinicians identify reinforcers (Catania, 1998; Timberlake & Farmer-Dougan, 1991). Timberlake and Allison (1974) extended th e Premack principle with the responsedeprivation hypothesis. Like Premack (1959) Timberlake and Allison conceptualized reinforcement as involving contingent access to responses. Unlike Premack, they argued that the critical feature of a reinforcement contingency was not that contingent responses be more probable than instrumental res ponses (those that produce access to the contingent responses). They stipulated th at contingencies like those used in common operant experiments disrupt the ratio of responding that would be observed if no

PAGE 12

4 contingency were in place (e.g., a continge ncy involving access to water produced by the wheel running of a rat typically involves rest ricting access to drinking such that the ratio of running to drinking is great er than that which would be observed under a free-operant baseline) and that organisms are motivated to respond in a manner that makes relative response allocation more similar to that observed during free-operant baselines. Timberlake and Allison cite this disequilibr ium produced by operant contingencies as the critical factor that determines a rein forcement contingency. They describe the necessary condition for a reinforcement effect as c iO O C I where I and C represent measures of instrume ntal and consequence responses respectively and Oi and Oc represent the operant level, or baseline level, of instrumental and consequence responses. A reinforcement effect is predicted if a contingency is arranged such that the ratio of instrumental to consequence re sponses is greater than that observed during baseline. It is important to not e that as long as the units of analysis are constant across baseline and contingency condi tions, they do not need to be the same for instrumental and consequence responses. For example, if, in a pair ed-operant baseline, a child read five pages of a book and watche d television for one minute, the responsedeprivation hypothesis predicts a reinfo rcement effect given the following TV read pagesC I min1 5 Thus, a contingency arranged such that re ading 10 pages is required to produce one minute of television, or

PAGE 13

5 TV read pages TV read pages min min1 5 1 10 would be expected to produce a reinforcem ent effect, whereas a contingency arranged such that reading three pages is require d to produce one minute of television, or TV read pages TV read pages min min1 5 1 3 would not be expected to produce an increa se in pages read over what was observed during the free-operant baseline. The manner in which the response-de privation hypothesis expands upon the Premack principle becomes more apparent if both responses are measured using the same units. If, instead of pages read, minutes of reading were meas ured, the contingency required for reinforcement may be something like the following TV readingC I min min1 4 An arrangement of TV reading TV reading min min min min1 4 1 5 should produce a reinforcement effect for r eading. Thus, contingent access to a less probable response should reinforce a more pr obable response. Th e relative baseline probabilities of the responses ar e thus shown to be unimportant The critical variable is the arranged ratio of instrumental to cons equence response as compared to the ratio obtained during baseline (Timberlake & Allison, 1974). The response-deprivation hypothesis has been demonstrated to be a useful heuristic in identifying conditions under which access to responses will serve as

PAGE 14

6 reinforcers in multiple studies in both basi c and applied behavior analytic literature (Allison & Timberlake, 1974; Amari, Grace, & Fisher, 1995; Charlop, Kurtz, & Casey, 1990; Hanley, Iwata, Thompson, & Lindbe rg, 2000; Homme, de Baca, Devine, Steinhorst, & Rickert, 1963; Konarski, J ohnson, Crowell, & Whitman, 1980; Mitchell & Stoffelmayr, 1973; Osborne, 1969; Timber lake & Allison, 1974). A measure of responding on a paired-operant ba seline can allow a pr iori identification of a reinforcer. An independent method of reinforcer iden tification helps allevi ate the philosophical concern regarding the circularit y of a purely functional definiti on of a reinforcer and also provides clinicians an additional met hod of reinforcer identification. Some of the most promising areas for application of the response-deprivation hypothesis are complex, multi-operant, environm ents (e.g., homes) in which there are often a large number of available respons es (e.g., watching tele vision, playing video games, listening to a radio, talking on the phone, reading a book, or doing homework). Reinforcers identified through standard preference assessment methods, which are typically conducted in restricted -operant environments in wh ich the number and type of responses available are well controlled, may be less effective in environments containing numerous alternative responses. The re sponse-deprivation hypothesis suggests measuring baseline levels of responding in th e environment in which the contingency is to be put into place. Thus, potentially comp eting sources of reinforcement are taken into account. Surprisingly, however, very little resear ch has explicitly evaluated the responsedeprivation hypothesis in multi-operant enviro nments (Bernstein, 1998). Most of the research on the response-deprivation hypothe sis has been done using paired-operant

PAGE 15

7 preparations, and the degree to which the same results would occur in multi-operant environments remains largely unknown. For ex ample, the following would be expected to result in a reinforcement effect for read ing given four minutes of reading and one minutes of watching television du ring a paired-operant baseline TV reading TV reading min min min min1 4 1 5 However, if, for example, baseline data collected in a multi-operant environment indicated that reading occurr ed for four minutes, television watching for one minute, playing video games for eight minutes, shooti ng pool for seven minut es, and talking on the telephone for 10 minutes, it is unclear if the above contingency would produce the same result. Perhaps the probability of r eading would increase, or perhaps responding would be reallocated to playing video ga mes, shooting pool, talking on the phone, or some new response. A paired-operant base line does not permit the detection of such reallocation. In the following section, basic research on response deprivation in multioperant settings is discussed. Basic Research on Contingent Access to Responses as Reinforcement in MultiOperant Environments Research on the effects of response re striction and continge nt access to responses in multi-operant environments with humans was pioneered in Bernsteins human-operant laboratory (Bernstein, 1998; Bernstein & Ebbesen, 1978; Be rnstein & Michael, 1990). Bernstein and colleagues created a laborator y apartment in which human subjects lived, 24 hrs a day, for the duration of an experiment Subjects made a list of desired activities upon initiation of a study and thos e activities were typically av ailable in the laboratory apartment. Access to external stimuli and activities inside the apartment was controlled

PAGE 16

8 by experimenters and data were collected on time spent engaging in various activities across experimental conditions. The first major study to come from this la boratory was an explicit investigation of the effects of response restriction and c ontingent access to responses when multiple response options were available (Berns tein & Ebbesen, 1978). Three subjects participated, ranging in age from 19 to 39 years. The subjects lived in the experimental apartment from between 21 to 34 days. Each selected several activities before the beginning of the study and required materials fo r these activities were made available in the apartment. Data on the duration of e ngagement in these various activities were collected by observers who monitored the s ubjects from behind one-way mirrors. The experimenters attempted to examine the e ffects of response restriction alone and contingent access to responses in a multi-operant environment. Responding was measured under three conditions : a free-operant baseline, in which there was no restriction on response availa bility; a contingency condition, in which a high-probability response, de signated the contingent res ponse, was restricted and provided contingent upon a low-probability response chosen as the instrumental response; and a matched-restriction conditi on, in which access to the response that was the contingent response in th e contingency condition was rest ricted such that the access time was yoked to the time of access in the contingency condition. The contingencies and restrictions were described to the subj ects in each condition. The times at which responses were restricted were signaled to the subjects by the illumination of labeled lights on a panel in the apartment.

PAGE 17

9 For all three subjects, response contingenc ies resulted in increases in instrumental responding. For one subject, restriction of the consequence response alone produced reliable increases in instrumental responding. For the other two subj ects, the effects of restriction alone were less reliable. On e subject displayed no effect of response restriction alone; the other demonstrated an increase in one topography, but this effect was not replicated. The Bernstein and Ebbesen (1978) study represented a move toward examining response restriction and contingent acce ss to responses in a naturally occurring environment. Although the experiment took place in an operant laboratory and the number of available responses was still lim ited, subjects had access to more than two responses, and these were responses that are common in the natural environment. Their findings suggest that response contingencies hold promise for application in complex, multi-operant, human environments. They al so suggest that, in some circumstances, response restriction alone may be sufficient to increase targeted responses in these environments. The effects of response restriction on response allocation in a multi-response environment were explicitly examined by Lyons and Cheney (1984). Subjects were four adult, male rats and sessions took place in a chamber divided into five areas. Each area contained one of the following sources of re inforcement: turning off the light, viewing a female rat, turning off a white noise genera tor, water, or food. Free-operant baselines resulted in similar patterns of responding for all subjects, with th e most preferred area being that which terminated the white noise. For each subject, two restriction conditions were tested. In one, the most likely free-operant response was re stricted. In the other, the

PAGE 18

10 least likely free-operan t response was restricted. The experimenters used the outcomes restriction conditions to eval uate four rules for how re sponding might be re-allocated following response restriction. The first was th e constant ratio rule which states that the remaining responses will increase to o ccur during the same proportion of available time that they did with the high-probability response present. Th e second was the most probable remaining response" rule, which stip ulated that the sec ond most likely response during baseline will increase relatively more than other remaining responses. The third was the equal redistribution rule, which stated that responding will be equally reallocated to all available responses. The final rule tested was the sequential dependency rule, which hypothesizes that the re striction of a response will also decrease the response most likely to occur following the restricted response. The results replicated those of Bernstein and Ebbesen (1978). Re -allocation of respondi ng following restriction was idiosyncratic to the participant and vari ed depending on whether it was the highest or lowest probability response that was restrict ed. These data suggest ed that none of the four rules discussed was sufficient to predic t the results of respons e restriction for any given subject. The reliability of the results, however, is unc lear. For each subject, each restriction phase was conducted only once. Wh ile no a priori rule was found to describe the results, it may be that restriction effects are reliable within subject and that the effects of previous restrictions can be good predictors of the effects of subsequent restrictions. Application of Response Restriction a nd the Response-Deprivation Hypothesis Osborne (1969) evaluated the effects of contingent and noncontingent free time on the in-seat behavior of stude nts in a school for the deaf. The six participants were female students between 11 and 14 years old w ith severe to profound hearing loss. The number of times each student got out of he r seat during the daily recording period was

PAGE 19

11 targeted for intervention. The experiment wa s designed as a reversal, with phase changes occurring simultaneously for all participants. In each condition, con tingencies in effect were described to the participants. Duri ng baseline, there was no free time provided either contingently or noncontingently. In the first treatment condi tion, each participant was allowed 5 minutes of free time contingent upon 15 minutes without leaving her seat. This was immediately successful, with the numbe r of times each participant left her seat reducing to near zero. Following this phase a yoked-control condition was implemented in which 5 minutes of noncontingent access to free time was provided every 15 minutes. The level of responding in this condition was si milar to that obtained in the contingentaccess condition with very slight upward trends in response frequency for three participants. The noncontingent-access phase con tinued for five days for all participants. This was followed by 51 days of the conti ngent-access phase during which the frequency of participants leaving their seats remained near zero. Data collection was terminated at this point and a one-day follow-up probe was conducted approximately six-weeks later. In the intervening time, the free-time contingency remained in place and no participants were observed to leave their seats during the probe. In the week following the probe, a one day contingency reversal was implemente d. Five minutes of free time during this reversal was contingent upon participants leav ing their seats at le ast once during each 15minutes period and the contingency was describe d to the participants. The frequency of seat leaving during the reversal was elevat ed during the contingency reversal for all participants. Osborne (1969) was one of the first to use contingent access to responses for behavior management in a natural environment. This study demonstrated the

PAGE 20

12 effectiveness of contingent access to respons es in a multi-response environment and also illuminated some important issues with rega rd to manipulating access to responses in these environments. One potentially critical variable in this study is the presence of a possible negative reinforcement contingenc y. In the study conducted by Osborne, access to restricted responses was also paired with escape from a work c ontext. The degree to which the obtained treatment effects were a f unction of contingent access to responses or contingent escape from work is unclear. It may have been the case that a break from the work context into a relatively impoverished environment would have been equally successful. In fact, the eff ectiveness of noncontingent free time periods in suppressing seat leaving suggests interpreta tion of the effects of treatme nt as an abolishing operation (Laraway, Snycerski, Michael, & Poling, 2003) in that frequent breaks may have made the demand context less aversive. An equally plausible explanation is that the free-time periods resulted in frequent acc ess to responses otherwise on ly available contingent upon seat leaving and thus decreased the reinforc ing efficacy of these responses. The relative effects of these potential abolishing operati ons and the omission contingency are also unclear. Another important point of considerati on is that the data collection, and the contingencies used, converted an envir onment in which numerous responses were available, presumably of different operant cl asses, into one in wh ich only in-seat and outof-seat behavior were available. This aggr egation may have concealed some potentially interesting findings. The authors noted that ou t of seat activities in cluded, but were not limited to, talking with classmates, conversi ng with the teacher, sharpening pencils, drinking water, and using the restroom. If seat leaving was maintained by access to

PAGE 21

13 multiple responses, for example peer interact ion and access to the restroom, contingent access to only one of the two may not have b een sufficient to suppress responding. This is a particularly important case. If data collected in a free-operant condition indicated that interacting with peers was the most pr obable response, both the Premack principle and the response-deprivation hypo thesis would predict that contingent access to this response would serve as reinforcement. Ho wever, access to a restroom may be potent enough to maintain seat leaving even if access to peer interaction were contingent upon the absence of seat l eaving. A finding of this sort woul d contradict the Premack principle and the response-deprivation hypothesis; however, the aggregation of responses into large classes, while experimentally convenient, disallowed the detection of this sort of effect. Mitchell and Stoffelmayr (1973) demonstrat ed the effectiveness of contingent access to a high-probability response as a reinfo rcer for the behavior of two individuals diagnosed with schizophrenia. Both particip ants resided in a psyc hiatric ward and were employed stripping copper wire from coils. Thirty-second partial in terval data were collected on coil stripping over the course of 30-minute sessions and, during baseline, both participants engaged in no coil stri pping for six consecutive sessions. The experimenters tested the effects of instructi ons alone, which consisted of one prompt to work delivered at the beginning of each se ssion, and instructions plus reinforcement, which consisted of identical instructions and 90 seconds of sitting contingent upon the removal of wire from three coils. Sitting wa s used as a consequence response as it was identified as a high-probability response during pre-experiment data collection. Instructions alone were found to have no in creasing effect on res ponding. Instructions

PAGE 22

14 plus reinforcement, however, re sulted in high and stable leve ls of coil stripping. Like Osborne (1969), Mitchell and Stoffelmayr demo nstrated the use of contingent access to responding to increase the lik elihood of a socially signifi cant response. They also demonstrated that the generality of the rein forcement effects predicted by the responsedeprivation hypothesis extends to the popul ation of individuals diagnosed with schizophrenia. The effects of contingent sitting, however, were confounded with potential effects of restriction of sitting. It may have been the case that restriction of sitting alone would have resulted in increases in coil stripping. It is also difficult to assess the degree to which the results of this study speak to the effectiveness of contingent access to responses in multi-operant environments. All sessions took place in the room of the psychiatric ward where all of the work of the patients was done. The number of alternatively available activities was not clearly specified. The authors noted that the subjects had previously refused to engage with multiple potentially reinforcing items. Also, the fact that sitting rather th an a more active response was chosen as the consequence response suggests th at the environment was relati vely devoid of reinforcing activities. A final concern involves the con ceptualization of acce ss to sitting as an instance of reinforcement in the form of access to a response. In the contingency condition, if the participants sa t, their chair was tipped forward until they were forced to stand. Standing for extended periods may have been effortful for the participants and it may have been more appropriate to conceptu alize the reinforcement effect as negative reinforcement in the form of escape from standing rather than positive reinforcement in the form of access to sitting.

PAGE 23

15 The applied studies reviewed thus far have viewed restriction effects as potential confounds to be controlled for. Green and St riefel (1988) evaluated these effects directly in a multi-operant preparation composed of six available responses. Four children between the ages of six and 10, all diagnosed wi th autism, participated. The six activities were chosen by the childrens classroom teacher and were either play or educational activities. The analysis test ed four hypotheses regarding the redistribution of responding following restriction of the highest-probability response. The constant ratio rule, the equal distribution rule, and the most probable alternative rule, as tested by Lyons and Cheney (1984) were also tested by Green and Stiefel. The final hypothesis tested, the selective substitution rule, was that the pa ttern of redistribution may differ for each subject. The selective substitution rule is, e ssentially, that the outcome of restriction of the most probable response cannot be predic ted based on baseline re sponse probability. For each participant, Green and Striefel (1988) conducted an initial free-operant baseline and then sequentially and cumulativ ely restricted access to the most probable response such that five responses were availa ble in the second condi tion, four in the third, three in the fourth, and two in the third. Following the final restri ction condition, another free-operant condition was conducted. Th e experimenters found that response redistribution followed no single pattern, t hus supporting a selective substitution description of the effects of re sponse restriction. These data replicated the idiosyncratic effects of response restricti on alone discussed previously (Bernstein & Ebbesen, 1978; Lyons & Cheney, 1984). Some potential limita tions exist concerning the generality of these findings extending into applied setti ngs in which responding is not limited to experimenter-selected responses. The first po tential limitation relate s to questions about

PAGE 24

16 the reliability of the effects of response restriction for any give n subject. Although no general pattern of redistribution of res ponding was observed across participants, the degree to which patterns of redistribution for individual subjects would remain stable across multiple replications was not examined. It may be th at reliable rules describing response reallocation can be generated for indi vidual subjects given certain restrictions. A second concern involves th e fact that respond ing could only be allocated to the responses arranged by the experimenters. Th e experimental prepar ation did not address the potential emergence of previously unobserve d responses following restriction. In an environment with more available responses the emergence of previously unobserved responses is always a possibil ity. A final potential limitation is that data were collected using 30-second partial interval data collec tion. Following completion of this study, the experimenters directly compared the partial interval data collecti on used with duration recording in scoring video tape s of sessions similar to the ba seline sessions of the study. They found that the rankings of highest to lowest probability that would be assigned to responding based on the outcomes of these tw o types of measurement agreed 83.3% of the time. This suggests that the rankings obt ained in the study may have been in error nearly 17% of the time. In a study designed to explicitly examine components of a behavioral treatment involving contingent access to stereotypy as reinforcement, Hanley, Iwata, Thompson, and Lindberg (2000) evaluated the effects of re sponse restriction in isolation as well as contingent access to behavior on engagement with leisure items. The experimenters examined the effects of prompting alone, prompting plus blocking of stereotypy, and prompting plus response blocking and continge nt access to stereotypy. Participants were

PAGE 25

17 three adults diagnosed with developmental di sabilities who engaged in various forms of stereotypy. The provision of leisure items, with no further intervention, had no suppressive effect on ster eotypy and item engagement was not observed for two participants and occurred at a low level fo r the third. Prompting engagement with the items resulted in no decreases in stereot ypy for any participant and an increase in engagement for only one participant. Pr ompting plus response blocking produced decreases in stereotypy for al l participants and increases in item interaction for two participants. For the third participant, item interaction was increased by allowing access to stereotypy contingent upon item engagement. The effects of res ponse restriction alone in Hanley et al. have some important, and relatively novel, implicat ions for clinicians attempting to produce increases in the appropriate behavior of their clients. Hanley and colleagues were able to increase a targ eted, socially releva nt, response for two participants without having to place a contingency on the response. While this study represents an important initial demonstrati on of the use of res ponse restriction as a method for increasing a socially desirable response, it is im portant to note that sessions were conducted in relatively barr en therapy rooms. Previous research has demonstrated that the effects of response restriction on reallocation of responding are idiosyncratic across participants when multip le alternative responses ar e available (Bernstein and Ebbesen, 1978; Green and Striefel, 1988). The degree to which clinic ians can rely on response restriction as a method for increas ing targeted responses in less controlled environments remains unknown. Purpose The present investigation was composed of three studies. All took place in naturally occurring environments in which the number of available responses was

PAGE 26

18 uncontrolled (e.g., a childs home environment or a play area). These relatively unrestricted multi-response environments are those in which the transituationality of reinforcers identified in preference assessm ents is challenged, insofar as multiple potentially competing sources of reinforcement ar e available. Thus, an activity identified as highly preferred in an isolated pref erence assessment may not compete with the multiple sources of alternative reinforcemen t available in a childs home (e.g., television shows, movies, radio, or video games). A re sponse-deprivation approach to reinforcer identification has promise for application in these settings. Following a responsedeprivation model of reinfor cer identification, the environm ent in which the response to be used as a reinforcer is identified is the same environment in which the contingency will be implemented. Thus, alternatively av ailable sources of reinforcement are taken into account. As noted above, however, a num ber of questions remain to be answered before clinicians can reliably use res ponse restriction and response-deprivation contingencies in complex, naturally occurr ing environments. The present studies represent an attempt to answer the followi ng questions. One, can response restriction alone be used to reliably incr ease a targeted, socially desi rable, response in a naturally occurring, multi-response environment? Two, ca n contingent access to behavior be used to increase socially important behavior in these same naturally o ccurring, unrestricted, environments or will access to multiple comp eting sources of reinforcement impede its effectiveness? Study 1 was an evaluation of a brief respons e restriction analysis for identifying responses that may be competing with targ eted responses in the setting in which treatment will be implemented. Study 2 was an examination of response restriction alone

PAGE 27

19 as a treatment method for incr easing socially important beha vior. The purpose of Studies 1 and 2 was to determine if a brief restric tion analysis would identify responses that would be effective as part of a response-restrictio n treatment and to determine if response restriction alone would produce reliable and su stained increases in de sired behavior in a multi-response environment. Study 3 was an i nvestigation of the use of contingent access to responses as a treatment designed to incr ease targeted behavior. The purpose of Study 3 was to determine the effectiveness of c ontingent access to responses as a method for increasing desired behavior in a multi-response environment. The present studies extended previous rese arch in two ways. One, by explicitly evaluating use of response re striction as a means of increasing a targeted, socially desirable, response in a naturally occurring, multi-res ponse environment. Two, by explicitly evaluating the use of contingent access to beha vior to increase socially important behavior in these same naturall y occurring, unrestricted environments.

PAGE 28

20 CHAPTER 2 STUDY 1: AN EVALUATION OF A BRIEF RESPONSE-RESTRICTION PROCEDURE IN A NATURALLY OCCURING MULTI-RESPONSE ENVIRONMENT Previous research has demonstrated that the restriction of av ailable responses can result in increases in other responses even in the absence of a c ontingency (Bernstein & Ebbesen, 1978; Green & Striefel, 1988; Hanl ey et al., 2000; Lyons & Cheney, 1984). However, the effects of response restriction in multi-response environments have varied across subjects and no single rule has been pred ictive of the outcome of these restrictions (Green & Striefel, 1988; Lyons & Cheney, 1984) In Study 1, participants were exposed to a brief response-restricti on analysis, consisting of a short series of response restrictions. The outcome of these brief re sponse-restriction analyses were used to identify responses that, when restricted, may result in increases in targeted socially desirable responses as well as to evaluate the reliability of th e effects of response restriction on re sponse reallocation. Method Participants and Setting Participants were 4 indivi duals who had been referred by parents and caregivers due to their lack of independent engagement in academic and other socially desirable behavior. Kyle was a typically developing 13-year-old male. John, was an 8-year-old male diagnosed with autism. Sarah was a t ypically developing 14-year-old female. The final participant, Beth was a typically developi ng 7-year-old female. Each session was 5 minutes in duration and sessions were conduc ted 1 to 3 times per week. Sessions for

PAGE 29

21 Kyle, Sarah, and Beth were all conducted in their respective homes. Sessions for these participants began in the r oom of their choice, however, participants movement throughout their home was not restricted. Fo r John, sessions were conducted in a play area at his school, in which a computer, a CD player, a television with a VCR and a number of movies, as well as a wide variet y of games, toys, and other activities were available. For all participants, the mate rials necessary for the targeted academic responses were made available in th e room in which the session began. Response Recording and Reliability Observers collected data using palm -top computers on duration of time spent engaging in various activities. Engagement was defined as orienting toward and, when appropriate, being in contact with material s necessary for given activities. Data collectors scored the occurrence of any identi fiable response, if no identifiable response was observed the data collectors scored nothing. A second obser ver collected data independently on 37.8% of all sessions across all studies. Interobserver agreement was calculated by dividing each session into 10-second intervals, dividing the smaller duration of recorded responses by the larger across observers, averaging these scores across the session, and converting them to a percentage. Overall agreement scores for all participants across all measures averaged 97.3% (range 80.3% to 100%). Procedure The primary investigator met with the car egivers of each participant to identify responses to increase. For Kyle, Sara h, and Beth, age appropriate reading and mathematics activities were identified by caregivers as responses that were considered important for the children and which rarely happened at home. For John, working on

PAGE 30

22 mathematics worksheets and reading were iden tified as important responses that did not occur unless he was explicitly prompted in an academic setting. Each participant completed two to four br ief restriction analyses. These analyses consisted of a series of five 5-minute sessions The first 5 minutes were a free-operant period. Participants were inst ructed that they may do anythi ng that they like. Following the free-operant session, the activity that wa s engaged in for the greatest duration was restricted and participants were told that th ey may do anything that they like other than the restricted activity. In each subsequent se ssion, the activity engaged in for the longest duration in the previous session wa s also restricted and subjects were instructed that they may do anything that they like other than th e restricted activities. Responses were restricted by instruct ing participants that they may not engage in these response and no participants attempted to engage in the responses that they were told were restricted. Two brief-restriction analyses were conducted with Kyle, th ree with Sarah and Beth, and four with John. At the time the brief restric tion analyses were conducted, the duration of participation of Kyle, Sarah, and Beth in this series of studies wa s unclear. For this reason fewer brief restriction analyses were conducted with these participants to allow them to move on to the treatment studies in the series (Studies 2 and 3) more quickly. Only one brief restriction analysis wa s conducted per day for all participants. Results and Discussion Figure 1 shows the outcomes of the brief rest riction analyses for Kyle. In the top panel, during the free-operan t phase, Kyle spent his time playing video games. Following restriction of video games (Restric tion 1), Kyle shot pool. When access to video games and shooting pool were restricted (Restriction 2), Kyle watched television. When television was also restricted (Restr iction 3), Kyle spent his time reading.

PAGE 31

23 Following the restriction of reading as well (Restriction 4), he allocated his time to a mathematics workbook. The bottom panel show s the results of th e second restriction

PAGE 32

24 Figure 1. Seconds of responding during each minute of each session during brief response restriction for Kyle; sessions are denoted by phase change lines. Seconds of Responding Minutes Brief Restriction Assessment 1 0 10 20 30 40 50 60 0510152025 Free Operant Restriction 1Restriction 2Restriction 3Restriction 4 Playing Video Games Playing Pool Watching Television Reading Math Brief Restriction Assessment 2 0 10 20 30 40 50 60 0510152025 Free Operant Restriction 1Restriction 2Restriction 3Restriction 4 Hair Twirling

PAGE 33

25 analysis for Kyle. In the free-operant pha se, Kyle spent all of his time doing math. Following restriction of math (Restriction 1), Kyle allocated his time to reading. When both mathematics and reading were restricted (Restriction 2), Kyle played video games. Following the restriction of mathematics, r eading, and video games (Restriction 3) Kyle sat on the floor, engaging in no particular identifiable response Restriction 4 was identical to Restriction 3 as Kyle had emitted no identifiable response to be restricted. In Restriction 4, he remained seated on the fl oor and spent the session twirling his hair. Figure 2 depicts the results of the brief restriction analyses for John. Data from the first analysis are plotted in the top pane l. During the free-operant phase, John played computer games. Restriction of computer games (Restriction 1) re sulted in John playing on a therapy ball. When computer games and the therapy ball were restricted (Restriction 2), he played with a handhe ld electronic game. When the handheld game was also restricted in Restriction 3, John watched a video. When access to videos was also restricted in (Restriction 4), John initially allocated his responding to mathematics worksheets during the first 2 minutes of th e session, and primarily engaged in reading during the final 3 minutes. The second panel shows the results of John s second brief restriction analysis. In the free-operant phase, he played computer games. With access to the computer restricted in Restriction 1, John watched vi deos. When both access to the computer and videos were restricted (Restriction 2), John played on a therapy ball. When access to computer, videos, and the therapy ball were all restricted (Restriction 3), he played with a handheld electronic game. When the handheld game was added into the restriction in

PAGE 34

26 Figure 2. Seconds of responding during each minute of each session during brief response restriction for John; sessions are denoted by phase change lines. Brief Restriction Assessment 4 0 10 20 30 40 50 60 0510152025 Free OperantRestriction 1Restriction 2Restriction 3Restriction 4 Brief Restriction Assessment 2 0 10 20 30 40 50 60 0510152025 Free OperantRestriction 1Restriction 2Restriction 3Restriction 4 Brief Restriction Assessment 3 0 10 20 30 40 50 60 0510152025 Free OperantRestriction 1Restriction 2Restriction 3Restriction 4 Brief Restriction Assessment 1 0 10 20 30 40 50 60 0510152025 Free OperantRestriction 1Restriction 2Restriction 3Restriction 4 Playing on a Computer Playing on a Therapy Ball Playing an Electronic Game Watching Videos Reading Math Minutes Seconds of Responding

PAGE 35

27 Restriction 4, John spent time working on ma thematics during the first 3 minutes and reading during the last 2. The third panel depicts the results from J ohns third brief-restriction analyses. John played on the computer during the fr ee-operant phase. When access to the computer was restricted in Restriction 1, he watched a video. When access to the computer and movies were restricted in Restriction 2, John played with a handheld electronic game. When the elect ronic game was also restricted in Restriction 3, he spent the majority of the first 3 minutes playing on a therapy ball and the last 2 reading. As playing on the therapy ball was th e most probable response in the third restriction phase, this response was added to the restriction du ring Restriction 4. Duri ng the first minute of the Restriction 4, John spent time reading a nd doing mathematics and during the final 4 minutes John only worked on mathematics. The bottom panel shows the results of the fi nal brief restriction analysis for John. John played on the computer during the freeoperant condition. When the computer was restricted in Restriction 1, J ohn watched a video. With restri cted access to the computer and movies in Restriction 2, he played with a handheld electronic game. When access to the computer, videos, and handheld game were all restricted in Restriction 3, John played on the therapy ball. When the therapy ball was also restricted in Restriction 4, he read during the first 2 minutes of the session but th is decreased during th e third minute and he sat quietly on the floor for the final two minutes. Figure 3 depicts the results for Sarah. The top panel shows results of the first brief restriction analysis. In the initial fr ee-operant phase, Sarah al located most of her time to watching television. With television re stricted in Restricti on 1, Sarah watched a

PAGE 36

28 Figure 3. Seconds of responding during each minute of each session during brief response restriction for Sarah; sessions are denoted by phase change lines. Seconds of Responding Minutes Brief Restriction Assessment 3 0 10 20 30 40 50 60 0510152025 Free OperantRestriction 1Restriction 2Restriction 3 Restriction 4 Playing with a Toy Phone Reading a Magazine Playing with Clay Coloring Brief Restriction Assessment 2 0 10 20 30 40 50 60 0510152025 Free Operant Restriction 1Restriction 2Restriction 3 Restriction 4 Blowing Bubbles Listening to a CD Player Brief Restriction Assessment 1 0 10 20 30 40 50 60 0510152025 Free Operant Restriction 1Restriction 2Restriction 3 Restriction 4 Watching Television Watching Videos Reading Drawing Weaving

PAGE 37

29 video. With both television and movies restri cted in Restriction 2, Sarah allocated 100% of her time to reading. With reading also rest ricted in Restriction 3, she spent all of her time drawing. With television, videos, reading, and sketching restricted in Restriction 4, Sarahs responding was allo cated to weaving. The second panel shows the results for the second brief-restriction analysis. Sarah allocated all of her time to weaving in the freeoperant phase. With weaving restricted in Restriction 1, Sarah watched television. When weaving and television were restricted in Restriction 2, she watched a video. When weaving, television, and movies were restricted in Restriction 3, Sarah spent her time blowing bubbles. With bubbles also restricted in Restriction 4, she listened to a CD player for the majority of the first 3 minutes and this response decreased during the final 2 minutes. The bottom panel depicts the results for th e third brief-restri ction analysis for Sarah. In the free-operant phase, she spent most of her time watching television. With television restricted in Rest riction 1, Sarah played with clay. With restriction of television and clay in Restrict ion 2, Sarah colored some pictures with markers. With restriction of television, cla y, and coloring in Restriction 3, Sarah sat on her bed without engaging in any particular response for the firs t half of the session and played with a toy phone for the second half. With playing w ith the toy phone added to the restricted responses in Restriction 3, Sarah sp ent her time reading a magazine. Figure 4 shows the results for the brief re striction analysis for Beth. In the top panel, the data for the first analysis are depict ed. In the free-operant phase, Beth watched television. In Restriction 1, television was re stricted and Beth worked on mathematics. In Restriction 2, with televisi on and mathematics restricted, she played with dolls. With

PAGE 38

30 Figure 4. Seconds of responding during each minute of each session during brief response restriction for Beth; sessions are denoted by phase change lines. Seconds of Responding Minutes Brief Restriction Assessment 3 0 10 20 30 40 50 60 0510152025 Free Operant Restriction 1Restriction 2Restriction 3Restriction 4 Playing with Chalk Working on a Report Brief Restriction Assessment 1 0 10 20 30 40 50 60 0510152025 Free Operant Restriction 1Restriction 2Restriction 3 Restriction 4 Watching Television Math Playing with Dolls Playing with Cards Writing in a Diary Playing with a Ball Brief Restriction Assessment 2 0 10 20 30 40 50 60 0510152025 Free Operant Restriction 1Restriction 2Restriction 3 Restriction 4 Counting Money Playing with Clay Reading

PAGE 39

31 television, mathematics, and dolls restricted in Restriction 3, Beth initially spent some time writing in her diary, then played with cards, and ended the session playing with a ball. As playing with cards was the most likely response in Restriction 3, this response was included in the restriction for Restriction 4. In Restriction 4, Beth spent the first minute playing with a ball and th e majority of the rest of se ssion writing in her diary. The middle panel of figure 4 displays the results of the sec ond brief-restriction analysis for Beth. When television was restri cted in Restriction 1, Beth spent the initial portion of the session counting money in her pur se. During the end of the condition, Beth spent her time reading and playing with a ball. As money counting was the most probable response during Rest riction 1, both television and money counting were restricted in the second. In Restriction 2, Beth spent her tim e playing with clay. When clay was added to the restri ction in Restriction 3, she worked on mathematics for the majority of the condition. With mathematics included in the restrict ion in Restriction 4, Beth spent most of her time reading. The bottom panel of figure 4 depicts the results of the third brief-restriction assessment. During the free-operant phase, Beth worked on mathematics. When mathematics was restricted in Restriction 1, Beth watched television. With television and mathematics restricted in Restriction 2, Beth worked on a report for her social-studies class. When access to working on the report wa s also restricted in Restriction 3, Beth played with a ball. With television, mathema tics, working on a social -studies report, and playing with a ball restricted in Restricti on 4, she allocated her time to drawing with chalk.

PAGE 40

32 The brief restriction analyses produced samples of free-operant responding and responding under various response restrictions. For all subjects, one of the targeted responses (reading) was observed following th e restriction of certain responses and, for all subjects other than Sarah, both of the ta rgeted responses (reading and mathematics) were observed during response-restriction pha ses. Although there was variability in effects of sequential re striction across replications for all subjects, the fact that, during some brief-restriction analyses, certain respons es had to be restricted before targeted responses occurred suggests that these responses may also be competing with the targeted responses outside of experimental sessions. Al so, the fact that targ eted responses were observed in the absence of experimenter imposed contingencies suggests that these contingencies may not be necessary to produce targeted responding. The brief restriction analyses for Kyle suggested that playi ng video games, playing pool, and watching television may compete with reading and that playing video games, playing pool, watching television, and reading may compete with doing mathematics. For John, they suggested that playing on the computer, watc hing movies, playing a handheld electronic game, and playing on a therapy ball may comp ete with both reading and mathematics. For Sarah, watching television a nd movies were identified as responses that may compete with reading. For Beth, brief restriction an alyses suggested that watching television, playing with money, playing with clay, and doing mathematics may compete with reading and also that watching television, playing with money, and playing with clay may compete with doing mathematics. In Study 2, a response-restriction treatment was implemented for each of these participants in which responses identified as possibly competing with targeted responses in brief rest riction analyses were restricted for greater

PAGE 41

33 periods of time in order to determine if restri ction of these responses in the absence of an experimenter imposed contingency was suffi cient to produce increases in targeted responding in naturally occurring, complex, multi-response environments.

PAGE 42

34 CHAPTER 3 STUDY 2: AN EVALUATION OF RESPONSE RESTRICTION AS A METHOD FOR INCREASING SOCIALLY DESIRAB LE RESPONSES IN A NATURALLY OCCURING MULTI-RESPONSE ENVIRONMENT In Study 2, the efficacy of response restriction as a treatment method for increasing the likelihood of soci ally desirable targeted responses in naturally occurring multi-response environments was examined. The utility of the brief restriction analyses conducted in Study 1 in identifying responses that compete with targeted behavior was evaluated by restric ting access to these responses fo r greater periods of time and examining response reallocation. When restrict ion of these responses was insufficient to produce reliable increases in targeted respondi ng, the restriction of additional responses was evaluated. Method The participants and setti ngs were identical to thos e in Study 1. During freeoperant conditions, no restrictions were placed on responding. During the first responserestriction phase for all participants, those re sponses identified during the brief restriction analyses as potentially competi ng with one of the targeted re sponses were restricted. For Kyle, the restricted responses were play ing video games, playing pool, and watching television. For John, the restricted responses were playing on the computer, watching movies, playing on a Leap Pad, and playing on a therapy ball. For Sarah, the restricted responses were watching television and movies For Beth, the restricted responses were watching television, playing with money, and playing with clay. If restriction of these responses was not sufficient to produce in creases in targeted responding, additional

PAGE 43

35 restriction phases were implemented in whic h the most probable response observed in the previous restriction phase was al so restricted. As in Study 1, responses were restricted by instructing participants that they may not e ngage in these responses No participants attempted to engage in restricted responses. A reversal design was used in which freeoperant conditions were altern ated with each response-restri ction condition. Data were collected on duration of observed responses and plotted in five-minute units. Data were collected during 15to 30-minute periods, on one to three days pe r week depending upon the availability of the participants. Results and Discussion Figure 5 shows the results for the respons e-restriction treatment for Kyle. During the free-operant condition, r eading and mathematics occurr ed briefly, but mainly he watched television and played video games. During the first exposure to the restriction condition (R1), Kyle allocated time toward ma thematics during 100% of all sessions. In a brief reversal to free operant, he watched television exclusively. In the second exposure to the R1, he spent all of his time drawing. In another reversal to free operant, he began by engaging in mathematics, but ended the c ondition by watching tele vision. In the third exposure to R1, he again spent 100% of his time drawing. Thus, the second restriction condition (R2) involved restricti on of drawing in addition to re striction of the previously mentioned preferred activities. This modifi cation resulted in exclusive responding to mathematics. When mathematics was restrict ed in addition to othe r preferred activities (during R3), he allocated all of his time toward reading. Figure 6 displays the results for the respons e-restriction treatment for John. In the first free-operant condition, John exclusively spent his time playing computer games. In

PAGE 44

36 0 50 100 150 200 250 300 0102030405060708090 Playing Video Games Playing Pool Watching Television Math Reading Drawing FO R1FO R1 FO R1 FO R2 FO R2 FO R3FOR3 Figure 5. Seconds of responding during the resp onse-restriction treatment for Kyle. FO signifies free-operant phases. During R1 phases, access to video games, pool and television was restricted. Duri ng R2, video games, pool, television, and drawing were restricted. During R3, video games, pool, television, drawing, and math were restricted. Seconds of Responding Sessions

PAGE 45

37 0 50 100 150 200 250 300 05101520253035 FOFO FO F O R1R1R2R2 Playing on a Computer Reading Math Figure 6. Seconds of responding during the re sponse-restriction treatment for John. FO signifies free-operant phases. Duri ng R1 phases, access to the computer, videos, electronic game, and therapy ball was restricted. During R2 the computer, videos, electronic game, ther apy ball, and books were restricted. Seconds of Responding Sessions

PAGE 46

38 the first restriction phase (R1), reading initia lly increased but then decreased toward the end of the phase as he allocated more time to doing mathematics. In the following freeoperant phase, John played computer games for the entire condition. In a second exposure to R1, reading again increased at the beginning of the condition but decreased over time. In the subsequent free-operant phase, John spent all of his time playing computer games. In the second restriction condition (R2), reading was restricted in addition to the responses restricted in R1 a nd mathematics initially increased but did not maintain. In another reversal to free opera nt, John played on the computer during 100% of all sessions. In a second exposure to R2, there was an initial increase in mathematics but a downward trend across the condition. As with Kyle, reliable increases in targeted responses were observed following response re striction, however, for John, increases in targeted responses did not maintain across re striction conditions. Nonetheless, in the final R2 condition, he spent over 16 of the 20 total minutes working on mathematics. Figure 7 shows the results of the restrict ion treatment for Sarah. In the first freeoperant phase, most of her time was spent wa tching television. In the first restriction phase (R1), playing with a si nging toy, playing a handheld vide o game, playing with clay, and playing with cards were all observed, w ith the most likely response being playing with the singing toy. In a return to free operant, Sarah spent nearly all of her time watching television. In the following R1 phase Sarah primarily worked on crafts. In another reversal to free operant, she spen t 100% of her time watching television. In a return to R1, the most probable response wa s again doing crafts and in the subsequent free-operant phase she, again, spent all of her time watching television. During the

PAGE 47

39 0 50 100 150 200 250 300 161116212631364146515661667176818691 Watching Television Magazine Reading Playing an Electronic Game Playing with Clay Playing with Dolls Listening to a CD Player Playing with a Singing toy Reading Playing with Cards Doing Crafts Listening to a Radio FOFO FO FO R1R1R1R2FOR2FOR3 R3 FO Figure 7. Seconds of responding during the resp onse-restriction treatment for Sarah. FO signifies free-operant phases. Duri ng R1 phases, access to television and videos was restricted. During R2, television, videos, and crafts were restricted. During R3, television, videos, crafts, and the radio were restricted. Seconds of Res p ondin g Sessions

PAGE 48

40 second restriction condition (R2) television, videos, and crafts were restricted and Sarah exclusively spent her time listening to musi c. In the next free-operant phase, Sarah watched television during 100% of all sessions. In a return to R2, she again exclusively listened to music. In the following freeoperant condition, she spent all of her time watching television. In the third restriction condition (R 3), access to a radio was restricted along with those responses restrict ed in R2 and Sarah spent 100% of her time reading a magazine. In the subsequent freeoperant phase, she again exclusively watched television. In a return to R3, Sarah, again, re ad a magazine during 100% of all sessions. For Sarah, restriction of certain responses reliably increased certain other responses. While the response restrictions did not pr oduce increases in the initially targeted academic-related reading and mathematics activ ities for Sarah, her ca regivers indicated that they preferred her spending her time on the activities that were produced through response restriction (working on crafts, listening to music, an d reading a magazine) to her watching television (her mo st common response during fr ee-operant periods). Figure 8 depicts the results for the respons e-restriction treatment for Beth. During the first restriction condition (R1), her respond ing was variable, with playing with dolls, reading, and mathematics all observed. Duri ng the first free-ope rant condition, Beth predominantly watched television. In a retu rn to R1, she spent all of her time doing math. In the following free-operant phase, Beth allocated all of her time to watching television. In the next exposure to R1, Beth primarily worked on crafts during the first four sessions and mathematics in the final tw o sessions. In a retu rn to free-operant, Beths responding was variab le, with reading, doing craf ts, watching television, and playing a handheld video game all occurring. In the following restriction phase, Beth

PAGE 49

41 Figure 8. Seconds of responding during the resp onse-restriction treatme nt for Beth. FO signifies free-operant phases. Du ring R1 phases, access to television, counting money, and playing with clay was restricted. 0 50 100 150 200 250 300 0102030405060708090 Watching Television Playing an Electronic Game Doing Crafts Reading Math Playing with Dolls Writing in a Diary Playing with a Ball Watching Movies R1FOR1FOR1FOR1FO R1 FO R1 FOR1 Seconds of Res p ondin g Sessions

PAGE 50

42 spent most of her time reading. In the s ubsequent free-operant phase, she exclusively watched television. In the ne xt response-restriction phase, Beth allocated most of her time to math. During the return to free operant, she again spent 100% of her time watching television. In the next R1 phase, Beth spent most of her time watching a movie. In the following free-operant phase, she excl usively watched television. In the final response-restriction phase, Beth worked on math during the beginning of the condition and worked on crafts during the end of the condition. For Beth, while no reliable effects of response restriction were observed, increa ses in targeted behavi or were observed in various restriction phases. Response restriction produced increase s in targeted responding in naturally occurring, multi-response environments, for thr ee of four participants. However, these increases were only reliable and sustained for Kyle. For John, response restriction produced reliable initial increases in targ eted behavior, but th ese increases did not maintain. For Sarah, response restriction did not produce increases in targeted responses but did produce reliable increase s in other desirable responses. In Beths case, response restriction had variable effects, producing increases in targeted responding in some phases, but not reliably. In or der to evaluate an alternativ e treatment for Kyle and to identify a method for producing more desira ble treatment outcomes for John, Sarah, and Beth, the effects of contingent access to re sponses on targeted behavior in naturally occurring, multi-response environments for each participant were examined in Study 3.

PAGE 51

43 CHAPTER 4 STUDY 3: AN EVALUATION OF CONTINGE NT ACCESS TO RESPONSES AS A METHOD FOR INCREASING SOCIALLY DESIRABLE RESPONSES IN A NATURALLY OCCURING MULTI-RESPONSE ENVIRONMENT Study 2 showed that response restric tion alone reliably produced sustained increases in targeted responding for one of f our participants. In study 3, the effects of contingent access to responses on targeted behavior in multi-response environments were examined for all four particip ants. Initially, the effects of contingent access to multiple responses were compared to either free-operant or response-restricti on control conditions. Following this, the effects of contingent access to a single high probability response were examined. Method The participants and settings were identic al to those in Studies 1 and 2. A single response was targeted for increase for each part icipant, reading for Kyle, math for John, and reading for Sarah and Beth. For Kyle a nd John, contingent respons es were those that were likely to be observed under free-opera nt conditions and those observed to be probable following response restriction in St udy 2, but not those which, when all were restricted, resulted in an increase in the targeted responses. This ensured that the contingent responses had a high likelihood of occurrence and that th eir restriction alone would not result in increases in targeted beha vior. Because reliable increases in targeted behavior were not observed for Sarah and Bet h, the same responses used in the initial response-restriction phases in Study 2 were used as initial co ntingent responses in Study

PAGE 52

44 3. These responses were found to be likel y under free-operant condi tions or following restriction of probable responses, but thei r restriction did not produce increases in targeted behavior. For Kyle, the initial cont ingent responses were playing video games, watching television, playing pool, and drawing. For John, the in itial contingent responses were playing on the computer, watching movies and playing an electronic game. For Sarah, the initial contingent responses were watching television and movies. For Beth, the initial contingent res ponses were watching television, counting money, and playing with clay. For all participants, follo wing the evaluation of a multiple-response contingency, the effects of contingent access to a single response were examined. For Kyle, the single contingent response used in this analysis was playing video games. For John, it was playing on a computer. For both Sa rah and Beth it was watching television. The single contingent responses were those en gaged in during periods in which access to contingent responses was delivered in the multiple-response contingency phases. During single-response contingency conditions, only th e one contingent response was restricted and all other responses were concurrently available with the in strumental response. Free-operant and response-restricti on phases were conducted in a manner identical to those in Studies 1 and 2 except th at participants were told which responses were restricted and that they were free to engage in targeted responses, other nonrestricted responses, or nothing at all. For all participants other than John, during all response-contingency phases, access to cont ingent responses was restricted and each minute spent engaging in the targeted res ponse produced a token ex changeable for one minute of access to restricted responses fo llowing the days observation period. Poker chips were used as tokens for all participan ts. Observation periods lasted between 20 and

PAGE 53

45 30 minutes. At the initiation of each response-contingency observation period, participants were told that they could earn to kens that they could exchange for access to restricted responses by doing targeted responses and that they were free to choose to do the targeted response, anothe r activity, or nothing at all. Johns initial responsecontingency phase was conducted in a manner iden tical to those of the other participants. Following this phase, for John, a 21.6 by 27.9 cm plastic board was constructed with 20 Velcro squares arranged in four rows of fi ve on which John could place earned tokens. In this phase, tokens were poker chips with Ve lcro squares attached to one side. In all subsequent response-contingency phases for John, no tokens were used and each minute of engagement in targeted behavior produced one minute of access to restricted responses immediately. Session time was stopped during periods in which access to restricted responses was provided. Results and Discussion Figure 9 shows the data for the response-c ontingency analysis for Kyle. In phase 1, with access to playing video games, wa tching television, playing pool, and drawing contingent upon reading, Kyle spent all of hi s time reading. In the following free-operant condition, no reading was observed. Kyle spent 100% of his time during all sessions reading during a return to contingent access to video games, watching television, playing pool, and drawing. In the next free-operant phase, no reading occurr ed. We noted that during all token-exchange periods Kyle excl usively engaged in playing video games. When access to playing video games alone wa s contingent upon reading and Kyle spent all of his time reading. In the following video game restri ction phase, reading was not observed. When access to playing video games was again contingent upon reading in the

PAGE 54

46 Figure 9. Seconds of responding during the cont ingency treatment for Kyle. FO signifies free-operant phases. The abbreviations Cont. and Rest. signify contingency and response restriction conditions respectively. 0 50 100 150 200 250 300 0510152025303540 VG, TV, Pool, Draw Cont. VG, TV, Pool, Draw Cont. FOFO VG Cont. VG Rest. VG Cont. Sessions Seconds of Responding

PAGE 55

47 0 50 100 150 200 250 300 051015202530354045505560657075Computer, Video, and Electronic Game Rest. Computer, Video, and Electronic Game Cont. Computer, Video, and Electronic Game Cont. New Token Board Computer, Video, and Electronic Game Direct Cont. Computer, Video, and Electronic Game Rest. Computer, Video, and Electronic Game Direct Cont. Computer Rest. Computer Cont. Computer Rest. Computer Cont. Figure 10. Seconds of responding during th e contingency treatment for John. FO signifies free-operant phases. The abbreviations Cont. and Rest. signify contingency and response restri ction conditions respectively. Seconds of Math Sessions

PAGE 56

48 subsequent phase, no reading was observed duri ng the first four sessions, then responding increased for the remainder of the phase. Figure 10 depicts Johns resu lts. In the first restriction phase, John spent no time doing math. In the first contingency phase, doing math increased and was observed at variable levels. In the subsequent phase, John was provided with a board on which he could place his tokens and his doing math remained variable during the first half of the phase but was eliminated during the final thr ee five-minute periods. In the next phase, the tokens were removed and each minute of doing worksheets produced immediate access to restricted responses and John spent near ly all of his time on ma th. In a return to the restriction phase, no math was observed. In the following contingency phase, John spent almost all of his time doing math. As was the case with Kyle, John exclusively engaged in a single response during all periods in which contingent access to responses was permitted. For John, this response was playing on a computer. In phase 7, the effects of computer restrict ion alone were examined and John spent no time working on math. When access to the computer was made contingent upon doing math in the following phase, doing math increased in the third five-minute period and John exclusively engaged in this re sponse for the remainder of the phase. In a return to computer restriction, John spent no time was spent on math. In the final contingentcomputer phase, John allocated all of his time to doing math. Figure 11 depicts the results for Sarah. During the first phase, access to watching television and videos was c ontingent upon reading and high levels of reading were observed. When watching televisi on and movies were restricted with no contingency in the following phase, no reading was observed. In a return to contingent television and

PAGE 57

49 Figure 11. Seconds of responding during the contingency treatment for Sarah. FO signifies free-operant phases. The abbreviations Cont. and Rest. signify contingency and response restri ction conditions respectively. Seconds of Reading Sessions 0 50 100 150 200 250 300 01020304050607080 TV and Video Cont. TV and Video Rest. TV and Video Cont. TV and Video Rest. TV Cont. TV Rest. TV Cont. TV Rest. TV Cont.

PAGE 58

50 videos, Sarah spent all of her time reading. In a second restricti on phase, no reading was observed. During all periods in which access to restricted responses was delivered during contingent access to responses phases Sarah ex clusively watched television. In phase 5, contingent access to watching television alone resulted in Sarah spending all of her time reading. When access to watching television wa s restricted, with no contingency, in the following phase, reading occurred during the fi rst four five-minute periods and then did not occur for the remainder of the phase. In a return to contingent television, reading was not observed during the first seven five-m inute periods, but occurred during all subsequent sessions of the phase. In a s econd television-restrict ion condition, no reading occurred. In a final contingent-television c ondition, Sarah spent all of her time reading. Figure 12 depicts the data fo r the response-contingency an alysis for Beth. In the first television, counting money, and clay re striction, no reading was observed. When these responses were contingent upon readi ng, in the following condition, high levels of reading were observed. In a return to restric tion, reading again decr eased to zero levels. When the contingency was reinstated, Beth spen t nearly all of her time reading. As was the case with Sarah, Beth exclusively watche d television during token-exchange periods in which access to previously restricted responses was permitted. In phase 5, access to watching television was continge nt upon reading and high levels of reading maintained. When access to watching television was rest ricted in the following phase, time spent reading decreased. In a return to contingent television, reading init ially increased during the first four sessions, decreased to zero le vels during the follow ing eight sessions and then increased to high levels for the remainder of the phase (12 sessions). In a return to

PAGE 59

51 Figure 12. Seconds of responding during the contingency treatment for Beth. FO signifies free-operant phases. The abbreviations Cont. and Rest. signify contingency and response restri ction conditions respectively. Seconds of Reading Sessions 0 50 100 150 200 250 300 0102030405060708090100 TV, Money, Clay Rest. TV, Money, Clay Cont. TV, Money, Clay Rest. TV, Money, Clay Cont. TV Cont. TV Rest. TV Cont. TV Rest. TV Cont.

PAGE 60

52 restriction, reading was not obs erved for the entire phase. In another reversal to contingent television, reading increase d and occurred at variable levels. For all participants, contingent access to multiple responses produced reliable increases in targeted behavior in naturall y occurring multi-response environments. When responses used as contingent responses were re stricted in the respons e-restriction analysis (study 2), no increase in reading was obser ved suggesting that th e response contingency was responsible for the increas e in, and maintenance of, ta rgeted behavior during the response-contingency analyses. Another findi ng common to all participants was that, during periods in which access to restrict ed responses was permitted in responsecontingency phases, responding was allocated exclusively to a singl e behavior (playing video games for Kyle, playing on a computer for John, and watching television for both Sarah and Beth). While there was more vari ability in instrumental responding than was observed when multiple contingent responses we re used, contingent access to this single response resulted in increases in targeted behavior for all participants even though multiple alternative responses were simultaneous ly available. These increases were not observed when the contingent responses were re stricted but no continge ncy was in effect. This finding suggests that contingent acce ss to a single high-probability response can serve as reinforcement for soci ally significant behavior in a complex, naturally occurring environment even with concurrently avai lable access to multi ple other potentially competing responses.

PAGE 61

53 CHAPTER 5 GENERAL DISCUSSION With the advent of preference assessment methodology, the empirical pre-treatment identification of reinforcers for use in clini cal application has become standard in the field of applied behavior analysis (Han ley, Iwata, Lindberg, & Conners, 2003) and the utility of these methods in identifying rein forcers that are effective components of behavioral treatments has been demonstrat ed numerous times. Typical empirical reinforcement identification methods involve a relatively simple response being exposed to contingent access to a number of items or activities in an environment in which potentially competing responses are largely unav ailable. The effectiveness of reinforcers identified in these preparations in app lication depends upon their generality extending across all of the variables that differ be tween the assessment and the application environments. Research on the response-d eprivation hypothesis sugge sts an alternative method of reinforcer identification. Baseline levels of responding are measured in the treatment environment and access to a conse quence response is provided contingent on an instrumental response such that the ratio of instrumental to consequence responses is greater than that observed dur ing baseline. The response-de privation hypothesis has been shown to effectively identify re inforcers and mitigate to so me degree concerns regarding the circularity of post hoc reinforcer iden tification (Timberlake & Farmer-Dougan, 1991). Furthermore, as baseline levels of respondi ng are measured in the environment in which the reinforcement contingency will be put into place, there is less risk that the generality of the identified reinforcers will not extend into the treatment context. The use of the

PAGE 62

54 response-deprivation hypothesis for reinforc er identification w ould seem to have particular promise for use in naturally occu rring environments in which access to many responses is available and in which the mu ltiple sources of reinforcement maintaining these responses may compete with reinforcers id entified in more restricted environments. Much research in this area, however, has occu rred in the context of restricted operant preparations in which only a limited number of responses are available. The present series of studies extends prior work on res ponse restriction, the Pr emack principle, and the response-deprivation hypothesis by exp licitly examining th e use of response restriction and contingent access to responses to increase socially desirable behavior in complex, naturally occurring, multi-response environments. In Studies 1 and 2, the effects of resp onse restriction alone were examined. Previous work has demonstrated that the restriction of access to behavior, in the absence of contingencies, can result in an increase in other behavior in both restricted(Hanley et al., 2000) and multi-response (Bernstein & Ebbesen, 1978; Lyons & Cheney, 1984; Green & Striefel, 1998) environments. In prev ious studies, however, such increases in behavior have been idiosyncratic across s ubjects, with no single rule adequately describing response reallocation following the re striction of a response. The results of the present studies replicated this finding. The brief re sponse restriction analyses conducted in Study 1 produced results that we re both variable across participants and across replications with in participant. Under conditions of response restriction in Study 2, with no response contingency in place, increases in targeted responses were observed for three of four subjects (Kyle, John, and Beth). For Kyle, reliable and su stained increases were observed in both

PAGE 63

55 reading and math. For John, reliable initial in creases were observed in both reading and math, however, these increases were not sustained over extended periods of time. For Beth, increases in both math and readi ng, among other responses, were observed following restriction of competing responses, ho wever, these increases were not reliable or predictable. One potentially important finding was the emergence of reasonably desirable responses that were not initially targeted for incr ease. For Kyle, an increase was observed in drawing following restriction of television and video games. For Beth, working on craft projects increased with re striction of televisi on, counting money, and playing with clay. For Sarah, when television and movies were restricted, increases were observed in working on crafts, listening to mu sic, and reading magazines. These findings suggest that the restriction of perhaps a less desired res ponse (e.g., watching television) may result in the serendipitous increase in a more desired response (e.g., drawing) in the absence of imposed contingencies. A lthough the particular outcome of response restriction was idiosyncratic acr oss participants, these data im ply response restriction as a potentially effective method for use in multiresponse environments. For example, at time when school work is not necessarily requ ired, response restriction might be used to produce a greater diversity of leisure activities. When effective, response restriction may provide a treatment option for clinicians or parents that is both relatively easy to im plement and not prone to failures in treatment integrity (i.e., response restri ction does not require counti ng of responses, timing, or the delivery of stimuli leaving re latively little room for erro r). The general methods of Studies 1 and 2 may serve as an initial guide for clinicians attempting to make use of response restriction as a treatment, Kyles ou tcome is an example of the most favorable

PAGE 64

56 result. Although the brief analyses of Study 1 produced variable outcomes, they were successful in identifying some responses that may have been competing with targeted behavior. This may provide a starting point for a more extended analysis which might identify further responses that may be competing with target behavior. The factors responsible for changes in re sponding under response restriction remain to be identified. It may be that responses that increase in frequency are members of the same operant class as those restricted and might be cons idered members of a responseclass hierarchy (Harding, Wacker, Berg, Barretto, Winborn, & Gardner, 2001; Lalli, Mace, Wohm, & Livezy, 1995; Richman, Wack er, Asmus, Casey, & Andelman, 1999). Alternatively, it may be that responses that in crease in frequency are either automatically or socially reinforced and that restricted responses produced more potent reinforcement and thus competed with them. A third possi bility is that some responses may never increase in frequency as a function of restric tion of alternative responses alone (i.e., those that are not in a participants repertoire or those that are not automatically or socially reinforced). Sources of cont rol that may produce variability in behavior under conditions of response restriction have also not yet been identified. It is possible, for example, that fluctuations in motivating opera tions, the presence or absence of relevant discriminative stimuli, or changing contingencies on any num ber of responses may af fect the outcome of restriction of access to responses in multi-operant environments. In Study 3, the effects of contingent acce ss to responses on socially important behavior in naturally occurri ng, unrestricted environments were examined. Both the Premack principle and the response-depriva tion hypothesis would predict increases in instrumental responding given th e contingencies arranged in Study 3 (i.e., more probable

PAGE 65

57 responses were made contingent upon the occurrence of less proba ble responses in a manner that would make the ratio of instrume ntal to consequence re sponses greater than that observed in free-operant phases). Ho wever, both the Premack principle and the response-deprivation hypothesis ar e silent with regard to an y possible effects of several alternative sources of reinfor cement being available. In Study 3, contingent access to multiple high-probability responses and a si ngle high-probability response arranged such that the ratio of instrumental to consequence responses wa s greater than that observed during free-operant periods produced reinforc ement effects for all participants. This finding further extends the generality of the Premack principle and the responsedeprivation hypothesis, showing reinforcement e ffects for socially desirable behavior in complex, naturally occurring, unr estricted environments. The effectiveness of the methods used in Study 3 suggests some promise for general application in multi-re sponse settings. The initia l consequence responses were chosen based on the outcomes of Studies 1 a nd 2, but it may not have been necessary to conduct both studies to identify effective con tingent responses. The single consequence response effective as a reinforcer for Ky le was playing video games, playing on a computer for John, and watchi ng television for Sarah and Beth. The results from the free-operant phases of the brief response restriction analyses, conducted in Study 1, indicated that the most probable responses fo r Kyle were video game playing and math, playing on the computer for all assessment s for John, watching television during two assessments and math during one for Sa rah, and watching television during two assessments and weaving during one for Bet h. When the data obtained during freeoperant periods in Study 1 are aggregated (i.e., the durations of each response in each of

PAGE 66

58 the free-operant periods are added), the most likely free-operant response observed in Study 1 is the same response that was demonstr ated to maintain targeted behavior in Study 3 for all subjects other than Kyle. Fo r Kyle, only two brief restriction analyses were conducted and playing video games was th e most likely response in one of the two. Whether or not playing video games would have emerged as the most likely response in further brief free-operant periods is unknown, but seems likely based on subsequent data. These data, for three of four participants, s uggest that effective contingent responses for use as reinforcers may be identified thr ough relatively brief free-operant observation periods in the targeted environment. Furthermore, the most probable response observed in the initial free-operant phase s in Study 2 for all participan ts was the response shown to serve as an effective reinfor cer in Study 3, again suggesting that a pre-treatment measure of free-operant responding in a particular environment may be useful in identifying responses to use as reinforcers. Research is needed in numerous areas to determine the conditions under which response restriction and contingencies will be mo st effective. First, because the current studies and previous research have found response restrict ion to produce idiosyncratic results, research is needed to examine the influences of potentially relevant variables including motivating operat ions, response effort, the presence or absence of discriminative stimuli, and contingencies for alternative respons es under conditions of response restriction. Second, because both the Premack principle and the responsedeprivation hypothesis predict th e obtained reinforcement effe ct in Study 3, evaluations are needed using a less-probable response with the ratio of instrumental to consequence responses being greater than that observed during free-operant peri ods (the two models

PAGE 67

59 make different predictions). Third, future re search is needed to directly compare the effectiveness of Premack/response-deprivation approaches to identifying reinforcers to those based on relatively standard preferen ce assessment methods (e.g., DeLeon & Iwata, 1996; Fisher, Piazza, Bowman, Hagopian, Owe ns, & Slevin, 1992; Pace, Ivancic, Edwards, Iwata, & Page, 1985; Roane, Vo llmer, Ringdahl, & Marcus, 1998; Windsor, Piche, & Locke, 1994). The outcome of such a comparison would sp eak directly to the applied utility of using contingent acce ss to behavior observed in a treatment environment as reinforcement.

PAGE 68

60 LIST OF REFERENCES Allison, J. & Timberlake, W. (1974). Instrume ntal and contingent saccharin licking in rats: Response deprivation and reinforcement. Learning and Motivation, 5, 231247. Amari, A., Grace, N. C., & Fisher, W. W. (1995). Achieving and maintaining compliance with the ketogenic diet. Journal of Applied Behavior Analysis, 28, 341-342. Bernstein, D. J. (1998). Establishment of a laboratory for continuous observation of human behavior. In K. A. Lattal & M. Perone (Eds.), Handbook of research methods in human operant behavior (pp. 509-539). New York: Plenum Press. Bernstein, D. J. & Ebbesen, E. B. (1978). Re inforcement and substitution in humans: A multiple-response analysis. Journal of the Experimental Analysis of Behavior, 30, 243-253. Bernstein, D. J. & Michael, R. L. (1990). The utility of verbal and behavioral assessments of value. Journal of the Experimental Analysis of Behavior, 54, 173-184. Catania, A. C. (1998). Learning. Upper Saddle River, NJ: Prentice Hall. Charlop, M. H., Kurtz, P. F., & Casey, F. G. (1990). Using aberrant behaviors as reinforcers for autistic children. Journal of Applied Behavior Analysis, 23 163181. DeLeon, I. G. & Iwata, B. A. (1996). Eval uation of a multiple-s timulus presentation format for assessing reinforcer preferences. Journal of Applied Behavior Analysis, 29 519-532. Fisher, W., Piazza, C. C., Bowman, L. G., Ha gopian, L. P., Owens, J. C., & Slevin, I. (1992). A comparison of two approaches fo r identifying reinforcers for persons with severe and profound disabilities. Journal of Applied Behavior Analysis, 25 491-498. Green, G. & Striefel, S. (1988). Response restriction and substitution with autistic children. Journal of the Experimental Analysis of Behavior, 50, 21-32. Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of problem behavior: A review. Journal of Applied Behavior Analysis, 36, 147-185. Hanley, G. P., Iwata, B. A., Thompson, R. H., & Lindberg, J. S. (2000). A component analysis of "stereotypy as reinforc ement" for alternative behavior. Journal of Applied Behavior Analysis, 33, 285-297.

PAGE 69

61 Harding, J. W., Wacker, D. P., Berg, W. K., Barretto, A., Winborn, L., & Gardner, A. (2001). Analysis of response class hierar chies with attention-maintained problem behaviors. Journal of Applied Behavior Analysis, 34 61-64. Homme, L. E., de Baca, P. C., Devine, J. V ., Steinhorst, R., & Rickert, E. J. (1963). Use of the Premack principle in controlling the behavior of nursery school children. Journal of the Experimental Analysis of Behavior, 6, 544. Konarski, E. A., Jr. Johnson, M. R., Crowell, C. R., & Whitman, T. L. (1980). Response deprivation and reinforcement in applied settings: A preliminary analysis. Journal of Applied Behavior Analysis, 13, 595-609. Lalli, J. S., Mace, F. C., Wohn, T., & Livez y, K. (1995). Identification and modification of a response-class hierarchy. Journal of Applied Behavior Analysis, 28 551-559. Laraway, S., Snycerski, S., Michael, J., & Poling, A. (2003). Motivating operations and terms to describe them: Some further refinements. Journal of Applied Behavior Analysis, 36, 407-414. Lyons, C. A. & Cheney, C. D. (1984). Time reallocation in a multiresponse environment: Effects of restricti ng response classes. Journal of the Experimental Analysis of Behavior, 41, 279-289. Meehl P. E. (1950). On the circ ularity of the law of effect. Psychological Bulletin, 47, 5275. Mitchell, W. S. & Stoffelmayr, B. E. (1973). Application of the Prem ack principle to the behavioral control of extrem ely inactive schizophrenics. Journal of Applied Behavior Analysis, 6 419-423. Osborne, J. G. (1969). Free-time as a reinfo rcer in the management of classroom behavior. Journal of Applied Behavior Analysis, 2, 113-118. Pace, G. M., Ivancic, M. T., Edwards, G. L., Iwata, B. A., & Page, T. J. (1985). Assessment of stimulus preference and re inforcer value with profoundly retarded individuals. Journal of Applied Behavior Analysis, 18 249-255. Postman, L. (1947). The history and pres ent status of the law of effect. Psychological Bulletin, 44, 489-563. Premack, D. (1959). Toward empirical behavior laws: 1. Positive reinforcement. Psychological Review, 66, 219-233.

PAGE 70

62 Richman, D. M., Wacker, D. P., Asmus, J. M., Casey, S. D., & Andelman, M. (1999). Further analysis of problem behavi or in response class hierarchies. Journal of Applied Behavior Analysis, 32 269-283. 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. Sidman, M. (1960). Tactics of scientific research. New York: Basic Books. Skinner, B. F. (1935). The generic nature of the concepts of stimulus and response. Journal of General Psychology, 12, 40-65. Skinner, B. F. (1953). Science and human behavior. New York: Macmillan. Timberlake, W. & Allison, J. (1974). Response deprivation: An empirical approach to instrumental performance. Psychological Review, 81, 146-164. Timberlake, W. & Farmer-Dougan, V. A. ( 1991). Reinforcement in applied settings: Figuring out ahead of time what will work. Psychological Bulletin, 110, 379-391. Windsor, J., Piche, L. M., & Locke, P. A. (1994). Preference testing: A comparison of two presentation methods. Research in Developmental Disabilities, 15, 493-455.

PAGE 71

63 BIOGRAPHICAL SKETCH Jason Bourret was born in Milwaukee, Wisc onsin. He entered the University of Florida in 1995 and graduated in 1999 with a degree in psychology. During his undergraduate studies, he served as a volunteer in a basic behavior analysis research lab working on choice research and two applied be havior analysis labs working primarily on research focusing on the treatment of severe behavior problems in individuals diagnosed with developmental disa bilities. In 1999, he enrolled in the behavior analysis program at the University of Florida to pursue a docto ral degree in psychology. He worked as a graduate research assistant unde r the direction of Dr. Timoth y Vollmer. Over the course of his graduate career at the University of Florida, he has been i nvolved in research on verbal behavior, quantitative analysis of be havior, response allocation in multi-response environments, descriptive analysis of behavi or, and schedules of re sponse-contingent and fixed-time reinforcement delivery. Expecti ng graduation in December, 2005, he plans on continuing his career in behavi or analysis through conducting behavior analytic research with implications for application in clinical settings.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20110318_AAAALW INGEST_TIME 2011-03-18T21:12:50Z PACKAGE UFE0012800_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 18009 DFID F20110318_AACEEC ORIGIN DEPOSITOR PATH bourret_j_Page_70.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
aa8aaa958a5ac75d74ec436371ba0c4f
SHA-1
0cea528b5829839b26c2a4ff0f18ae4d7ae3dade
94947 F20110318_AACDSH bourret_j_Page_39.jpg
1719d15f408ea39f803bc4b03a7186e1
f4f6ca43dca21cf127e0c16b18fce20973d2a9eb
16146 F20110318_AACDNK bourret_j_Page_41.jpg
676895585c5df95a51c6eda4d9211bf5
f826648e8ee0ff3e0a1da1c0a7a59238c25d7262
1051979 F20110318_AACEED bourret_j_Page_04.jp2
1113f0a8e767e1737fbad5bbd2f93075
4157e005b57057d4445778bb54f2fc9bf5b3717d
2300 F20110318_AACDXF bourret_j_Page_69.txt
955e47eed1039f195a318f56c7f044cc
82a33288cd83b38b44fa822590b0563ce6a45240
103811 F20110318_AACDSI bourret_j_Page_65.jpg
447cfe8781cc5377a29b6c326ee61b08
5caecf2facd96db922555831cc15d4199491383b
112072 F20110318_AACDNL bourret_j_Page_66.jp2
250d865265af4f19a6d6a6c91f087a66
e7708779c675c5b89754dc681e7a3de4d2da6a8a
292648 F20110318_AACEEE bourret_j_Page_05.jp2
80943cf045b9015b82883e4727a760a4
caa09e0351b7ef30124c40e2147f5c1ac60c2c8c
1541 F20110318_AACDXG bourret_j_Page_49.txt
1ef5f1cc393457b02e55f11b6434bc27
ba2d2e1cdd39d97ed86a6ff22909f7c6b29fdd90
8323 F20110318_AACDSJ bourret_j_Page_57.QC.jpg
993004d0e7e1c7dad8e50f07ccd589d2
c2a53bf7931c42dcec5323a87f5ea936fc828cdb
51734 F20110318_AACDNM bourret_j_Page_66.pro
227125c47fb24aeadf128a6f008fc7ce
086a514d61c5a4b81bd624716e079125fc610dca
110077 F20110318_AACEEF bourret_j_Page_21.jp2
21c5cdac4e5ff3b67e7e268c653e7899
098935f6fb412b9aa3a4410e5a22f4c445386927
31399 F20110318_AACDXH bourret_j_Page_69.QC.jpg
d267dc62570ea84245f7bc935be02577
9472b422ad97f1bebf4922117982d3ac0e86f152
105181 F20110318_AACDSK bourret_j_Page_29.jp2
970ef4374f23f3cbf00303f3633e0685
9169d849ac56b82ec2d04bbd6d87fc23a01ac75e
1613 F20110318_AACDNN bourret_j_Page_41thm.jpg
6df281ccce0a1a8ccfe91a10ca66574f
730c1506a6fd50cf6d14860d2162664ac7115dda
109812 F20110318_AACEEG bourret_j_Page_25.jp2
3dec7d9af3252813d805e5952e404bb7
01f1c51f99c69a03f13feb903d34eccf2aa7080c
3782 F20110318_AACDXI bourret_j_Page_44thm.jpg
b0ddb1db42797b3970986e7f1ca1745f
9daada613364e903621626b2796c4e21edce6da5
33108 F20110318_AACDSL bourret_j_Page_46.QC.jpg
d7837469e94fde26652a7e2458fa9f41
d1448cf53e58fa0f61c77839000a43722a802bbc
50050 F20110318_AACDNO bourret_j_Page_11.pro
e07c451edf96aa5406b5afc558a5abd8
f9e80e06f4cc85d98ebb1e4adc84867de9e18c0e
36635 F20110318_AACEEH bourret_j_Page_32.jp2
4ab4c99508b4e041f6d66241ac9cbf65
ba30a80e103462b6372031e4cada18d5df793c5f
10422 F20110318_AACDXJ bourret_j_Page_57.pro
9176aefcd7ba6ae57281ca37bccc2f61
6df10840f391b60574535e304c2605a8598443b5
28049 F20110318_AACDSM bourret_j_Page_28.QC.jpg
0bb136dfa34b9b33eb91be224e7a8f93
5bcc63341d85ca01e2282f21afd354aa3d1c7fd9
658 F20110318_AACDNP bourret_j_Page_59.txt
4995145c616ab5a0729a26576b37d446
baef5734677c795f2ac3d427ec2d6fe2aa6a56c0
99859 F20110318_AACEEI bourret_j_Page_39.jp2
43575938ab0269a1a480bae592723c6b
17e3d8a473995cb5e1c88031268e68bad6abb965
1053954 F20110318_AACDXK bourret_j_Page_57.tif
1b1f048281316e52d8bea1581a3913cc
689b739972aefc3f8f0454ee122b1b1b8399574a
1727 F20110318_AACDSN bourret_j_Page_60.txt
26f8d5fc21655c9658b1323d52c6959e
e3f0b60b9a82cbf1eb129d59f66bbbbc64b3edc4
111091 F20110318_AACDNQ bourret_j_Page_58.jp2
fb69d9681cab72e60a610feb348dd2c8
5f7d725f4f6f67703885b665bd4927a3e860d20b
18169 F20110318_AACEEJ bourret_j_Page_41.jp2
788ada35b8dbb71e8830adcf7616d670
d932d5884b7179570206e8a5125aa4711cf10c9a
51425 F20110318_AACDXL bourret_j_Page_65.pro
a2c194559991281ed38e8d5b11e5cc1d
8846c66744cfa3f2adbf5c2e18d648a539c0f486
7981 F20110318_AACDSO bourret_j_Page_26thm.jpg
6f6a0906ad91e3682da53416af5bffa0
3f9cf6ccc887d449039da3b2130c313595630f3f
103224 F20110318_AACDNR bourret_j_Page_24.jpg
31babc397f40511b8bae31e1f7ef68f5
ae4a46daca60b98a02779bc5f150446839c20b53
99014 F20110318_AACEEK bourret_j_Page_42.jp2
72b7e953c909e79162cf8f328adf27d6
375340438dfadf30f06ec4827cc1543fa3cc0bbd
1868 F20110318_AACDXM bourret_j_Page_39.txt
c731beb55a262a09ba306aed813d76de
6ad67232d876d7746251153d352a0c2d473a8ef0
50706 F20110318_AACDSP bourret_j_Page_25.pro
3ed550b5e1552452a174a976af530770
749e69fbe39596cf9d87f1ccaf9d6982e3fc11ee
3105 F20110318_AACDNS bourret_j_Page_67thm.jpg
d7689a458da785693ae20fa3def6c9d7
c3b832565739a0859249b4a438f7cdce34843d98
107721 F20110318_AACEEL bourret_j_Page_43.jp2
080db61cd17bcac06190a1dfec6a666b
4a3ef170cbb45b90581dacfac0cd2bcdddf39294
105606 F20110318_AACDXN bourret_j_Page_17.jp2
d54a88642140ea77526cbf63301813bf
3f006540a83c4d2b98923414e666e3a23d524e26
F20110318_AACDSQ bourret_j_Page_16.tif
26c1a13a31666afe1511cb9c273d2dbd
91643d31dcf5750f4ad89b4b2e08beacbdefb44e
31448 F20110318_AACDNT bourret_j_Page_68.QC.jpg
13cc51958c51addb86f853e4106eb053
1a5d2fc9e7360f27b02233c9ee2f3ba96ac49e12
35630 F20110318_AACEEM bourret_j_Page_45.jp2
0a3f9f97cf0ffa0c3314e0fbbaf09463
224aef11d97f8cfb96b1de6ffd7e85f814d0a358
38610 F20110318_AACDXO bourret_j_Page_49.jp2
469b2060ead6799fc30ab322add67fc0
59a900b526a6ab0ae24b65e3a8956a4e44e8a75e
1984 F20110318_AACDSR bourret_j_Page_56.txt
93bd94cda4e43b9e167f0b5c194835fa
69dabbc007fe80458be3025d9b88d60e98a28adb
15782 F20110318_AACDNU bourret_j_Page_32.pro
1a016271a7c0b1c9608bf92bfb31d654
a72380f6cccaca441044098b3c5ddb08daf6f1d4
107833 F20110318_AACEEN bourret_j_Page_46.jp2
13b37bceacc8a804dd67487637035690
9ca88fbbc7151043a48fe996f37e64b7e79df35b
4833 F20110318_AACDXP bourret_j_Page_31.pro
a4e902bc9e91693d367087a734744e43
2e135cf958635f212d65ad24cd916bbe95cb5782
F20110318_AACDSS bourret_j_Page_50.tif
22aade060bfd2da468dd6d28ea8b59cb
dc9c2aad370a34d3b4472eef388f2717f78b8d14
28807 F20110318_AACDNV bourret_j_Page_09.QC.jpg
f054a7640565948cc8903e303fd72d07
055947f2ba57ff3c38b294d56d3a8ec79da1e41f
114536 F20110318_AACEEO bourret_j_Page_52.jp2
c665e79886db60b404e449b27ff3262d
76f1761ec5a1955946c6ef6b95531f30811de51a
66999 F20110318_AACDXQ bourret_j_Page_70.jp2
21e6a2d46c40811d2ecd83fa128d8bab
b2f6a074be54fb3926db53a329560ac06fa174ce
68009 F20110318_AACDST bourret_j_Page_71.jpg
d78e6315d714bc6243810d3ee40aecd7
53af0fd5f96b058a303d3abc2355fdb9a5186a4e
5936 F20110318_AACDNW bourret_j_Page_41.QC.jpg
374011d62c17a9306a983029b8301d2a
8f389ccf8f87ffaead2394e79675699d86f6206b
109334 F20110318_AACEEP bourret_j_Page_53.jp2
4043b091ba98f0daa58b3f8e2c8cf22f
417d7555d129a8c0a23cfd05a2ea550276f98106
30593 F20110318_AACDXR bourret_j_Page_61.QC.jpg
43d22f7b74e92f168c440c6f01ed7030
b2ca30f47029217ed59d5a172fbc1b7c6918edec
103661 F20110318_AACDSU bourret_j_Page_58.jpg
d6aaf8e14ba52b017565f76bc87778d6
4300d8214419b012e54d4b42b31e18cdd439b4f6
8748 F20110318_AACDNX bourret_j_Page_18thm.jpg
266d159463d4a9fcf73eec9b53b8214b
3c67a1da424d9529bdab02e38e21a7e1d13d27c0
26681 F20110318_AACEEQ bourret_j_Page_54.jp2
23a599d93f21617e5b9a73b5442d30a3
2b18810f301fb8ace87f1c68a913b4a22bfca3de
110594 F20110318_AACDXS bourret_j_Page_69.jpg
fa8bdce8d6c1f97009fcedcc82a59367
5e432b7ec02685b51334867f74ea739dbfa5fef3
F20110318_AACDSV bourret_j_Page_56.tif
e442348cc2169e4b80d8ae0726b31851
0695eea5ef4092f9926c5a7a3340d06d93121985
60968 F20110318_AACDNY bourret_j_Page_70.jpg
05ccbeacaae4cb8f0c5e6cf5e4e02177
3ed8f5993fb7b971b853bd0845531ebe726f9acd
F20110318_AACDSW bourret_j_Page_01.tif
8739c3722b0f92d2e00603a3ce276eeb
d276062ba9e45cd5003d72b88cac486de9e1200c
99077 F20110318_AACDNZ bourret_j_Page_35.jpg
df2b257f326d03e561a4964b98cc7f40
aa0b269db38341d244502e66eaf842594d46f72d
17756 F20110318_AACDXT bourret_j_Page_55.pro
6fd4b483d2cc0d756ffd3c2035450a8d
18ad4387c3c001467fbc2b924eca2ae2feac6bee
113079 F20110318_AACEER bourret_j_Page_64.jp2
4c1429f6399929d0f1b8f33cc4f52afc
48825ef1870e6f68bcabfca567b99c205582e348
77325 F20110318_AACDSX bourret_j_Page_13.jp2
e415fd24aa936f4a923b64df7dd7154d
435274dd4b94b64571a1b98a35f2aec19da2091c
1932 F20110318_AACDXU bourret_j_Page_22.txt
885bdfa3dbe184fe024a0c57b61d8d24
78e36e502dc75add79ac1ad37307783d21fc5824
112720 F20110318_AACEES bourret_j_Page_65.jp2
5435f38aa5db13c5fef2f2ff2814e7bd
1e11ff254b6b5b6e8b4d9269e480f6ac446b7255
104231 F20110318_AACDSY bourret_j_Page_35.jp2
d0c53bc071102ee9dcc9fa65c837af29
d61fad913060fe298a50b150d9daf1d9e41e48a1
8512 F20110318_AACDXV bourret_j_Page_65thm.jpg
93f88936c9eb8c38dd9aaadb79121eb4
3a06667e40a27973e57a8afc6dc306cc506d3c88
6390 F20110318_AACEET bourret_j_Page_04thm.jpg
2e5c2a46d5c21c148de011ebfffca9d1
adc95bc5cd050d97cf999fc8ae7fda7717e8b66d
100610 F20110318_AACDQA bourret_j_Page_26.jpg
f6c65ac8630b6f6235ad77f83d6b28bf
1d6e95624ae135989e6e1553912798e1b4a2fbbb
95236 F20110318_AACDSZ bourret_j_Page_50.jp2
47537898d0b2c3ef2ce1c55398c01826
29dab3578ff12d44b4573e2a171588675fd8c7ab
1825 F20110318_AACDXW bourret_j_Page_13.txt
dd4ce12cf914a13be786711619739b79
3017f635ad28c3f6387fb35397ba17484a748da1
1473 F20110318_AACEEU bourret_j_Page_05thm.jpg
ad4277db756f22131b56a58b88102374
b78bfb775ff2efe881851f2e65b84312d8728333
3756 F20110318_AACDQB bourret_j_Page_32thm.jpg
b8eec939fb0f38498ec0a6757f532dae
0195e47c7f0f6fa599980c67ad2cc77c1e2cac55
12727 F20110318_AACDXX bourret_j_Page_67.QC.jpg
7f1e1126a713086fa868d6cd045fda4a
448477276e7162af2a6e86f956e7e051bef48a92
5708 F20110318_AACEEV bourret_j_Page_06thm.jpg
32bd93a13abcb08077c213ec91b3c794
2158d702e0d46a0433b7e63fa38b420d3e1343d6
1781 F20110318_AACDQC bourret_j_Page_28.txt
ef883cfed3a13f5c9f1d4a2459c57ee4
09e589ca03b4e09b9c358c9f1cc8c4b82765dfc6
F20110318_AACDVA bourret_j_Page_20.tif
4ce8f9c74371edf853e90b96fbdf551a
33ee11b0916c42485153df3a1a9524ba4f62d550
50105 F20110318_AACDXY bourret_j_Page_08.jp2
e5e5852447bd2e406b848f936c4c39c9
b1e11c0e51672dffacd89dee5b982d6f64825a24
8158 F20110318_AACEEW bourret_j_Page_14thm.jpg
f3bff497ad90d22e29668575d862a374
89b426e62b042a87d3777fdd27469bf62fa3c5df
113012 F20110318_AACDQD bourret_j_Page_48.jp2
9d201668e7a234c8ccadd6a5369f5964
f995431bb6a5a9643b3304764087d02f96c575e9
33495 F20110318_AACDVB bourret_j_Page_65.QC.jpg
b71e809cef4315eccea9d452b19162e2
f3569679fa12fcad6a7bc83f3fa7a7fa867bbe8e
52636 F20110318_AACDXZ bourret_j_Page_63.pro
7fe8c1f077f43ac77eccb990a6da9f83
4d9d76bbbdc485435a68ca65f33e2116ac6f13aa
7499 F20110318_AACEEX bourret_j_Page_16thm.jpg
bc41ad52321b1770f2ca1fca75fb1896
ad944297f7cb3ad55fc1a4630487571fc984a8ef
2200 F20110318_AACDQE bourret_j_Page_18.txt
1bb212453728f257467850e34f851c71
6614c32e9276b33fe6de1fd50a93982dfa24b5cb
32054 F20110318_AACDVC bourret_j_Page_35.QC.jpg
ec60c158228bd49c4a79a3576cec6e12
18b94fa3ad6fb8e75358903dfd6626aaf9efed65
7152 F20110318_AACEEY bourret_j_Page_28thm.jpg
b5bb4ed8646c82e8171ad5a99b74ce81
0b49cb0a2dbc9186d06d3563ec7c38eabf513106
859 F20110318_AACECA bourret_j_Page_08.txt
7981de166d7b0ce46f84c5a01ba94e35
0484483da096fbf7c254fb5dd2bc51daa29b9719
3219 F20110318_AACDQF bourret_j_Page_55thm.jpg
68544a5d393c371c6b5a81b3da525b10
9e9f6f32fcc9bfc7e207eb9e9b1e205cf28622b1
4426 F20110318_AACEEZ bourret_j_Page_34thm.jpg
155eb51e34608d49cc5c4514551fd328
af7f8df7cc7903dd7b1fc6ca14680cfd920deb70
1865 F20110318_AACECB bourret_j_Page_09.txt
83bcc94f98b79224871c961865260b65
fcebc524168e2c2a6768aa23fa1118392896d972
51187 F20110318_AACDVD bourret_j_Page_20.pro
5f76ca505cd4e972865b9c89bb2591be
d7f1be3793cd094b181ef3499f71dc76455b163c
48841 F20110318_AACDQG bourret_j_Page_33.pro
7a20f1611dd737c0b03aab3ac0f9ce4a
7abec9b6b4cc01001f60c3cf986e4488cf8b4f8e
1918 F20110318_AACECC bourret_j_Page_10.txt
6859b629002dba41719301fcd28e6046
465230418793ec983e3c30596793803c8204cd96
8325 F20110318_AACDVE bourret_j_Page_56thm.jpg
d5b49dec727b87767d339bf33be723e4
37a20d7ebfdc5da589126ad681bbb506af1e582d
F20110318_AACDQH bourret_j_Page_37.tif
01b88ebb11d2652cca0d6c5340540340
6e9a08c7acf99c64f9bf11b7dbd474db173454a5
1826 F20110318_AACECD bourret_j_Page_16.txt
b025c4aa958330ff7086fce80722bb56
2249fef747ed08ce1a4bfebef72c7bb949ed0687
11393 F20110318_AACDVF bourret_j_Page_03.QC.jpg
a7259645215678af00d807d22e5b838f
d6cb3252f2e8c37a0919fdd3cf47229f99cb1054
F20110318_AACDQI bourret_j_Page_31.tif
35ba3d77abca7fe326e588d89751ccab
17e46f6846ef69e09816f8b540ce9bbfc1f15284
2020 F20110318_AACECE bourret_j_Page_24.txt
3dce9f1fd2f5f64a000eb145afd5888e
88f6e2ca7ec8875e5776369a6fbcc588185b0e87
29606 F20110318_AACDVG bourret_j_Page_70.pro
840ff2d2a4c67aff398d2a7131d9bfe0
f5fcae1b957c097c230649e49ec54ee2c115c8fb
21145 F20110318_AACDQJ bourret_j_Page_36.pro
2154a0acb2f8a49fe3062efbc63b9b42
38041da38355cf419498995679cad8dfcb8127d4
2126 F20110318_AACECF bourret_j_Page_30.txt
2fe88ce36fbf388d7e77f53b3d865eae
6bae8a427e5d75997d9b024215c8ff37e3264f5f
35493 F20110318_AACDVH bourret_j_Page_13.pro
90bf80ff4a075294f4ee4ac7b62eea06
df2404935ced51da05d3926c976a1138e2661431
41057 F20110318_AACDQK bourret_j_Page_06.pro
e5cc24550bed7eab8561affd37c443c9
948ece7b7e9cc335e05e51cd50a1048b5f181cc3
854 F20110318_AACECG bourret_j_Page_32.txt
4000ea90d5c4a7283c24c040b6b0db0c
2a480ca54ea70fec16ba4ed0e33f3be110a320b7
98211 F20110318_AACDVI bourret_j_Page_22.jpg
5298be4cc70fc72968a11c34f85493eb
b029bd3d21ee3ac7edea1cde61f57c55d12baf68
47275 F20110318_AACDQL bourret_j_Page_61.pro
6e98bc5168378054ac76aee5e73a814e
cc093e3e527baaafff1fa775c5d6dcc780e2b59f
1930 F20110318_AACECH bourret_j_Page_33.txt
32e1e80c0c34a06b5b9dc93396636209
0e7dd947835aa27f1ee10c3e6d00e0acff815680
26418 F20110318_AACDVJ bourret_j_Page_01.jpg
cc20bcbc592eb978509e4f6f51db24cb
631d07c93bbdd2f835c3696957bb57ef84a25160
F20110318_AACDQM bourret_j_Page_21.tif
1495d5d588d1a44c6f0b81210c37f309
f199f818c39656fde37e755191758e757f823205
1274 F20110318_AACECI bourret_j_Page_34.txt
f50e502abc09f000faed15048fac9d08
5538225504c731a60f9146aec5d014c519a83225
29127 F20110318_AACDVK bourret_j_Page_42.QC.jpg
99ded57862442e23aeb6d913e370ac4d
883a8c016df832eb9934747ce4c08528e92b87a5
30882 F20110318_AACDQN bourret_j_Page_59.jp2
ac432b3d01c5bca8a5bb1849cd818139
3a6586fd7e17f8bfb9d25f2c55809f3fe3f0bd70
269 F20110318_AACECJ bourret_j_Page_41.txt
dc7159d15d582809b49e0539902aee33
9bb38a24d39f8522cde5f76200ffb74718f6dfe5
50390 F20110318_AACDVL bourret_j_Page_26.pro
c4bd786cd1e4ead32193d34fedb7ce2a
a1ae97e079174e5a03c07f38ddd1b82dbb69ed35
1306 F20110318_AACDQO bourret_j_Page_71.txt
02d3be35a4622d8e533773fbd8de710f
2cb675f95d2a5f6334f977d060b53ea1aef30d1b
792 F20110318_AACECK bourret_j_Page_45.txt
ccc652d6c235f25d9045a118d421a060
bc7d803a671811764996f86210894d14a097fb59
1987 F20110318_AACDVM bourret_j_Page_62.txt
94e57c7550e62684ae95bbfcac099571
c25a1795f75ba28a0dbeec923710475b2906e29e
21975 F20110318_AACDQP bourret_j_Page_71.QC.jpg
a4f418ce08c06f471573092570239143
1da2d1e4f5cca73c68d40d7be2de2d2d9e5a2906
2085 F20110318_AACECL bourret_j_Page_48.txt
b8d1807fb0a22b4a8807b72db40701f6
81a1af62880bb8f345878deda5dcd2cf2bbd1a2c
107947 F20110318_AACDVN bourret_j_Page_56.jp2
c220a8cca9c3142abe86dd26d0e930e0
d87e1cd003031aad259dcee28be1e7cf1ae2943c
55630 F20110318_AACDQQ bourret_j_Page_69.pro
68aec6959fb431424c1d86f39d955e26
fd191ffe6e89754a8bdf212b781215f98a85a3ea
2032 F20110318_AACECM bourret_j_Page_58.txt
51684fe958940ff46de9ab7af4131502
5a2613ceb5c3b84ec8a6ee19c0df5c763af096b4
F20110318_AACDVO bourret_j_Page_32.tif
99e69c5f63d87b8eb6503fe1928186a8
bae12235a39807fea4cca6e4935e95cdd3ef2433
46118 F20110318_AACDQR bourret_j_Page_08.jpg
f679247f189b69b7669dc25e287661b6
560d19fbeedcc3fd7d29072a3cc21fd077ff9f05
691 F20110318_AACECN bourret_j_Page_67.txt
6fbcda7cd3e704ea04f337d2d7e8a2cb
9981f6b6eb4d820bfc2c53f5dc3d25988329cc75
110047 F20110318_AACDVP bourret_j_Page_19.jp2
cf5212cf78a3cb5143a8e219ed2f8510
4ba21141d6a93f930b79042db19ea6047e1cfb88
30900 F20110318_AACDQS bourret_j_Page_39.QC.jpg
ed3aa4e06fc0c28929ccda2cfe5bf76b
28010296ac6c8e8bf24f60a90095d6c0c3dc16dd
2315 F20110318_AACECO bourret_j_Page_68.txt
d8999398b1a593569210230d855abb3f
4b830678cd2f1ed9ee36d50368f344922ee3151f
88358 F20110318_AACDVQ bourret_j_Page_60.jpg
911d2e0e59d3638a180b905d3fffb3bc
a7ff422aba3f7297fcf02064f296430dc0a46ace
118400 F20110318_AACDQT bourret_j_Page_69.jp2
b4defe36cdd8aac1d848e4c8514d1dce
8b77f0aa4ecad47fabf020de00d34933edf6d631
47072 F20110318_AACDVR bourret_j_Page_39.pro
5b6868d44858e251d307ef4a43d0e0db
812d0ced52821060e14c5986fc702b937b3e8476
96572 F20110318_AACDQU bourret_j_Page_60.jp2
880f7b331d6bfc01207488f620cc2285
c1b35a23333b11a90c6280de2c2d7e1931278490
1223 F20110318_AACECP bourret_j_Page_70.txt
67e045befa022c0ae9b129a89f6b81eb
80c9f0c2185f6c0e50f6db525602456443fd38a3
47750 F20110318_AACDVS bourret_j_Page_15.pro
c99c175102553d0565132dc39ecea790
61f9f55e18fcc08e5efc3c7ae1043fd00568afb6
44328 F20110318_AACDQV bourret_j_Page_09.pro
3291ac6812d642829246505dd3ba4e09
844a316f87220899e11fabe828996a26acb40a45
14821 F20110318_AACECQ bourret_j_Page_03.pro
03b91fc4ef0696ac39a3c580f9e03b09
f3d52ad643863efdacc52adb78016885db100d97
106702 F20110318_AACDVT bourret_j_Page_40.jpg
821681d3d910825ea7681524224c6870
f4d92f64495ccd32880852847f4e6a0aebc13cc8
8102 F20110318_AACDQW bourret_j_Page_25thm.jpg
177f28d538ee363e0e7f6bc53a12ba9d
7ed720b2797b36b99bd363b00623b8e1161e0504
39516 F20110318_AACECR bourret_j_Page_07.pro
fa8eae99b4ee8d66395e0ac3d235a694
b8822ef0017aefe01a0711edcf27cd1c054d38c8
F20110318_AACDLZ bourret_j_Page_08.tif
4ea14c401641fc9c82c571188f037fd1
e709a8083b44fb85bd9c6bdb107d8831371d0035
28205 F20110318_AACDVU bourret_j_Page_54.jpg
fe340a10b12698193fc0be28b11cd2bb
26e888b99cef160b2584865971cdd192e21eede9
13842 F20110318_AACDQX bourret_j_Page_36.QC.jpg
6dee32e8aab5de6839cf7c7d12cc0c18
3450629b1b89786821b293ec331659a37e7c9e94
48686 F20110318_AACECS bourret_j_Page_10.pro
15d261865e1e851744a4e5392364ac61
3d60fe1bc92f20c7ef227b7ff29f83924a9ffbf6
34067 F20110318_AACDVV bourret_j_Page_48.QC.jpg
12e9dc6d646000fb2aba3d5a0637210d
fe81a0a9f193c65f79121ad4bf5ea27f55b5d3c1
3821 F20110318_AACDQY bourret_j_Page_08thm.jpg
18b8e27b2360cbf16915554a01e65fd1
58467dfcc479a53d19417910c0482864299007de
47528 F20110318_AACECT bourret_j_Page_29.pro
91919f26ce4bccf9bc1ab6927f5c2fe6
65c85ed71edf963694d4e9ab574b49e8c13649a6
25271604 F20110318_AACDVW bourret_j_Page_04.tif
1748bbf5c8a0f5f7a43a4108672e1a17
f9e01e0f86f666a326485f121d3fc040774c221f
1862 F20110318_AACDOA bourret_j_Page_12.txt
17c6657ca564c1ae59b73930dcbc27fd
a60cddb47ea486a902f1dd61f6bcb071140ffb9e
23245 F20110318_AACDQZ bourret_j_Page_13.QC.jpg
61cc3f8d7dc4414393f322c0289fc60e
b4cd5807d4ac56ab87e29ae6f7c54d3c71eedf13
50760 F20110318_AACECU bourret_j_Page_46.pro
2371de07bb343527375506452782a9a5
6f38d14c280cc1fc25fc682195105774e198d92d
30913 F20110318_AACDVX bourret_j_Page_10.QC.jpg
63605b46981b0760f624babf449f8b85
78f66c60141fcab24a31fe89217b1ecc7c6f0cb2
96265 F20110318_AACDOB bourret_j_Page_61.jpg
6a1eb84c863e4b4cf2bd1bd2bcecd0e1
d114477835fe4ab009d56a3ff7de709316db23ef
53157 F20110318_AACECV bourret_j_Page_48.pro
3fcaa1a180cca18a4a518d1c1480c727
c655845d26eb140fa35cfddc1e82c57e4892400b
28780 F20110318_AACDTA bourret_j_Page_60.QC.jpg
0c44aaa15efaaeb3c8175eb9d70d87a5
1874642893a128d4f8b1f523d6336cdb34d19534
108792 F20110318_AACDVY bourret_j_Page_37.jp2
b20a097e2fe981c440b7cd14f844f51b
54bb3b306315c122875cd6e1599be8a426581b48
F20110318_AACDOC bourret_j_Page_10.tif
8f3406aac3492a656297bf37a756c641
53b314999221eca1f92345e18308194d23d35976
22230 F20110318_AACECW bourret_j_Page_49.pro
b7297b8ca9b40782d23291195e959baf
45ea7420b340ded48e57d28c949ba445bc54391c
F20110318_AACDVZ bourret_j_Page_68.tif
bf2139c06e5674d3099db7b15a4c6cbb
5c38e6fdc112ba2b5e46f15b8aec9357c4e7fd4a
1737 F20110318_AACDOD bourret_j_Page_07.txt
bc9d754b519c5d936a5eaa46498a265e
b002131a7e20f885a4a6eec7c06862330b1f640b
42683 F20110318_AACECX bourret_j_Page_51.pro
cb9d9ebc3f2df6d3a4e2cba72aa6dba5
e0e233e5d4dcf440bfd69c1f404c4288cec11f33
F20110318_AACDTB bourret_j_Page_49.tif
fe58ea6ac291426404259a720886125c
f489bfe55de7c77a66f647c35eb87e076d2c2991
F20110318_AACDOE bourret_j_Page_36.tif
38877aa4ccd646e0f489b236f60144dc
af9aa855d031f514bb1899aa26e4b37d9465d920
52729 F20110318_AACECY bourret_j_Page_52.pro
3ed2880a8e1198d638b96080543f0253
63924dc0d3d51678da1474fe1966cc398dd19895
30747 F20110318_AACDOF bourret_j_Page_17.QC.jpg
1207b53c5ea3d6f5c7d93c5f0767bbe7
72b0d270f501d2b6c0ef960d5f89b39f1615efdf
113529 F20110318_AACDYA bourret_j_Page_30.jp2
b561d6b6a2e596c7c0cfa8ca680f70c7
6d38f6751351ca467a705c46336fadc2d879a0e4
1994 F20110318_AACEAA bourret_j_Page_21.txt
0be33979b9931e20c5485fcd94fa28cf
33620c68b9f5c8b618d14f47d0c6bfbf7e76a56b
1973 F20110318_AACDTC bourret_j_Page_11.txt
0ae51a85a8a643a42ce6829eb5b7406a
24218353c8e923e08a78cd30061dac89062716f1
56329 F20110318_AACECZ bourret_j_Page_68.pro
9d8a76fa4164a21a14391cb73f30eebd
434554745d86760d7cae8a016883e1a5ea03fc71
12850 F20110318_AACDOG bourret_j_Page_31.jpg
bc367662aacf3c8d9d77081ec1fdeece
1a08fb6fee4056e1413833ab93d6c2c1fdfc2fba
8366 F20110318_AACDYB bourret_j_Page_21thm.jpg
ef3348717936518cb4e96cea31bd7242
72e47b12d261f67207e825013a83986996ff6adb
43410 F20110318_AACEAB bourret_j_Page_60.pro
244f91607e5ea9db426661f4cda44b2a
c4cb0305d869c1510a1d092f8c522d6399361bc4
1200 F20110318_AACDTD bourret_j_Page_02.pro
27b50ba06032d46f00cf666a1e585644
d30bddd9a3ad78c7acfe051fea3c6d6e34e5efc5
102437 F20110318_AACDOH bourret_j_Page_37.jpg
65ecdb6093bac77954fbd87e90d3dccb
7dd42e038e7b530aae3e8396e36c1cd9e4a0774f
F20110318_AACDYC bourret_j_Page_06.tif
821b1f24a47892b87d5862fc1b93005f
7cfb288ae816ba0b9e74cfc98cd44e75d8a0d870
F20110318_AACEAC bourret_j_Page_51.tif
8d5de51df761908abf6f495cfd462516
8b5b4fb89b31f5cd2f3d4e3bba48990271c5d4a2
28911 F20110318_AACDTE bourret_j_Page_27.pro
40ee111c67a41161621c0cda2e3fdaef
e6289a4bc3fcb65a1a4dc41c7bc071fa957bec56
3656 F20110318_AACEFA bourret_j_Page_49thm.jpg
a13e7d6b7100bcf02bed55d333fb5c56
48fc80c56da43544143a403bd91ef55fdb696dbc
F20110318_AACDOI bourret_j_Page_66.tif
7407ede87530cfa256731414ccc6c496
fecacdae703cdd6ddb9805b5b434c085ce45512e
102411 F20110318_AACDYD bourret_j_Page_61.jp2
7812434b2085037c69a2b4d7e800fa0b
dd43d74335946f687fd609953a30a7f2e4da7aa2
102659 F20110318_AACEAD bourret_j_Page_15.jp2
57fc881ec3399a537a2cafb5a32ba6f1
6e84843371e5c806a9fff9cf972a8c141cf522e3
F20110318_AACDTF bourret_j_Page_14.tif
e1208eec4ca1c0d6bb3ba26b6fe60e36
f7616755d3b8b543f7d9d8082e845096fbecb8c5
2848 F20110318_AACEFB bourret_j_Page_57thm.jpg
89458aaa98aa84902189c5f796fdbf50
1e17aaf001383f004fb437d60b508395d9d47250
11203 F20110318_AACDOJ bourret_j_Page_55.QC.jpg
b6c8e2a6f4eaced010bb87314ed31e48
b68f99bb6d98be41ed4b5425570bec919ad20806
8205 F20110318_AACDYE bourret_j_Page_11thm.jpg
9000cee41e3819e3b817b4204863e705
7b5e42a943f504b80a7eda7a81afd32d9cac776d
2006 F20110318_AACEAE bourret_j_Page_20.txt
7a240dfb45500abaff09520494ef515c
2200ae095867d0adab5b6c23b2c04aaaa6b443c5
1153 F20110318_AACDTG bourret_j_Page_27.txt
0a4e8de3683984cf833d275b37851dd1
d0697ba784d26b61eb87140fec400e5a0ba9fc05
3201 F20110318_AACEFC bourret_j_Page_59thm.jpg
a2720cd558604402aceb616df53870ca
35f56dbced31a8fcd86922cc435f9659f6a10d0e
5458 F20110318_AACDOK bourret_j_Page_71thm.jpg
4dc9b67268034027b5b90e5f57ff9728
d0b4d4ca9cdcf95600f74e3dc6611ddd5cdc53d7
2103 F20110318_AACDYF bourret_j_Page_23.txt
6bb03dce8addf5ec56e8634f5e53ead9
eb4e9eaf70a3695b284886934d568dd8a5ec3e59
103459 F20110318_AACEAF bourret_j_Page_66.jpg
1e36e50534d08da3dd17e37bff5e6bff
8ba2525ded9037c812ee2d4668345acecd504d84
8277 F20110318_AACDTH bourret_j_Page_01.pro
d83e447d6ec2a7a1e5929699387ba3b4
be033638d32ea3b2dcbee680f456fae5ae55492a
7475 F20110318_AACEFD bourret_j_Page_61thm.jpg
a83894321df09220da504e35c5a56e68
66edc420b9d11c68d644c4aaf412f944eeaf67c6
34491 F20110318_AACDOL bourret_j_Page_55.jpg
5df1e46180b18c6b2a7b32a21c70db57
08c185baebbe11c3669f46e7689c49372577ecdf
114188 F20110318_AACEAG bourret_j_Page_23.jp2
97c992f2d9bfae217263802c0d655326
4a9dd9a61979e34e2c68fc6a6d2eac4ba6b768de
2002 F20110318_AACDTI bourret_j_Page_43.txt
a4e10c95c64ac23c03888d2477c17577
1188d0953cdaf27f56c004d2505064a461d6f171
8282 F20110318_AACEFE bourret_j_Page_66thm.jpg
e3f9d0879553e445c755bc93f335dccd
953ebd1a8694fedddd436ae6c6986c5d84f0ab31
23115 F20110318_AACDOM bourret_j_Page_06.QC.jpg
52e5df58bebe8c733f5b4a0e79657442
324f9942ce2e22a534d59731e94a1c08be92a17d
50336 F20110318_AACDYG bourret_j_Page_56.pro
78abcad31beb55c9a90ebcc103f807a3
8e256f60b6d04c0b6d6803ee117703def6e364e5
93706 F20110318_AACEAH bourret_j_Page_28.jp2
c512d734a6f40686c01e5d69930e7407
0826d537f8b327324271d0a4f15307b89479c77f
101537 F20110318_AACDTJ bourret_j_Page_53.jpg
7c1781288eebfe003a6458b64a064b04
da52236df887b752b862dfa7028dd17c4b72418c
7762 F20110318_AACEFF bourret_j_Page_69thm.jpg
a082fdf2f69e2fb7fbf82dfb04d350ac
8234fe57a014adaca8795c73fe7ef92d95fe1ead
4726 F20110318_AACDON bourret_j_Page_27thm.jpg
dd4c48333abd2858d304f91f20216d2e
7fb1b39e0c486c74ead4c091ddc91f69cbbbadb7
116565 F20110318_AACDYH bourret_j_Page_40.jp2
31143b1158d0a29d1dd88b06b03d5ebb
d9dbf2b1dbb96beed2954f3587240bfb89d9bec6
2029 F20110318_AACEAI bourret_j_Page_66.txt
4ac41acb4be3606bbefd2197457fbfd1
cb4dc4eb732a8c280cf196aad67decbb66ae9a7f
622 F20110318_AACDTK bourret_j_Page_02thm.jpg
63e699977f040c2269deb9c33cafdb24
b3f45aba0a84044fa42775db6311d5bcbba6327f
85177 F20110318_AACEFG UFE0012800_00001.mets FULL
ad2e7cf430857ce63d294ac4a6389b16
f92ad9529f598002bdb537e54c11727a905c4403
F20110318_AACDOO bourret_j_Page_61.tif
94d39b407407d4a930b337f6b3ee45ef
6a4e4dd694655b577b0559372bec541129b564e2
95469 F20110318_AACDYI bourret_j_Page_09.jp2
e9f0c31fbf8be57acfc118b0dd617446
5b82c3a0b0ff100aefe6b671e35200aa6df713ab
51698 F20110318_AACEAJ bourret_j_Page_58.pro
08ee97f5a5c1ea5d4b111b4910d38d47
5775584a2aba434cd4536e2ab4d2bb0f732f1c4a
7486 F20110318_AACDTL bourret_j_Page_42thm.jpg
cc4c47dfc21a75c8757bfffbc97433c4
c03eb6b60a839e30212baa02639c0af75e01e04b
49152 F20110318_AACDOP bourret_j_Page_22.pro
ffa7802b3e77b2d4af0f642bbda7abfa
1b7d8ac21289ad8cc88141f6602403ac9ed532ff
34654 F20110318_AACDYJ bourret_j_Page_30.QC.jpg
743651c638172cb47c05f174c0a2fea3
9449396b9b261b1132bc9c7574c59d8dfc8d84e6
2068 F20110318_AACEAK bourret_j_Page_52.txt
07c6e5e6d2b00219470726cbbf599faf
1c5bf1a4ee76b148ab26ea6440aa7bfc9901ff5c
31034 F20110318_AACDTM bourret_j_Page_15.QC.jpg
b63a6c268c20daeabb918f130e7f7a50
c60feffbcdd8e3857fa27dcff7db1f0845fc1a25
7848 F20110318_AACDOQ bourret_j_Page_17thm.jpg
1edcbc1252165a637f10535355cb9c59
0a60e467764b59caa710f114c370eeacfc5f9f8a
1252 F20110318_AACDYK bourret_j_Page_31thm.jpg
5d9d0e1765ba60f2875797f790fa456d
573391a784b4efe9af61fed307b1c07f9559c67d
F20110318_AACEAL bourret_j_Page_41.tif
3ab257f7e7c4bc7e24100ca638f90f4e
c07cfaed5b616f73ec293d222150403923e0d137
41820 F20110318_AACDTN bourret_j_Page_12.pro
b4e1bd9db37cc11566211edbf70fd5af
679b06998f137a352a7f9abd3d253c466aa8ba8e
F20110318_AACDOR bourret_j_Page_11.tif
1b4a4302c1e9bdff7d6d927a26e73f36
9c070b23d07f20f5c2eccac2090860bc6de4f7d7
53497 F20110318_AACDYL bourret_j_Page_23.pro
d15c37a692b546155a0df8e9270268dc
c24e16243696c0b9f39b78ace5af80a331ae90cc
F20110318_AACEAM bourret_j_Page_58.tif
ad75acd817eb1664ffe8f813b36eb069
25a2c8eb62a04173273364243df7e4b813d93c15
100562 F20110318_AACDTO bourret_j_Page_14.jpg
c78dcfbcf39ae1c958960520067c8ff8
e7e3ec3a1ad028162d185b32acb7bc8a64313d31
108468 F20110318_AACDOS bourret_j_Page_26.jp2
c96e37201889c4737b38f433932dd57e
464bbdb5a4efda89753fa27dd77b732c8fbe6fd8
8577 F20110318_AACDYM bourret_j_Page_40thm.jpg
05d1d969156908e6b6b63db70077f5f0
5aba2ae1aaa0f3f4ce7631e99f6fc36e7aa2ca8a
92043 F20110318_AACDTP bourret_j_Page_12.jp2
32ca652765223f600f802435860ce431
2a6333ba6fa30ab37ba2e474a7db08a6af528049
79810 F20110318_AACDOT bourret_j_Page_06.jpg
af225ba99c87433429d12e40d4128e8c
da7675cc9621750468adb0a26d9d136dc42c04c5
F20110318_AACDYN bourret_j_Page_05.tif
0938c9b60f639f8dd169e4992d6438ee
c5a7cfdde1175bd3be07cefadb26e77ab59fb35c
50806 F20110318_AACEAN bourret_j_Page_21.pro
d5625925a4b63db048c8e44865c6aab0
d1b9396c27329b8405a635b63ba7f9f0839ec914
2025 F20110318_AACDTQ bourret_j_Page_15.txt
ad640b38dc78e674f29cd66920a59021
f1886be9953fbcbdc2f080b86b71ae271038553d
105536 F20110318_AACDOU bourret_j_Page_22.jp2
d184a3269cfc44b7c862d0c081b65eb3
643d61b522f6082a01a8a11cd206059701b5e65b
6951 F20110318_AACDYO bourret_j_Page_12thm.jpg
9671a21084c5c6e19a2b60334378acab
20708ea34047e36587221c28a0538ab627cd1774
21612 F20110318_AACEAO bourret_j_Page_08.pro
763a58e182dcb408e963fc9422ec35e4
0ec7b2fa973359f46f12167a6f625759149d8509
45631 F20110318_AACDTR bourret_j_Page_44.jpg
1159d2da392e01874f974baac95c9a30
6291223e6cd622d5293a6b43516db12b7b353f27
4538 F20110318_AACDOV bourret_j_Page_38thm.jpg
a9692f2f34c6baaa0a86c279bbe0742b
12fff3f3f91e7b91fb606a4b6a36eaa5bee10522
120371 F20110318_AACDYP bourret_j_Page_68.jp2
12591ca754df3cbee7564459ad623441
ee7186c17cd7224f07f9c3d8d3fdb24a4b05da1c
F20110318_AACEAP bourret_j_Page_43.tif
a1ae5e97bf446b7fa02bee7eb27da4c5
f8a4301e0e872582099f68657ce67d26bea7d44d
2011 F20110318_AACDTS bourret_j_Page_19.txt
b0c85a02169a118fa71ad474d13b8f57
3e7a375bd9d06703814a8f5b37487bf608e7422b
34207 F20110318_AACDOW bourret_j_Page_40.QC.jpg
5d23c1931ac5c09e7a2b75849c660180
cdcbe5eecba44b91f49c8bbe46e6200d67fd56ec
8349 F20110318_AACDYQ bourret_j_Page_43thm.jpg
6236dac98181984cec7328edfe765c06
5fc4084faac003585baa0e3c373c7b45a75102ae
45377 F20110318_AACEAQ bourret_j_Page_47.jp2
080e640cf226e47f226b86d4a55d1631
3804c2cdabd3350dda3af79461682fc253f73192
4846 F20110318_AACDTT bourret_j_Page_70thm.jpg
653aeae92fe7afb66c38b2dc22b2788d
597f27c31ac6446b2d7b5eecd832c96226571c1e
1581 F20110318_AACDOX bourret_j_Page_02.QC.jpg
841379fd3f0b135c00eae5591fae1c81
4afb841bb67216f25bc19aa30aca7759b7d97245
8236 F20110318_AACDYR bourret_j_Page_46thm.jpg
9ea4a64c198979b31da1d39ce1ab02a0
8de065733928fc09b407aa8d4df2c218de8f940b
F20110318_AACEAR bourret_j_Page_53.tif
012faef27202f8b74a93db83aaeea7a1
d919b9a7cceff64d38ed03ffd3b93b63d6e6112e
34036 F20110318_AACDTU bourret_j_Page_52.QC.jpg
15bce29d8aabd38fd08d83747fc16b0a
66d50cf86a2ec77f07ece55c5a8ca8a3a9182243
285109 F20110318_AACDOY bourret_j.pdf
7a775d97204ce811d1c91013b4a74175
cb32032cdab783c2e9259b6d0f876baf7332d3f0
53308 F20110318_AACDYS bourret_j_Page_30.pro
a8e51aa65a188627ec9cc43e649f74d9
549e2ea17d8cdff66faaf9ab98a4abef4f760cf6
1983 F20110318_AACEAS bourret_j_Page_26.txt
8b3bcda40a7a03a1258c5797eb3110d1
b115f46b990538602d47352dbb44fe17a011d7df
105906 F20110318_AACDTV bourret_j_Page_52.jpg
e0bc320611030221c3eaedd2fb57e1d8
95ebbca43c875fbf4b393ccf365df32df1975094
40203 F20110318_AACDOZ bourret_j_Page_67.jp2
c86ba0187db4141901ac5e1a1528ccac
5ac0dae5b4ddc1bd070c5278cfe64ece44e3f7af
8542 F20110318_AACDYT bourret_j_Page_63thm.jpg
628b16676d6fe0f25b777f4aee11334d
3fb4c3e60de429b8df227bb36294f16038ed5e50
111441 F20110318_AACEAT bourret_j_Page_20.jp2
58d7cf851b41ef09abeee51de0bd24c2
223f80e0faa1685b907afac1f517f8cf3cb703fe
102812 F20110318_AACDTW bourret_j_Page_10.jp2
2d693b338201d6560a73d1f21cd75c37
eb2af513726565dc55ee4ac9b4cfa2b15b01472e
F20110318_AACDMA bourret_j_Page_07.tif
cfc6d850b7c36dc2ce24fb6a79b9f51a
389e28f099e1caff9c613078e131c59e1f702d85
483028 F20110318_AACDYU bourret_j_Page_44.jp2
287250350110211ba267f1b83409acf4
fc6fb0c28aa7d49cd36e1611b81b0af4a91b7237
8469 F20110318_AACEAU bourret_j_Page_20thm.jpg
b153974cbe0b2d5709fe211ba1f0498f
9fca3a363ee8e85312780c63db4333e9e9da99e5
F20110318_AACDTX bourret_j_Page_19.tif
0e13affa34cb9edc3400fb182875490a
b3b69b3203a019402e08bec9a9304d0a6df6566c
41971 F20110318_AACDMB bourret_j_Page_28.pro
de11dfdc8bc6b0ea20894adfcf6cd1e9
f5d4252f587a820dd00f7cbb24dfca26d98b07ee
11478 F20110318_AACDYV bourret_j_Page_32.QC.jpg
d9fd2d3999556223724435160c110a09
a9fd52c7235d282406f1a4299e2c43b291e2b405
33878 F20110318_AACEAV bourret_j_Page_63.QC.jpg
d6ba405e1770aba80410f6b6c13df7c1
442199d4130cdcd5920902618bbdcf54bac3eb9a
F20110318_AACDTY bourret_j_Page_45.tif
00e29f8d8d64f1e329bc175f962973fe
cd21e36b8ac6e8c337070be733ba9fb7e72dadb3
F20110318_AACDMC bourret_j_Page_35.tif
a2fbe8b21cb458e70adc1004a2cb2065
1f0486d835825e19b692a350768855b7f0e24b71
25591 F20110318_AACDYW bourret_j_Page_07.QC.jpg
7d28980945ecf992f0a5dc6a549af32f
bb090beb11f21819229fe7bae5c4a63d5f42eb3e
45571 F20110318_AACEAW bourret_j_Page_34.jp2
f51283eb942884a67daae65da5fbe845
bcdd4e12c56cd5473588e9d5ff10a2da87526cbc
105962 F20110318_AACDRA bourret_j_Page_30.jpg
03a4b069646120d77e42a70c80ba4eae
8505794dee09b6099e7ffb896ebd6e42bfb4488b
105812 F20110318_AACDTZ bourret_j_Page_48.jpg
0b3a48c9cb9743a7c74da6ff4f585244
15c585d117a887e78bca61bd394e8183091511d0
1947 F20110318_AACDMD bourret_j_Page_14.txt
8c1b33373444bd388a1b7ae71c0bcdd3
76a95086cf50e0a1363a513920705e26a3012bc9
112684 F20110318_AACDYX bourret_j_Page_18.jpg
f3cee64db227b821e30ed9f5a2d140cb
e6c0544cde6b35151c523128d569cda4152d0607
F20110318_AACEAX bourret_j_Page_42.tif
93e2307468ba9f5bc6e8ffd135625dda
3512457567a51a569f293dd265a7e2078e108e4b
8132 F20110318_AACDRB bourret_j_Page_15thm.jpg
236d774a19288374de23de3f6cc5074a
ff9360b013c461a6874c161819d950816a9d1c13
36861 F20110318_AACDME bourret_j_Page_18.QC.jpg
c565b2642f5cf5f987db8cc22e781778
01aff6c3e23bc6d04d195ffc0fdb273b84b1bb65
1923 F20110318_AACDYY bourret_j_Page_61.txt
40e45ec4f1668a0344c35d0d53566d9f
0ee8f4e45b50bc253efc04a0614cd4991299c510
7745 F20110318_AACEAY bourret_j_Page_29thm.jpg
eceeaf86f2aa1e45d13e25b6b0212648
4fdd468e8aafed7452135d013f19c5e166cf6eaa
30026 F20110318_AACDRC bourret_j_Page_16.QC.jpg
9b08f9c1a71a07a50a4d94100097f95b
d8bd7580ec34e0a518927bb17e527c8275614e60
1487 F20110318_AACDMF bourret_j_Page_44.txt
708b49605672e48e037748732a1ed21d
34cc645cb5a1ac05869382232fc36dc8c2f11e83
F20110318_AACDWA bourret_j_Page_71.tif
fe2d8b1b2e018ec69bc151f9dcafab7a
0fcb1e4e3a3386ca7becd43e4ddaa4f7cda35c9d
F20110318_AACDYZ bourret_j_Page_62.tif
3692ea5021303d5b39db7653b7a88e0c
82964a93573169394e451d0d383ba97912a33bc8
16389 F20110318_AACEAZ bourret_j_Page_67.pro
75482303ffda7c819a6ab73cbefa0040
609859f790e3afd13c24241fdb751ebc0ae768c0
44019 F20110318_AACDRD bourret_j_Page_50.pro
597d6be40a8ae97efff4c2e60888a85a
4a47201284c95a97433b4984495d8e46bc74e782
F20110318_AACDMG bourret_j_Page_62.pro
6c6f5d00c87f29b5f5336e8161d4c634
3cf1db841636ab6a083f9ce3d22242cd2c79ff71
109087 F20110318_AACDWB bourret_j_Page_11.jp2
7fa5c889fb1eb673eb143d0a2705c9d5
374185709ecc5327f42c9bea3b06ac85458af127
F20110318_AACDRE bourret_j_Page_23.tif
b95054e2980e81d972206d818dcb5338
2cde456d24d43f9a0350b33babe29126b9635a5a
32591 F20110318_AACDMH bourret_j_Page_56.QC.jpg
ff26df95c1e1332b9f1e2dafd87b1c78
e7638ffc26230d3199c6df963fd891b4f9551a69
F20110318_AACDWC bourret_j_Page_48.tif
3f30b75aca90dd5904def0bbd37ba5c3
f8ffc2ff908583e7da04bc7f52bc1b1763e4cd8b
8069 F20110318_AACEDA bourret_j_Page_01.QC.jpg
f7f3f42c9f19a951e52d54b79414ff03
6c85497e6a636e182e2dabc98e42e85f4050649d
2535 F20110318_AACDRF bourret_j_Page_04.txt
65df18e1404540c5f96a9d171362cd99
4fb64b87bdadefe8e4c0282aa5e03c9b105b1298
8201 F20110318_AACDMI bourret_j_Page_35thm.jpg
ac9f0f5558df4c7c89b77101958602a0
c342a83d8232c3f16b6b18821531c3658ae1cf57
1879 F20110318_AACDWD bourret_j_Page_29.txt
5551e468cd5a5dcca07aa318db5401bf
7e0ca01474ab96877b398eeb5556b69d138e9dbb
32241 F20110318_AACEDB bourret_j_Page_11.QC.jpg
cfa10d8a1d0d87ed3c6e8dc545723ff7
76483df1139f3eca14f389512c329d7fe938e59e
8460 F20110318_AACDRG bourret_j_Page_23thm.jpg
e468897de11e718ffc01643d5e3c0678
d5f2396ae9ef7d78bfe5628ae518cb7d6dc35d54
F20110318_AACDMJ bourret_j_Page_54.tif
f64769ad8406d36ea2c81cfcce4dd74c
18b3fe9500a74c61095ddaaa7f8a8d766f50efd8
84376 F20110318_AACEDC bourret_j_Page_12.jpg
9f73a2e3c1c1afa1ef401d8f70a8cea3
195a2e0d20569a089e41475abf2dfff5c42a0795
7928 F20110318_AACDRH bourret_j_Page_22thm.jpg
aca1be55426b4c882849fac6b88562ec
c857e2bb42de78b9fbbaba2767cd10025c0bad52
33741 F20110318_AACDMK bourret_j_Page_66.QC.jpg
c46246c2d5dddb9706b790301273033a
f59752053cd01bf815d83233382910da1461b050
91837 F20110318_AACDWE bourret_j_Page_51.jpg
6591f4b2c47d307390a5d842a42cdacb
4e931cbf13d5ecac06c008da10aa5c5aaf015252
27567 F20110318_AACEDD bourret_j_Page_12.QC.jpg
9ccf06d263a8c799fbd099d7a3f938ef
97daf6c1737abecb951c9810195451fda8567c0b
6712 F20110318_AACDRI bourret_j_Page_41.pro
cce1a013372d59bdf090570c82971559
28f499996d3c079719dd31a4383435973987e7fd
1978 F20110318_AACDML bourret_j_Page_53.txt
7d11fcc9fb750ee1dbbe3777832a713f
0a3de99997d42e082b2046c31feacd29a36ab494
1960 F20110318_AACDWF bourret_j_Page_35.txt
bef47ca6f2a6e8320f9d47b487ecc72b
68f9094c2fb6d7213beb51b7f81604b4ba33aec4
32332 F20110318_AACEDE bourret_j_Page_14.QC.jpg
8126235adadbb4aa20f5843fc3b1cfca
0b0eb6442212a5f357e1a82d03f0c512f4fd5aeb
F20110318_AACDRJ bourret_j_Page_59.tif
a5dda774abb3dad121dac37138fc6d87
aa4cbc31e6ae3bec4c567f9575658a6ea87e35e5
7255 F20110318_AACDMM bourret_j_Page_50thm.jpg
368059a056c75fb46a1d98aa4c501cea
3677bcf8ca99192fc8a64c3b84362ff60be4bc83
33375 F20110318_AACDWG bourret_j_Page_53.QC.jpg
2d4e032eaa6373f2dc23c2b73936fb38
4d9ee6b9eb10d16dc0bbd9b4c2415e2ac097331b
93564 F20110318_AACEDF bourret_j_Page_16.jpg
eb4e06eb78f72f52769a5a09baa25d75
b24c6af86076ce200f803aad3efe05fbf3a050f6
83110 F20110318_AACDRK bourret_j_Page_07.jpg
df9b020b2e53711dd4a3b0019c833361
768311bd0bf68ba28b1eff93e176ebaf24bbcff2
F20110318_AACDMN bourret_j_Page_30.tif
3bc4a92e839a4278d57d355c97025af4
bbdbc064206d12dd6abf51df17926e835f8188e4
1885 F20110318_AACDWH bourret_j_Page_47.txt
12aefe536a7c907c37def11315d6191d
17ac3194e02a3daeb808135ab02c2e3829a58d59
32926 F20110318_AACEDG bourret_j_Page_20.QC.jpg
4466555efb66a9fc4f0573dd5e970086
d9aa9afca1108c4062969e6156d166bff877f617
2018 F20110318_AACDRL bourret_j_Page_65.txt
56d80d9978a15511cbfe3ec38ac1b11d
7d18a4f6a28759d4f91328ac7f4b7692c5fa79e8
14906 F20110318_AACDMO bourret_j_Page_08.QC.jpg
ba2aff06e0b66d9aa102de2564319ad5
26e4541caaf39b0826eb7e05e7f2b2fd2e345948
8566 F20110318_AACDWI bourret_j_Page_52thm.jpg
663af3d7903e3c20300c1718f60eb5e2
d6756902d59689df905ddba3fd7351413928de8b
32393 F20110318_AACEDH bourret_j_Page_21.QC.jpg
4b26fca1f3f7877b2cd59b92fc54223a
f17278548b76c1214ca9242641b4578129450116
6539 F20110318_AACDRM bourret_j_Page_07thm.jpg
6681ab1f8fc39d7d20e4fe1a3465a26f
17201c459d87eed369351f3e86358a05490a7057
F20110318_AACDMP bourret_j_Page_24.tif
990e195a02f5abe8d46d2dc30c636b1d
a808c8e4739d8cb6a7cd9c2d2740e8d1a9010e7a
8345 F20110318_AACDWJ bourret_j_Page_19thm.jpg
6a8e5d2106e9c537b6ca498e6aa7976c
43f1a58e7620175d2c526afb032b8b9f51ace7af
105731 F20110318_AACEDI bourret_j_Page_23.jpg
1f88e99555a707f6010a0f0c965878a1
496c082c2f63c5052f5ad74a9bab54d3a291aefe
33498 F20110318_AACDRN bourret_j_Page_37.QC.jpg
1d972521fc03f04fda3cc2c06b99a406
2cc7614da4b6936adcc53c37658693688b59ced8
101615 F20110318_AACDMQ bourret_j_Page_25.jpg
182589a94bb65e13c08317ca7fe3bd9a
cf9579d087bcdddcfffa1bae1f913c1553cfa7fc
14989 F20110318_AACDWK bourret_j_Page_34.QC.jpg
ed1a19aeb6686a22313dbbba62213a72
6bdc6b9bc0e3fee6e8719311507a38a27722c9f8
33852 F20110318_AACEDJ bourret_j_Page_25.QC.jpg
47e5fd219775abb2ea9a59547d485461
52f166d5c1925a6bc1e20299ff4be5a386616515
23510 F20110318_AACDRO bourret_j_Page_34.pro
07af9be30100b3501f66006d40789aa7
7df1d4dfcd5dfb281a0d4af64aa8b88b33579024
71805 F20110318_AACDMR bourret_j_Page_13.jpg
c38d27b66a91f056a88a7d53499c0e94
84b767526c50ac98a95dd4460567aa246f6d144c
6115 F20110318_AACDWL bourret_j_Page_13thm.jpg
08d977a5fd23b944ef11b5a544232b2f
756bc98f0d52a9cb08a8c740a87920f278898488
60027 F20110318_AACEDK bourret_j_Page_27.jpg
9b86173374ff5f69c9c4269ec9bf8608
f2dc1a149fa4803548ef20366a213f0aabb206fa
31938 F20110318_AACDRP bourret_j_Page_22.QC.jpg
18d9e4fb827709389d4895810f14e558
0539cc2b0367095700ea12cb2f30a040bedc8343
51328 F20110318_AACDMS bourret_j_Page_37.pro
29fd65ca6b72c19dfbc02d09635527af
e83934dcc66096957dcfb75c3eb8dfc146e39911
34785 F20110318_AACDWM bourret_j_Page_55.jp2
73a1abb9671e53c6984f00520cba1e98
056522d5a5757716db384814d59e6675c58a4d56
95951 F20110318_AACEDL bourret_j_Page_29.jpg
d207ebe176287484b601287d525917ef
7561cca1b249e603e97a827737428013b2d06658
105835 F20110318_AACDRQ bourret_j_Page_04.jpg
e04896c93d16703a82e7479cda29f95f
f57c4e5209dd6d834addef9197c06cd6b38251b7
97307 F20110318_AACDMT bourret_j_Page_10.jpg
0b259bed9e3c974cc5cddd0db24894a6
a703d8016fe1d9024abe672b78a0e7ed5e76a135
F20110318_AACDWN bourret_j_Page_38.tif
44921c563e6516f118bf210d3ff01e34
14b955ae3de6db262678d7a3b0eff6208375548c
31401 F20110318_AACEDM bourret_j_Page_29.QC.jpg
2e2812d0bcdd683ff0dfe64f49c7bd1a
ba109d54875d6a5aa3fc6bd32b3c2f85eec961b1
9569 F20110318_AACDRR bourret_j_Page_54.pro
ddecd369add14ebdd9aa3a64804b2668
566325e4a52fed2547679e3f71aeff369265953d
112216 F20110318_AACDMU bourret_j_Page_24.jp2
9cdb1ff8bb9b7ef2c74fe57d43e5ed18
e80154e8ffe6de7bd135984cbaa12cc9c65a771f
4681 F20110318_AACDWO bourret_j_Page_02.jpg
75910ba096af87f23ee9f74b1e050f22
56c94064abbb648903969095cc23e03d2f8120e9
36458 F20110318_AACEDN bourret_j_Page_32.jpg
aa968219aaed6edc74d9240b73fbad1e
c7a2177b7d84904d3cef7eed74cf41668fd504f7
1778 F20110318_AACDRS bourret_j_Page_51.txt
f5961179490ae7d42e6a6be316a39ff8
272450468bfe69d498260b73b780019e4bb29f44
3963 F20110318_AACDMV bourret_j_Page_47thm.jpg
fa840eb6dda849d5cbc521df93f069b5
683028d5ee9631a37bf11cf3ac79d2538e6dc31a
28533 F20110318_AACDWP bourret_j_Page_57.jp2
35ab373e0cc1ba51d8d29c398c58dee6
12393df4a70341d0501f6f30a7a37c1c5aba1695
32165 F20110318_AACEDO bourret_j_Page_33.QC.jpg
ec5a8b79f3221059f36ed43b2bec2b70
7afc30872e00e3e2ad405f54cdf0f9d58a8fbda8
F20110318_AACDRT bourret_j_Page_55.tif
bdb84d045e5ae766942a3e511a201a1f
749ae8e61a355e45b1228214262e1ff2483cd69b
8222 F20110318_AACDMW bourret_j_Page_33thm.jpg
cd7f13032a0e944b2e36db3fd9868e9e
6d414b895476c88a37c8517b56dba3fadbac1a13
4794 F20110318_AACDWQ bourret_j_Page_31.QC.jpg
1e1fe9e151eb1e195cc3f1c82689d421
243204d54ed7416c04580b584168ed947ce16ad5
43269 F20110318_AACEDP bourret_j_Page_36.jpg
345307c42f71e0ad2ae55a2d23d900a3
559b9d10276268de5bd72f05e91f1acf09266032
100838 F20110318_AACDRU bourret_j_Page_16.jp2
13da8c99fc91a7721801ef5ab0c12270
67bd7128b3e950eea85f98369a8ecf1f6bcc245a
2060 F20110318_AACDMX bourret_j_Page_40.txt
2a71b3b5b4a65725a853f38d877a2d18
8b696808127caf0c9b03319a545b4f4f4dda27ab
1051964 F20110318_AACDWR bourret_j_Page_06.jp2
d5a478363562db19539a12d72e47f7ba
d852de8d9e5041b8ee7c97ec52315efddca4650c
1141 F20110318_AACDRV bourret_j_Page_36.txt
984103dfdd8bd91147f4a5a499b36a30
2bb3dbe59ec420198b701f1772f26316eec417af
13251 F20110318_AACDMY bourret_j_Page_44.QC.jpg
9e74df8b2c2bf39cc183d3a9b71b0808
3ecbd78b32b2c2757ccd8e5783887f620948c93a
1105 F20110318_AACDWS bourret_j_Page_38.txt
00ea4ba48512c856b4820337c0ae2f31
a6a5a789dbd09e91b48ed239d947ed606ee884c7
47155 F20110318_AACEDQ bourret_j_Page_38.jpg
ee8a29dc89191bdbb9f76a5d38f002f5
df3d1a600280f92d663d2631610768fd3b0294b5
8549 F20110318_AACDRW bourret_j_Page_48thm.jpg
b332f450faba26f9df534206361893d3
75710ee964348ab5e52f61431681e19f4314de35
F20110318_AACDMZ bourret_j_Page_67.tif
99136acd5c4cf1b3326f7a537b9bb5c0
441ec4fe412b52c92a3e6553cd4c0c308b37e708
30088 F20110318_AACDWT bourret_j_Page_47.pro
ce1cc5fc4a81eb25c4748b62412f7941
30673c04748f77cb287c077663bea1e8068e95fa
93902 F20110318_AACEDR bourret_j_Page_42.jpg
68b945668ba50dd65f454b0e802be3b8
8f7884ccdaa8f36502fa9837e2d0d9894a1c7942
101376 F20110318_AACDRX bourret_j_Page_11.jpg
da2dfe29dbbbb26a12c1e8ba5905d266
92eb18d6730f0f370f4b3c790a0a7351b8ae0ccc
8174 F20110318_AACDWU bourret_j_Page_53thm.jpg
539d9b4aee4f1caaa2beeed2ead1faa2
97b6671310ff6a49e442ecc9b9a6ee0767f6eddd
102619 F20110318_AACEDS bourret_j_Page_46.jpg
cecf694585cabacd78601593947fde8d
c38f01762ed620abfa8e211c8e03e2e0029e8304
F20110318_AACDRY bourret_j_Page_40.tif
50f16f947af69cf9784c8f782e49f64b
cb736ad556811b2e6b1bd16858d07491c039ec25
49949 F20110318_AACDWV bourret_j_Page_43.pro
249836ee572003d3902cc8e3f97f4e3d
9e7810a97fb8fb9628c836697ec829c5a33262fa
46681 F20110318_AACEDT bourret_j_Page_47.jpg
e0602a888d20608d8857963230a21642
74146e9eeba44561316689e7f9f72174127c8ac7
49586 F20110318_AACDRZ bourret_j_Page_53.pro
19ee64a8040142eb38d8027920da6f0a
c7a1b25b4087d40fbaf1c1ab4a78b26a2c929a7d
F20110318_AACDWW bourret_j_Page_65.tif
15b69d2b717eb4b76788e60a49f425a2
233147fc942e8f35bda9d4ad83050e421cc9117c
96970 F20110318_AACDPA bourret_j_Page_15.jpg
127af5829c14e69b2f881478aa04242b
fed2034139ac1239b758ddb5ba250647df64cebd
14056 F20110318_AACEDU bourret_j_Page_47.QC.jpg
ac087f03c059f4416a68aaaf28f51550
62de22444b181c6e8bc2e88e6f14954ff8f89b9f
7466 F20110318_AACDWX bourret_j_Page_51thm.jpg
d71656151105c41bebea0663e926c46d
3fefaa660d16e332a74539288dfe1d7be097cf9c
87339 F20110318_AACDPB bourret_j_Page_07.jp2
29f65b47ac4ca6b610c2d7bd04f51ece
c236e902f7f9fdcb286750e1589a14bd1e569793
40165 F20110318_AACEDV bourret_j_Page_49.jpg
86f04ce5fedc34dd4e8e5336a5192d1c
f0d65b870aa617a3eba41fd376fe0d1d756efab2
94818 F20110318_AACDUA bourret_j_Page_51.jp2
f267c850613b7c1b9f040981b34860e9
1d00ad1c2103ef042f47bdb64ee8a75a71604d51
11867 F20110318_AACDWY bourret_j_Page_45.QC.jpg
ea7da66601c79a64337e356ba61e3caf
8d14c5b6c06483c4c05e21ad83298c1e2195ea94
100366 F20110318_AACDPC bourret_j_Page_43.jpg
f13bfda02b104755fb801b3119f0b6ea
ca6c72fe6720bea9c81791b13d0567a4e11ac567
89703 F20110318_AACEDW bourret_j_Page_50.jpg
246d1ce7e3bf4431ad107eed3c8a548d
8d5e3d5d369043c69c4e1432128a2a128bc4aaa1
108416 F20110318_AACDUB bourret_j_Page_14.jp2
e4a5953d3a2962c08f15c461f57fd7fd
d71623c700f78a20fda6b6ca955feb2255302382
1857 F20110318_AACDWZ bourret_j_Page_42.txt
8d5ff47c57c30088272308ef435ed277
0e11dbfb7932bc0654f78e661f3947c34f729524
49236 F20110318_AACDPD bourret_j_Page_14.pro
b9778a1289218c3bd47c82a6dd4a7ac0
2f3e2a098be13ee2b37f6fe64caf77c099dd9c5e
29722 F20110318_AACEDX bourret_j_Page_50.QC.jpg
15d94c37c2212078cadff3d248780af6
ed28b60a6cfb0289ce1ec1ee3efe7424414a19cd
14855 F20110318_AACDPE bourret_j_Page_38.QC.jpg
dcad7e141d5ef8e537277cd9bca789a7
1ac84cb32fe39b870c4f571e1ffc44971ff985a9
28355 F20110318_AACEDY bourret_j_Page_51.QC.jpg
8e163168894a1803e30d415f96c7adc2
62048a2039255f68f9bbeb625403e1b0dc7f51ea
64976 F20110318_AACEBA bourret_j_Page_27.jp2
92e669ea3a1ac141da5da2ec004d1cd0
3cff8bf9412675eac7ffeb8506e92b3f70266790
8053 F20110318_AACDUC bourret_j_Page_68thm.jpg
50b77eb02568c07cc8c76c7b6749bddb
fcd5fbb6ae42a8e18876231b0053124bfee592ef
26268 F20110318_AACDPF bourret_j_Page_04.QC.jpg
3e5a27c8e4fa8165092dc9858bf5f048
10fe67277aeaf9911c1f3d1128dde6e02e4cd517
105288 F20110318_AACDZA bourret_j_Page_33.jp2
ca7f1563e8a5089c4ee6d75aa43af6f7
541de846c7a848a82a414364f531e7cf6049b573
101117 F20110318_AACEDZ bourret_j_Page_56.jpg
bd9dceb7d17507415649d375fd5cb3c0
c2aa4a29d1d6aeaebe6824156c30ae3a0c9b5c6d
2933 F20110318_AACEBB bourret_j_Page_54thm.jpg
005dcbaf5317b0e8d81d4de86a9610ff
f9661148ba9468bb5e9f028d58e3fc09d46ebada
5538 F20110318_AACDUD bourret_j_Page_02.jp2
36e56b78f507b699854b6596447e085c
3585b7fe926e13f719021b4e95cdc07b92dc9ffb
49490 F20110318_AACDPG bourret_j_Page_35.pro
33cb8c977b68cb7a5762c9a602c006b5
169322910fd88fdd6e94f91c0ec78b4c8ca63f4b
8420 F20110318_AACDZB bourret_j_Page_24thm.jpg
e3ac1066bf7c869c0bf11584b1a557c3
a88d140112820d6fc92a62c50a95b6406a277e03
34616 F20110318_AACDUE bourret_j_Page_23.QC.jpg
fd6e89e9e4a4905b18c891e49784d35c
fd5249bd7e098eb87f6545e7144e41ab48081ac9
F20110318_AACDPH bourret_j_Page_15.tif
949ffee3caa8251b50f95dd6570d36d0
18bab42e28782f0b908eaa8bdeea260ad9c33111
104471 F20110318_AACDZC bourret_j_Page_64.jpg
82a0f14657bf73cb897cfa52a974a617
62fe4abf9adba76cdf37e74f180383c5971d80cb
F20110318_AACEBC bourret_j_Page_64.tif
1a6b8238759cee8d2d30777ea5319d59
9774121570dc79129d7719f5d6a4f62da62b51ba
F20110318_AACDUF bourret_j_Page_27.tif
2cd8bda6215497eaf75edc9bb2c4b6d3
dcfda93eaae3a75a91f07b1cca5bf6608b424073
2036 F20110318_AACDPI bourret_j_Page_25.txt
04f72535bf98fcbd05da0b054f053ed0
fc8a1e2a5b0ebbde874e54223d313d832e1d8bfa
33197 F20110318_AACDZD bourret_j_Page_58.QC.jpg
89a327226b5aeb0c60b54c73cca25047
c876faa66942fd7637a35f000d98c604d31e5e6e
1722 F20110318_AACEBD bourret_j_Page_06.txt
6277315c5a028d71ecc96be082db283f
e2dd306d67671304c22243a50eff29f121b9c557
F20110318_AACDUG bourret_j_Page_52.tif
7a286b207e9ece926b9d072b38742bf6
a5ca2f97454f8652524d57e38e2a89b8a06b94ea
96739 F20110318_AACDPJ bourret_j_Page_17.jpg
b1a36a12767324543bdf53ff1c7b2bdd
28febf11cf256313087a8ea47c34c8daee7daf9e
19751 F20110318_AACDZE bourret_j_Page_44.pro
11ecbd5eb0090c2309e4434fcfd946e1
e14cc23cd5ee0cc0dc17c288cc0e7cd5fa4bb390
8557 F20110318_AACEBE bourret_j_Page_37thm.jpg
979374d4667f000a831a148533b0cd38
d323ac64fb517bd56d02bfd962176701cc1c364e
32665 F20110318_AACDUH bourret_j_Page_59.jpg
3e870080d8a9e4e7c18e84fb2d797448
1710465fabde9be8459cc800706d338a6b8bdce3
89820 F20110318_AACDPK bourret_j_Page_28.jpg
e6fe4f7783f9bda9f48cc99a4e8a5470
df5b31d308fb13471f6826d3538d9486e42c6fd7
102707 F20110318_AACDZF bourret_j_Page_19.jpg
ee47fcae3b1fa11208fd5024c6c2f08c
4b0c6eb1872bdce09e3251315b724d1568ef6db8
7121 F20110318_AACEBF bourret_j_Page_09thm.jpg
b81cb4a96eda24ccf612bd64e438e02e
cbdc21aa3246ce07bc314e1fd58a3df813708986
F20110318_AACDUI bourret_j_Page_22.tif
4c6f70a38b5a54c6e5053aba90a3ef52
6e174ab5681aac7f085d46e5de0ac546c508a942
1746 F20110318_AACDPL bourret_j_Page_50.txt
3680afad9b3f93cf75fc76d18718e608
c6334dc5db698ed0b3d5c415526aa4a30537ee98
F20110318_AACDZG bourret_j_Page_17.txt
e72c61dbfa86ca5295789c78c1e4fd6c
cdf084d7bbb2842bf30bf15086d0974ebd832c9d
33344 F20110318_AACEBG bourret_j_Page_43.QC.jpg
ba1df3c01e2880988a22ab242b3dc4d9
9d12eebae4cb853a9a18af44d39d723096eaccf0
F20110318_AACDUJ bourret_j_Page_26.tif
c03af9ea8865d751bb4eb8c96e084137
55cfb15d74de4381fb10407852e36f5580ad875e
48853 F20110318_AACDPM bourret_j_Page_17.pro
80c43f9ee57428ceab6e72c5d5bec22f
a19a2ec84d894df2de500d95629c740deac63962
2960 F20110318_AACEBH bourret_j_Page_03thm.jpg
590601f5ab59220ce0fcc0180427f1b2
2bc2ae3f9da733a5dddf98a804d2ed32724bdf84
44761 F20110318_AACDUK bourret_j_Page_34.jpg
f11fe8bb5e5a294fa43af3cdc351214e
8af614429145872b4cef927ca0019a0b3429480f
F20110318_AACDPN bourret_j_Page_34.tif
d9356533707f84fbb1b5dfea3e991c02
aa942b8e3576c6eca570c1fea41377c213cfbe80
43691 F20110318_AACDZH bourret_j_Page_36.jp2
eac5854e8eef9fdc75b78c558b9c9d39
09035be2c15527b55e448b22f2f97f9b7f9b28f0
2063 F20110318_AACEBI bourret_j_Page_63.txt
a4a18937a6bc9137d9ac00b6b164b266
d8959360bdb7b0114adf929f4263eba341a2e57a
101209 F20110318_AACDUL bourret_j_Page_21.jpg
9d98f851f82a5a7cef9d5c9d0065f3c0
36c7be7e77a2ef6c7b9d4f2ee4ebff9df98ebeea
108686 F20110318_AACDPO bourret_j_Page_62.jp2
4965c6c220207f1673bf06af1bd6b076
5cbc87cc91dfc81a1bdc70ece2797c62853f2649
35360 F20110318_AACDZI bourret_j_Page_03.jp2
1857472a23ce84dd6611fb0b83c9a1b2
dc31a75f570439e84ff319155195b0d41882c57e
116440 F20110318_AACEBJ UFE0012800_00001.xml
70a0ec37e46eb968159ec4afda7d94fa
b20df671d861cd649feaadd944ccb6699304fa20
F20110318_AACDUM bourret_j_Page_17.tif
a5a0159db00ac7af0ac39d8703db11e9
64b70db2d85dfd9175b9bcb8726d778c20adf60a
31971 F20110318_AACDPP bourret_j_Page_26.QC.jpg
dedb08a65f5b81ba0a9717f572370a1f
06e3b586cb2122080c1c040e7b584f87751fcd44
19473 F20110318_AACDZJ bourret_j_Page_27.QC.jpg
1cf6b75f78952dadb4fe221b8a4b3eb9
72a2b12df163674b41a9c653b6382d8ebaac13ad
63187 F20110318_AACDUN bourret_j_Page_04.pro
22d0ee0d7b1a7926068c425638781ba3
03ee4b53c511f010ab3c71fd4f278556d41cbc6f
922 F20110318_AACDPQ bourret_j_Page_55.txt
9a91fa299555ccb4160470b3a9716dd5
5dc0af0e0d6cd56212cc9eb991fa3a7607c9bde6
7685 F20110318_AACDZK bourret_j_Page_39thm.jpg
be8a680a84e8e5794d88f9ff4871f24e
6a79ce7945cd7c5d68db56be4604cfabcb3eb023
F20110318_AACEBM bourret_j_Page_02.tif
373c2e51787018fd0ba8cee06ccdf912
d1f21eeb0aac460e5e00c215076bc48e5912fa77
8202 F20110318_AACDUO bourret_j_Page_62thm.jpg
7c2e772d5073bfda305259eb70af13ed
ee40074530e412969e71b46dcb671f6185a02b04
2051 F20110318_AACDPR bourret_j_Page_64.txt
a73cf8ce364e18eb49f1ede967900c49
c8b346b4da1d6da9bf7645ac65db23e780baa125
33818 F20110318_AACDZL bourret_j_Page_03.jpg
0b26496805dc6e6a5d18d53ab3f0d475
5853302691df86c84c77b226310ebcda14ed9f3a
F20110318_AACEBN bourret_j_Page_09.tif
52bbec680d9ad47097a356124633cf1f
32ef34db3faf6dc560d084877a775ac3869a47eb
113071 F20110318_AACDUP bourret_j_Page_63.jp2
19a6d59b35e8ca46f8cb717d5a3fb044
2c9ccad7b12ace979c46faf71c5b0b36f371b1ac
F20110318_AACDPS bourret_j_Page_18.tif
034664819ff71dd5ff276a1bcb0f66d8
d78042ab78353b5aa799177c2e190930b803552a
13915 F20110318_AACDZM bourret_j_Page_45.pro
6b55138480f7810a9b385bf3b438633b
0efbbbc379ef57629f7d8c12177b131e25271acf
644 F20110318_AACDUQ bourret_j_Page_03.txt
20d35074282f1defc8acef6236e5a5b0
dd95a1b4e5302d144a8d3c998355a09b9680b5c8
46081 F20110318_AACDPT bourret_j_Page_16.pro
368a43d89e86ca3cd723643bca3426ff
88ab59d36c86d85d70a76b5da4b5b1f69926af1b
F20110318_AACDZN bourret_j_Page_03.tif
7ee64dc240f49664d04eb210ba08a52e
4f59f090ce1360752cf880d8d48b9799bb5153ba
F20110318_AACEBO bourret_j_Page_12.tif
be235e7ee6cf9d90b7480d07ca4bba88
2456bd2a15b9c7a91b02e2c64d01d7e1a51cfef7
8281 F20110318_AACDUR bourret_j_Page_58thm.jpg
424018d4015afd963431c618cdc5b8f4
7ebdf2b858b40624dcea34dc45f877fa200f885d
33486 F20110318_AACDPU bourret_j_Page_24.QC.jpg
a1b301085c7be9f38cfc0b46dc797c9f
c0a38a71b11c36af91091aece1bc3e2b248c6921
10541 F20110318_AACDZO bourret_j_Page_59.QC.jpg
d38f724857b3069bee8b942785423690
ece5dc9b85007d29f87aa789495e92618c4be33f
F20110318_AACEBP bourret_j_Page_28.tif
40286ebd60301300487ae84c0e031693
5a53e828ac033d086cfc8f1953f53ca441c6c875
44518 F20110318_AACDUS bourret_j_Page_42.pro
2d27e03d31531800a69510af86f79d53
24ad831860ae98a60822da06d64148eb2d4499c8
51482 F20110318_AACDPV bourret_j_Page_24.pro
4aa4a293531927fbe9eecd50afd7b2da
050c11b8c732a93058d86e29156c9630ef5c3161
37074 F20110318_AACDZP bourret_j_Page_67.jpg
a5589b40eab5206daed900eec8a265e9
2bc2af7225fe12d78c0e79738e55905aea6197bc
F20110318_AACEBQ bourret_j_Page_29.tif
5be48031d9aff2001c80b5ac63a9310e
e2ad182291a83b19a58802c1e90d8f58fbb4f953
19665 F20110318_AACDUT bourret_j_Page_05.jpg
9f37455ebfda8cf035d77229fb09fe1f
ee190a27313595e079629f960e17ef9a7ccc92c5
108980 F20110318_AACDPW bourret_j_Page_68.jpg
0793ae052631d9d09e9fad3e817f899e
f5c060ceed1e50a7dfdd378e45a97e5921bc43ee
102502 F20110318_AACDZQ bourret_j_Page_20.jpg
800dff993cb1b281d2bdef47449e2dd0
56912b0545be67ccef8f21f67f08c17ba44f108e
F20110318_AACEBR bourret_j_Page_33.tif
117aad9fcfc1c8dbba9fe67a196d5ed0
c542b7a1e55f7ce968de3200ec3dd7eb10ef2df8
234 F20110318_AACDUU bourret_j_Page_31.txt
96cd71375047d61b9d17db9ef335d769
0ddbce1ba83ff72cf6b89a1b199e1860fc7d9ec6
5412 F20110318_AACDPX bourret_j_Page_05.QC.jpg
49ffb49a9752bc7e2971ad345f6df6ba
31f36c24bd69ecbe7c9dde3fefb814903954a93b
484 F20110318_AACDZR bourret_j_Page_54.txt
53756fd91ceb99f76a97c77b8700aaba
071bb82929e1aba01c8aad6bc9ab74cede4bf954
F20110318_AACEBS bourret_j_Page_39.tif
3cbac868a19ba71e19e19e7d45bc0e25
fc5a89191cecdd8708c5252036476af20de2c9f2
F20110318_AACDUV bourret_j_Page_25.tif
5733407c3eeefde5eca2578cb61bb055
d687c974b056bc697437fd5cf2af629c8fc04f81
F20110318_AACDPY bourret_j_Page_46.tif
8faa0b2b7d51e416fe8507697f361fee
045ec5dc9bee6a5f1d8bee0a968aed6b94959995
32189 F20110318_AACDZS bourret_j_Page_62.QC.jpg
4c50caa39d782116bc10754ecfc6f08f
147069a7923709cd48bee2c0e1f256eb3e80c823
F20110318_AACEBT bourret_j_Page_44.tif
0379154f7a3fbb9ae6118b85c337ad51
34dfdd766adaacfcc0b31be21e05778c4c300666
F20110318_AACDUW bourret_j_Page_13.tif
9855f3fb9d8ba177e31845f73511853f
c4a76201c1874a0d431051ceb0e5de585c78d3be
37520 F20110318_AACDNA bourret_j_Page_45.jpg
47241492a592e8d06fcee549e58fd567
40dcd1893fadb5df325f675e486e0bbccfb1ecfe
1997 F20110318_AACDPZ bourret_j_Page_46.txt
3eb4f94df0a053170ffa7e41f796d2a0
db8a03c7d6130d844714289eac0d3e86f4e80850
13316 F20110318_AACDZT bourret_j_Page_31.jp2
81a3e623402807fce54d25ff4bc7f44b
dd92e1d78d3a72b76a02c36455a5db3bdd836468
F20110318_AACEBU bourret_j_Page_47.tif
445475f6dcffa902baba201fe32d328f
c05b521768410f47f794bf011906d1ce4b1e5007
52134 F20110318_AACDNB bourret_j_Page_64.pro
6ed8a4b2169a438b7c1fd089382a9adb
f02cd789445e245b2ba3b55d3aed6ff77797f4f6
12821 F20110318_AACDZU bourret_j_Page_05.pro
a87fc742da0d573934a834757f47ab07
c2b583fb2b08235a2f7894c1bc8be248919d2699
91028 F20110318_AACDUX bourret_j_Page_09.jpg
fde49566f103152f462ea966c0724317
9172a554078a19d2c37dc50a3a49070bc4ec9388
F20110318_AACEBV bourret_j_Page_63.tif
cf023148f96bbce41a96609f54299e2a
66fcd0adc9e987d3b46db866f14c0ab4e9ea41c4
2028 F20110318_AACDNC bourret_j_Page_37.txt
e031b9bed55a11d344c1d461acaff184
46b67e64e28dfe898c2a68bbff1cf4fd45b58ae3
12120 F20110318_AACDZV bourret_j_Page_49.QC.jpg
96178d9a2e2db2dce3d43174aff8fccf
5877cfe47e1d083aed012c570516b26778d7b91d
4214 F20110318_AACDUY bourret_j_Page_36thm.jpg
9e6cad04838a0aba9dfc0f0b02f2d9d2
ee7fece9a53e3c37bf50014ed657766048d89729
F20110318_AACEBW bourret_j_Page_70.tif
5adc72e61b98dbe95ce797ba24057dfb
3f084a509316be4cabac5f8b20c7b2d3f9fb309c
8360 F20110318_AACDND bourret_j_Page_30thm.jpg
32740fc7a2b65a19fc15a8de45146f38
636b9c80dcb5d76900fb583b97212460fde1b286
500 F20110318_AACDZW bourret_j_Page_57.txt
5596b2e01241a4413ea152e9371fb208
efb899031c034978e48618775da09767f0255233
26982 F20110318_AACDSA bourret_j_Page_57.jpg
2fc5a513dac92834b82885f1dd6221c0
dfe625c47f81095a2a40ba68cd007db8f658cf8d
21265 F20110318_AACDUZ bourret_j_Page_38.pro
2cd808229556dadef89f8f78c874da83
cdb1f7438142de275a20a0a4d637e962ac3a2a28
466 F20110318_AACEBX bourret_j_Page_01.txt
441a65519d91fad009ef9beebb5f721d
6606f4a834f0b1a729eb139b093009b8794bfef0
118665 F20110318_AACDNE bourret_j_Page_18.jp2
24051e0417cb56dd6ba91276c3d08143
b13ce6daa6d108f9af90a758dd0dead01225fb0d
11940 F20110318_AACDZX bourret_j_Page_59.pro
65e79aef1d20f64b39919a37ee4acb50
726248030be11c8592235aa25f43051e86e09b81
7974 F20110318_AACDSB bourret_j_Page_10thm.jpg
c3977733018ad06f8a3e50e52bcb0ff8
328fd15f7d43041429a5aaff93533b3736c8c857
113 F20110318_AACEBY bourret_j_Page_02.txt
619f923d309d3704cc97494129efb0c4
d311ad39f2e75117f115a202e980d44940ea6f4c
100232 F20110318_AACDNF bourret_j_Page_33.jpg
d76e0d81969154e28f6fdf090307ae45
b441f36598a9e8afe93a185b1e61ebf84166d864
71799 F20110318_AACDXA bourret_j_Page_71.jp2
6d41b3ece8069cc2b74842caef8827e9
47c7912ec320d5a2a0d346d4ab34d2ed8754c785
51049 F20110318_AACDZY bourret_j_Page_19.pro
f3c407527b3867bf4ca0c1aad1f6b22b
94b6f56623462be35b892f01ab97a4ac445ba087
31723 F20110318_AACDSC bourret_j_Page_71.pro
2c31f5bdd35ad954bacc5bd4cc528340
28c8672e55d590fd075b7abb8df9a257cb70d39f
650 F20110318_AACEBZ bourret_j_Page_05.txt
6cc53d7c5d4de9ec9e31f2a6aeba5c50
36e66f8f743bc5947e4417cc81992bf74d7e52e7
8422 F20110318_AACDNG bourret_j_Page_64thm.jpg
c36e25a4c6a08b0b841a3096e420a04d
d50f965e1216227c2db7eb9665f45c27210b33fd
101150 F20110318_AACDXB bourret_j_Page_62.jpg
f69d98c2c8e0b1a108a02b9a19f0f9ba
406b7af7fe7bb824b342647e4dacb3ae4ab9800b
2288 F20110318_AACDZZ bourret_j_Page_01thm.jpg
445000c0402ae2c869ea375ab355a2f7
86bc36d312b99d401c7b5d39cb9feda1f091ce72
8967 F20110318_AACDSD bourret_j_Page_54.QC.jpg
e3ef2dae8cd8b34268b6e8197f538379
7921532792e89514271b81431a6a63ff3c76ea05
485621 F20110318_AACDNH bourret_j_Page_38.jp2
8499817838837e823700c15d5a3691a8
8150e7c6cc2c89a4982ba39584277f4922e38332
F20110318_AACDXC bourret_j_Page_69.tif
858a3204fc436fd6a624a50f59f4c4d1
13d2a6f33f71060d1450d75af54c4e3066a1e71b
3799 F20110318_AACDSE bourret_j_Page_45thm.jpg
d98c4d81ce1a108d1ab32750aee379a4
a4c7fb2eb20e3cfa6b4b1d49113347e4f928f1e7
104065 F20110318_AACEEA bourret_j_Page_63.jpg
ec17ed5737cdd6d54d0a2956a4f5c8fe
340cf12d50d7c4932b7de7d4718408403ecac473
24805 F20110318_AACDNI bourret_j_Page_01.jp2
184f8489b514ef9fe7b8d2a528e2a13e
257a34091e65768d78f710e1b850c42ab553fd72
33302 F20110318_AACDXD bourret_j_Page_19.QC.jpg
624db45a90c0bc4ccc9213db681e68e8
bbf1e5555fc115af74082e3be8dbe383b6a443a5
52508 F20110318_AACDSF bourret_j_Page_40.pro
633a7a293b180a3a2899dfe70024cb5b
761acac8bafd4b539bd842fc22e07189aa183d36
33597 F20110318_AACEEB bourret_j_Page_64.QC.jpg
bd5309ae64bfd53b383970e5405301e6
abdc1bae51b4b8df13981805eaf947d6978e6548
56149 F20110318_AACDNJ bourret_j_Page_18.pro
865ac21f97e0eaa54b0b0bc00286b0e5
9d32149fffd8755ea21755fe851ad07837af62ba
F20110318_AACDXE bourret_j_Page_60.tif
e0220991f63ee0bd3bca6d4342ece98b
e595690e9044cdf6536b621c5e90f9d176820b8f
7001 F20110318_AACDSG bourret_j_Page_60thm.jpg
a082bb322cfd0865cadb16e713174688
40018de39e96a84815d2a4b34fb9d4abd909f5d4


Permanent Link: http://ufdc.ufl.edu/UFE0012800/00001

Material Information

Title: Response Restriction and Contingent Access to Responding in Multi-Response Environments
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0012800:00001

Permanent Link: http://ufdc.ufl.edu/UFE0012800/00001

Material Information

Title: Response Restriction and Contingent Access to Responding in Multi-Response Environments
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0012800:00001


This item has the following downloads:


Full Text












RESPONSE RESTRICTION AND CONTINGENT ACCESS TO RESPONSES IN
MULTI-RESPONSE ENVIRONMENTS















By

JASON C. BOURRET


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


2005


































Copyright 2005

by

Jason C. Bourret

















ACKNOWLEDGMENTS

I would like to express my sincere gratitude to the members of my committee,

Drs. Jesse Dallery, Brian Iwata, Mark Lewis, Terry Scott, and Timothy Vollmer, for their

support, guidance, and mentorship. I would also like to thank my wife, Eileen Roscoe,

my mother, Patricia Machkovich, and my grandparents, George and Lucille Machkovich,

for their love, support, and encouragement in all my endeavors. I would also like the

thank the collaborators of this study, Timothy Vollmer and Luanne Witherup, for all their

contributions to the study.
















TABLE OF CONTENTS



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

LIST OF FIGURES ......... ......................... ...... ........ ............ vi

ABSTRACT .............. ..................... .......... .............. vii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

O overview ................. ............................................ ................... 1
Basic Research on Contingent Access to Responses as Reinforcement in Multi-
O perant E nvironm ents .............. ..................... ...................................................7
Application of Response Restriction and the Response-Deprivation Hypothesis......10
P u rp o se ............................................................................ 1 7

2 STUDY 1: AN EVALUATION OF A BRIEF RESPONSE-RESTRICTION
PROCEDURE IN A NATURALLY OCCURRING MULTI-RESPONSE
E N V IR O N M N E T ............................................................................ .....................2 0

M eth o d ..............................................................................................2 0
P participants and Setting .................................................................. ............... 20
Response Recording and Reliability ....................................... ............... 21
P ro cedu re ...................................... .............................................. 2 1
R results and D iscu ssion .............................. ........................ .. ...... .... ...... ...... 22

3 STUDY 2: AN EVALUATION OF RESPONSE RESTRICTION AS A METHOD
FOR INCREASING SOCIALLY DESIRABLE RESPONSES IN A
NATURALLY OCCURRING MULTI-RESPONSE ENVIRONMENT....... ........ 34

M ethod ................. .. ................................... ........................... 34
R results and D discussion .............................. ........................ .. ...... .... ....... ..35

4 STUDY 3: AN EVALUATION OF CONTINGENT ACCESS TO RESPONSES AS
A METHOD FOR INCREASING SOCIALLY DESIRABLE RESPONSES IN A
NATURALLY OCCURRING MULTI-RESPONSE ENVIRONMENT ...................43

M ethod ................. ................................... ...........................43










R results and D discussion ........................... .... .................. .. ......... .. .............45

5 GEN ER AL D ISCU SSION ............................................... ...... ........................53

L IST O F R E FE R E N C E S ............................................................................. .............. 60

B IO G R A PH IC A L SK E T C H ...................................................................... ..................63



















































v
















LIST OF FIGURES


Figure page

1 Seconds of responding during each minute of each session during brief response
restriction for K yle ................................................. .......... .....24

2 Seconds of responding during each minute of each session during brief response
restriction for John ...................... .. ........................ .. .. .. ..... .... ........... 26

3 Seconds of responding during each minute of each session during brief response
restriction for Sarah ........................................................................ ............. 2 8

4 Seconds of responding during each minute of each session during brief response
restriction for Beth .................................... .. .... ........ .......... .....30

5 Seconds of responding during the response-restriction treatment for Kyle.............36

6 Seconds of responding during the response-restriction treatment for John ............37

7 Seconds of responding during the response-restriction treatment for Sarah............39

8 Seconds of responding during the response-restriction treatment for Beth .............41

9 Seconds of responding during the contingency treatment for Kyle.........................46

10 Seconds of responding during the contingency treatment for John .........................47

11 Seconds of responding during the contingency treatment for Sarah.......................49

12 Seconds of responding during the contingency treatment for Beth .........................51















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

RESPONSE RESTRICTION AND CONTINGENT ACCESS TO RESPONDING IN
MULTI-RESPONSE ENVIRONMENTS

By

Jason C. Bourret

December, 2005

Chair: Timothy Vollmer
Major Department: Psychology

In this investigation, the effects of response restriction and contingent access to

responses in multi-response environments were evaluated in three studies. The same four

participants took part in all studies. Kyle was a typically developing 13-year-old male.

John was an 8-year-old male diagnosed with autism. Sarah was a typically developing

14-year-old female. Beth was a typically developing 7-year-old female. For all

participants, caregivers selected reading and mathematics as responses in need of

increase. In Study 1, participants were exposed to a brief response-restriction analysis,

consisting of a short series of response restrictions. For all participants, reading was

observed following the restriction of certain responses and, for all subjects other than

Sarah, both reading and mathematics were observed during response-restriction phases.

In Study 2, the efficacy of response restriction as a treatment method for increasing the

likelihood of socially desirable targeted responses in naturally occurring multi-response

environments was examined. The utility of the brief restriction analyses conducted in









Study 1 in identifying responses that compete with targeted behavior was evaluated by

restricting access to these responses for greater periods of time and examining response

reallocation. When restriction of these responses was insufficient to produce reliable

increases in targeted responding, the restriction of additional responses was evaluated.

Response restriction produced increases in targeted responding for 3 of 4 participants;

however, these increases were only reliable and sustained for Kyle. In study 3, the

effects of contingent access to responses on targeted behavior in multi-response

environments were examined. For all participants, contingent access to multiple

responses produced reliable increases in targeted behavior. Implications of the results for

research and application are discussed.














CHAPTER 1
INTRODUCTION

Overview

Classifying stimuli functionally is a hallmark of behavior analysis. As stimulus

operations are performed, their effects on behavior are the defining features used in

differentiating one type of stimulus from another (Skinner, 1935). Take, for example, the

classification of a stimulus as a reinforcer. Any stimulus that increases the probability of

a response being emitted when its presentation or removal follows the response is

classified as a reinforcer (Skinner, 1953). Thus, grain presented following the

keypecking of a pigeon that results in an increased likelihood of keypecking, a token

presented following completion of a math problem that results in an increased likelihood

of completing math problems, and a demand that is removed following hitting a teacher

which results in an increased likelihood of hitting the teacher would all be reinforcers.

These would be classified as positive reinforcers in the case of those that are presented

and negative reinforcers in the case of those that are removed.

The classification of stimuli, and reinforcers in particular, based on their function,

or the effects they have on behavior, has proven fruitful in keeping research in behavior

analysis conceptually systematic and engendering effective technologies for application

to socially significant behavior (Catania, 1998; Hanley, Iwata, & McCord, 2003). This

method of identifying reinforcers, however, has not been without criticism (Postman,

1947; Timberlake & Allison, 1974). Postman pointed out the circularity of defining a

reinforcer based solely on its effect on responding. This definition occurs after the









operation has taken place and allows for no independent means of classifying the

stimulus as a reinforcer. From a pragmatic standpoint this is troublesome. Without any

independent characteristics of a stimulus identifying it as a reinforcer, we are left with

identifying reinforcers through trial and error. For basic researchers, the fact that a

stimulus is not identified as a reinforcer until it has been demonstrated to have a

reinforcing effect may be an annoyance in that multiple stimuli may have to be tested

until a reinforcer is identified. In the application of behavior analysis to socially

significant behavior, it is sometimes critical to quickly and accurately identify reinforcing

stimuli and a failure to identify a reinforcer can result in the failure of an important

behavioral treatment.

Meehl (1950) proposed a partial solution to difficulties arising from post-hoc

identification of reinforcers. Meehl argued that reinforcers should be considered

transituational. A reinforcer identified in one context could then be used as a reinforcer

in another context. Thus, researchers or clinicians in need of a reinforcer may simply

make use of a stimulus that had previously been demonstrated to be a reinforcer. The

argument of transituationality depends upon the assumption that stimulus functions have

a certain amount of generality (Sidman, 1960). This assumption is commonly, and often

effectively, made by behavior analysts. The logic behind the use of preference

assessments to identify reinforcers for use in other contexts, for example, relies upon the

generality of the stimuli identified as reinforcers (Timberlake & Farmer-Dougan, 1991).

This generality, however, may be limited across varying levels of response effort,

schedule parameters, motivating operations, or the availability of substitutable or









competing reinforcers (Timberlake & Allison, 1974; Timberlake & Farmer-Dougan,

1991).

The probability-differential hypothesis, introduced by Premack (1959), suggested

an independent and a priori method of reinforcer identification. Premack conceptualized

reinforcement as involving contingent access to responses rather than stimuli. Rather

than grain as the reinforcer for the keypecking of a pigeon, according to this account, the

reinforcer is access to eating. The probability-differential hypothesis, sometimes called

the Premack principle, states that contingent access to a more probable response will

reinforce a less probable response. The initial probabilities are determined in a paired-

operant baseline. In this baseline, free access is permitted to both responses. Data are

collected on duration of each of the two responses in baseline in order to determine which

response is more probable. A reinforcement effect is then observed if access to the more

probable response is restricted and provided contingent upon the less probable response.

While there are conditions under which the Premack principle does not hold, it has

proven to have broad generality and can be a heuristic that may help clinicians identify

reinforcers (Catania, 1998; Timberlake & Farmer-Dougan, 1991).

Timberlake and Allison (1974) extended the Premack principle with the response-

deprivation hypothesis. Like Premack (1959), Timberlake and Allison conceptualized

reinforcement as involving contingent access to responses. Unlike Premack, they argued

that the critical feature of a reinforcement contingency was not that contingent responses

be more probable than instrumental responses (those that produce access to the

contingent responses). They stipulated that contingencies like those used in common

operant experiments disrupt the ratio of responding that would be observed if no









contingency were in place (e.g., a contingency involving access to water produced by the

wheel running of a rat typically involves restricting access to drinking such that the ratio

of running to drinking is greater than that which would be observed under a free-operant

baseline) and that organisms are motivated to respond in a manner that makes relative

response allocation more similar to that observed during free-operant baselines.

Timberlake and Allison cite this "disequilibrium" produced by operant contingencies as

the critical factor that determines a reinforcement contingency. They describe the

necessary condition for a reinforcement effect as

I 0,
C OC

where I and C represent measures of instrumental and consequence responses

respectively and 0, and Oc represent the operant level, or baseline level, of instrumental

and consequence responses. A reinforcement effect is predicted if a contingency is

arranged such that the ratio of instrumental to consequence responses is greater than that

observed during baseline. It is important to note that as long as the units of analysis are

constant across baseline and contingency conditions, they do not need to be the same for

instrumental and consequence responses. For example, if, in a paired-operant baseline, a

child read five pages of a book and watched television for one minute, the response-

deprivation hypothesis predicts a reinforcement effect given the following

I 5pages-read
->
C 'mm TV

Thus, a contingency arranged such that reading 10 pages is required to produce one

minute of television, or










pages-read pages-read

mmn-TV mln-TV

would be expected to produce a reinforcement effect, whereas a contingency arranged

such that reading three pages is required to produce one minute of television, or

3 5
pages-read pages-read
'mmnTV 'mmnTV

would not be expected to produce an increase in pages read over what was observed

during the free-operant baseline.

The manner in which the response-deprivation hypothesis expands upon the

Premack principle becomes more apparent if both responses are measured using the same

units. If, instead of pages read, minutes of reading were measured, the contingency

required for reinforcement may be something like the following

I 4mn-reading
->
C mmn-TV


An arrangement of

5 4
mm-reading mm-reading
'mmnTV 'mmnTV

should produce a reinforcement effect for reading. Thus, contingent access to a less

probable response should reinforce a more probable response. The relative baseline

probabilities of the responses are thus shown to be unimportant. The critical variable is

the arranged ratio of instrumental to consequence response as compared to the ratio

obtained during baseline (Timberlake & Allison, 1974).

The response-deprivation hypothesis has been demonstrated to be a useful

heuristic in identifying conditions under which access to responses will serve as









reinforcers in multiple studies in both basic and applied behavior analytic literature

(Allison & Timberlake, 1974; Amari, Grace, & Fisher, 1995; Charlop, Kurtz, & Casey,

1990; Hanley, Iwata, Thompson, & Lindberg, 2000; Homme, de Baca, Devine,

Steinhorst, & Rickert, 1963; Konarski, Johnson, Crowell, & Whitman, 1980; Mitchell &

Stoffelmayr, 1973; Osborne, 1969; Timberlake & Allison, 1974). A measure of

responding on a paired-operant baseline can allow a priori identification of a reinforcer.

An independent method of reinforcer identification helps alleviate the philosophical

concern regarding the circularity of a purely functional definition of a reinforcer and also

provides clinicians an additional method of reinforcer identification.

Some of the most promising areas for application of the response-deprivation

hypothesis are complex, multi-operant, environments (e.g., homes) in which there are

often a large number of available responses (e.g., watching television, playing video

games, listening to a radio, talking on the phone, reading a book, or doing homework).

Reinforcers identified through standard preference assessment methods, which are

typically conducted in restricted-operant environments in which the number and type of

responses available are well controlled, may be less effective in environments containing

numerous alternative responses. The response-deprivation hypothesis suggests

measuring baseline levels of responding in the environment in which the contingency is

to be put into place. Thus, potentially competing sources of reinforcement are taken into

account.

Surprisingly, however, very little research has explicitly evaluated the response-

deprivation hypothesis in multi-operant environments (Bernstein, 1998). Most of the

research on the response-deprivation hypothesis has been done using paired-operant









preparations, and the degree to which the same results would occur in multi-operant

environments remains largely unknown. For example, the following would be expected

to result in a reinforcement effect for reading given four minutes of reading and one

minutes of watching television during a paired-operant baseline

5 4
mm-reading mm-reading
mmnTV mnmTV

However, if, for example, baseline data collected in a multi-operant environment

indicated that reading occurred for four minutes, television watching for one minute,

playing video games for eight minutes, shooting pool for seven minutes, and talking on

the telephone for 10 minutes, it is unclear if the above contingency would produce the

same result. Perhaps the probability of reading would increase, or perhaps responding

would be reallocated to playing video games, shooting pool, talking on the phone, or

some new response. A paired-operant baseline does not permit the detection of such

reallocation. In the following section, basic research on response deprivation in multi-

operant settings is discussed.

Basic Research on Contingent Access to Responses as Reinforcement in Multi-
Operant Environments

Research on the effects of response restriction and contingent access to responses

in multi-operant environments with humans was pioneered in Bernstein's human-operant

laboratory (Bernstein, 1998; Bernstein & Ebbesen, 1978; Bernstein & Michael, 1990).

Bernstein and colleagues created a laboratory apartment in which human subjects lived,

24 hrs a day, for the duration of an experiment. Subjects made a list of desired activities

upon initiation of a study and those activities were typically available in the laboratory

apartment. Access to external stimuli and activities inside the apartment was controlled









by experimenters and data were collected on time spent engaging in various activities

across experimental conditions.

The first major study to come from this laboratory was an explicit investigation of

the effects of response restriction and contingent access to responses when multiple

response options were available (Bernstein & Ebbesen, 1978). Three subjects

participated, ranging in age from 19 to 39 years. The subjects lived in the experimental

apartment from between 21 to 34 days. Each selected several activities before the

beginning of the study and required materials for these activities were made available in

the apartment. Data on the duration of engagement in these various activities were

collected by observers who monitored the subjects from behind one-way mirrors. The

experimenters attempted to examine the effects of response restriction alone and

contingent access to responses in a multi-operant environment.

Responding was measured under three conditions: a free-operant baseline, in which

there was no restriction on response availability; a contingency condition, in which a

high-probability response, designated the contingent response, was restricted and

provided contingent upon a low-probability response chosen as the instrumental

response; and a matched-restriction condition, in which access to the response that was

the contingent response in the contingency condition was restricted such that the access

time was yoked to the time of access in the contingency condition. The contingencies

and restrictions were described to the subjects in each condition. The times at which

responses were restricted were signaled to the subjects by the illumination of labeled

lights on a panel in the apartment.









For all three subjects, response contingencies resulted in increases in instrumental

responding. For one subject, restriction of the consequence response alone produced

reliable increases in instrumental responding. For the other two subjects, the effects of

restriction alone were less reliable. One subject displayed no effect of response

restriction alone; the other demonstrated an increase in one topography, but this effect

was not replicated.

The Bernstein and Ebbesen (1978) study represented a move toward examining

response restriction and contingent access to responses in a naturally occurring

environment. Although the experiment took place in an operant laboratory and the

number of available responses was still limited, subjects had access to more than two

responses, and these were responses that are common in the natural environment. Their

findings suggest that response contingencies hold promise for application in complex,

multi-operant, human environments. They also suggest that, in some circumstances,

response restriction alone may be sufficient to increase targeted responses in these

environments.

The effects of response restriction on response allocation in a multi-response

environment were explicitly examined by Lyons and Cheney (1984). Subjects were four

adult, male rats and sessions took place in a chamber divided into five areas. Each area

contained one of the following sources of reinforcement: turning off the light, viewing a

female rat, turning off a white noise generator, water, or food. Free-operant baselines

resulted in similar patterns of responding for all subjects, with the most preferred area

being that which terminated the white noise. For each subject, two restriction conditions

were tested. In one, the most likely free-operant response was restricted. In the other, the









least likely free-operant response was restricted. The experimenters used the outcomes

restriction conditions to evaluate four rules for how responding might be re-allocated

following response restriction. The first was the "constant ratio" rule, which states that

the remaining responses will increase to occur during the same proportion of available

time that they did with the high-probability response present. The second was the "most

probable remaining response" rule, which stipulated that the second most likely response

during baseline will increase relatively more than other remaining responses. The third

was the "equal redistribution" rule, which stated that responding will be equally

reallocated to all available responses. The final rule tested was the "sequential

dependency" rule, which hypothesizes that the restriction of a response will also decrease

the response most likely to occur following the restricted response. The results replicated

those of Bernstein and Ebbesen (1978). Re-allocation of responding following restriction

was idiosyncratic to the participant and varied depending on whether it was the highest or

lowest probability response that was restricted. These data suggested that none of the

four rules discussed was sufficient to predict the results of response restriction for any

given subject. The reliability of the results, however, is unclear. For each subject, each

restriction phase was conducted only once. While no a priori rule was found to describe

the results, it may be that restriction effects are reliable within subject and that the effects

of previous restrictions can be good predictors of the effects of subsequent restrictions.

Application of Response Restriction and the Response-Deprivation Hypothesis

Osborne (1969) evaluated the effects of contingent and noncontingent free time

on the in-seat behavior of students in a school for the deaf. The six participants were

female students between 11 and 14 years old with severe to profound hearing loss. The

number of times each student got out of her seat during the daily recording period was









targeted for intervention. The experiment was designed as a reversal, with phase changes

occurring simultaneously for all participants. In each condition, contingencies in effect

were described to the participants. During baseline, there was no free time provided

either contingently or noncontingently. In the first treatment condition, each participant

was allowed 5 minutes of free time contingent upon 15 minutes without leaving her seat.

This was immediately successful, with the number of times each participant left her seat

reducing to near zero. Following this phase, a yoked-control condition was implemented

in which 5 minutes of noncontingent access to free time was provided every 15 minutes.

The level of responding in this condition was similar to that obtained in the contingent-

access condition with very slight upward trends in response frequency for three

participants. The noncontingent-access phase continued for five days for all participants.

This was followed by 51 days of the contingent-access phase during which the frequency

of participants leaving their seats remained near zero. Data collection was terminated at

this point and a one-day follow-up probe was conducted approximately six-weeks later.

In the intervening time, the free-time contingency remained in place and no participants

were observed to leave their seats during the probe. In the week following the probe, a

one day contingency reversal was implemented. Five minutes of free time during this

reversal was contingent upon participants leaving their seats at least once during each 15-

minutes period and the contingency was described to the participants. The frequency of

seat leaving during the reversal was elevated during the contingency reversal for all

participants.

Osborne (1969) was one of the first to use contingent access to responses for

behavior management in a natural environment. This study demonstrated the









effectiveness of contingent access to responses in a multi-response environment and also

illuminated some important issues with regard to manipulating access to responses in

these environments. One potentially critical variable in this study is the presence of a

possible negative reinforcement contingency. In the study conducted by Osborne, access

to restricted responses was also paired with escape from a work context. The degree to

which the obtained treatment effects were a function of contingent access to responses or

contingent escape from work is unclear. It may have been the case that a break from the

work context into a relatively impoverished environment would have been equally

successful. In fact, the effectiveness of noncontingent free time periods in suppressing

seat leaving suggests interpretation of the effects of treatment as an abolishing operation

(Laraway, Snycerski, Michael, & Poling, 2003) in that frequent breaks may have made

the demand context less aversive. An equally plausible explanation is that the free-time

periods resulted in frequent access to responses otherwise only available contingent upon

seat leaving and thus decreased the reinforcing efficacy of these responses. The relative

effects of these potential abolishing operations and the omission contingency are also

unclear.

Another important point of consideration is that the data collection, and the

contingencies used, converted an environment in which numerous responses were

available, presumably of different operant classes, into one in which only in-seat and out-

of-seat behavior were available. This aggregation may have concealed some potentially

interesting findings. The authors noted that out of seat activities included, but were not

limited to, talking with classmates, conversing with the teacher, sharpening pencils,

drinking water, and using the restroom. If seat leaving was maintained by access to









multiple responses, for example peer interaction and access to the restroom, contingent

access to only one of the two may not have been sufficient to suppress responding. This

is a particularly important case. If data collected in a free-operant condition indicated

that interacting with peers was the most probable response, both the Premack principle

and the response-deprivation hypothesis would predict that contingent access to this

response would serve as reinforcement. However, access to a restroom may be potent

enough to maintain seat leaving even if access to peer interaction were contingent upon

the absence of seat leaving. A finding of this sort would contradict the Premack principle

and the response-deprivation hypothesis; however, the aggregation of responses into

large classes, while experimentally convenient, disallowed the detection of this sort of

effect.

Mitchell and Stoffelmayr (1973) demonstrated the effectiveness of contingent

access to a high-probability response as a reinforcer for the behavior of two individuals

diagnosed with schizophrenia. Both participants resided in a psychiatric ward and were

employed stripping copper wire from coils. Thirty-second partial interval data were

collected on coil stripping over the course of 30-minute sessions and, during baseline,

both participants engaged in no coil stripping for six consecutive sessions. The

experimenters tested the effects of instructions alone, which consisted of one prompt to

work delivered at the beginning of each session, and instructions plus reinforcement,

which consisted of identical instructions and 90 seconds of sitting contingent upon the

removal of wire from three coils. Sitting was used as a consequence response as it was

identified as a high-probability response during pre-experiment data collection.

Instructions alone were found to have no increasing effect on responding. Instructions









plus reinforcement, however, resulted in high and stable levels of coil stripping. Like

Osborne (1969), Mitchell and Stoffelmayr demonstrated the use of contingent access to

responding to increase the likelihood of a socially significant response. They also

demonstrated that the generality of the reinforcement effects predicted by the response-

deprivation hypothesis extends to the population of individuals diagnosed with

schizophrenia. The effects of contingent sitting, however, were confounded with

potential effects of restriction of sitting. It may have been the case that restriction of

sitting alone would have resulted in increases in coil stripping. It is also difficult to

assess the degree to which the results of this study speak to the effectiveness of

contingent access to responses in multi-operant environments. All sessions took place in

the room of the psychiatric ward where all of the work of the patients was done. The

number of alternatively available activities was not clearly specified. The authors noted

that the subjects had previously refused to engage with multiple potentially reinforcing

items. Also, the fact that sitting rather than a more active response was chosen as the

consequence response suggests that the environment was relatively devoid of reinforcing

activities. A final concern involves the conceptualization of access to sitting as an

instance of reinforcement in the form of access to a response. In the contingency

condition, if the participants sat, their chair was tipped forward until they were forced to

stand. Standing for extended periods may have been effortful for the participants and it

may have been more appropriate to conceptualize the reinforcement effect as negative

reinforcement in the form of escape from standing rather than positive reinforcement in

the form of access to sitting.









The applied studies reviewed thus far have viewed restriction effects as potential

confounds to be controlled for. Green and Striefel (1988) evaluated these effects directly

in a multi-operant preparation composed of six available responses. Four children

between the ages of six and 10, all diagnosed with autism, participated. The six activities

were chosen by the children's classroom teacher and were either play or educational

activities. The analysis tested four hypotheses regarding the redistribution of responding

following restriction of the highest-probability response. The constant ratio rule, the

equal distribution rule, and the most probable alternative rule, as tested by Lyons and

Cheney (1984) were also tested by Green and Stiefel. The final hypothesis tested, the

selective substitution rule, was that the pattern of redistribution may differ for each

subject. The selective substitution rule is, essentially, that the outcome of restriction of

the most probable response cannot be predicted based on baseline response probability.

For each participant, Green and Striefel (1988) conducted an initial free-operant

baseline and then sequentially and cumulatively restricted access to the most probable

response such that five responses were available in the second condition, four in the third,

three in the fourth, and two in the third. Following the final restriction condition, another

free-operant condition was conducted. The experimenters found that response

redistribution followed no single pattern, thus supporting a selective substitution

description of the effects of response restriction. These data replicated the idiosyncratic

effects of response restriction alone discussed previously (Bernstein & Ebbesen, 1978;

Lyons & Cheney, 1984). Some potential limitations exist concerning the generality of

these findings extending into applied settings in which responding is not limited to

experimenter-selected responses. The first potential limitation relates to questions about









the reliability of the effects of response restriction for any given subject. Although no

general pattern of redistribution of responding was observed across participants, the

degree to which patterns of redistribution for individual subjects would remain stable

across multiple replications was not examined. It may be that reliable rules describing

response reallocation can be generated for individual subjects given certain restrictions.

A second concern involves the fact that responding could only be allocated to the

responses arranged by the experimenters. The experimental preparation did not address

the potential emergence of previously unobserved responses following restriction. In an

environment with more available responses, the emergence of previously unobserved

responses is always a possibility. A final potential limitation is that data were collected

using 30-second partial interval data collection. Following completion of this study, the

experimenters directly compared the partial interval data collection used with duration

recording in scoring video tapes of sessions similar to the baseline sessions of the study.

They found that the rankings of highest to lowest probability that would be assigned to

responding based on the outcomes of these two types of measurement agreed 83.3% of

the time. This suggests that the rankings obtained in the study may have been in error

nearly 17% of the time.

In a study designed to explicitly examine components of a behavioral treatment

involving contingent access to stereotypy as reinforcement, Hanley, Iwata, Thompson,

and Lindberg (2000) evaluated the effects of response restriction in isolation as well as

contingent access to behavior on engagement with leisure items. The experimenters

examined the effects of prompting alone, prompting plus blocking of stereotypy, and

prompting plus response blocking and contingent access to stereotypy. Participants were









three adults diagnosed with developmental disabilities who engaged in various forms of

stereotypy. The provision of leisure items, with no further intervention, had no

suppressive effect on stereotypy and item engagement was not observed for two

participants and occurred at a low level for the third. Prompting engagement with the

items resulted in no decreases in stereotypy for any participant and an increase in

engagement for only one participant. Prompting plus response blocking produced

decreases in stereotypy for all participants and increases in item interaction for two

participants. For the third participant, item interaction was increased by allowing access

to stereotypy contingent upon item engagement. The effects of response restriction alone

in Hanley et al. have some important, and relatively novel, implications for clinicians

attempting to produce increases in the appropriate behavior of their clients. Hanley and

colleagues were able to increase a targeted, socially relevant, response for two

participants without having to place a contingency on the response. While this study

represents an important initial demonstration of the use of response restriction as a

method for increasing a socially desirable response, it is important to note that sessions

were conducted in relatively barren therapy rooms. Previous research has demonstrated

that the effects of response restriction on reallocation of responding are idiosyncratic

across participants when multiple alternative responses are available (Bernstein and

Ebbesen, 1978; Green and Striefel, 1988). The degree to which clinicians can rely on

response restriction as a method for increasing targeted responses in less controlled

environments remains unknown.

Purpose

The present investigation was composed of three studies. All took place in

naturally occurring environments in which the number of available responses was









uncontrolled (e.g., a child's home environment or a play area). These relatively

unrestricted multi-response environments are those in which the transituationality of

reinforcers identified in preference assessments is challenged, insofar as multiple

potentially competing sources of reinforcement are available. Thus, an activity identified

as highly preferred in an isolated preference assessment may not compete with the

multiple sources of alternative reinforcement available in a child's home (e.g., television

shows, movies, radio, or video games). A response-deprivation approach to reinforcer

identification has promise for application in these settings. Following a response-

deprivation model of reinforcer identification, the environment in which the response to

be used as a reinforcer is identified is the same environment in which the contingency

will be implemented. Thus, alternatively available sources of reinforcement are taken

into account. As noted above, however, a number of questions remain to be answered

before clinicians can reliably use response restriction and response-deprivation

contingencies in complex, naturally occurring environments. The present studies

represent an attempt to answer the following questions. One, can response restriction

alone be used to reliably increase a targeted, socially desirable, response in a naturally

occurring, multi-response environment? Two, can contingent access to behavior be used

to increase socially important behavior in these same naturally occurring, unrestricted,

environments or will access to multiple competing sources of reinforcement impede its

effectiveness?

Study 1 was an evaluation of a brief response restriction analysis for identifying

responses that may be competing with targeted responses in the setting in which

treatment will be implemented. Study 2 was an examination of response restriction alone









as a treatment method for increasing socially important behavior. The purpose of Studies

1 and 2 was to determine if a brief restriction analysis would identify responses that

would be effective as part of a response-restriction treatment and to determine if response

restriction alone would produce reliable and sustained increases in desired behavior in a

multi-response environment. Study 3 was an investigation of the use of contingent access

to responses as a treatment designed to increase targeted behavior. The purpose of Study

3 was to determine the effectiveness of contingent access to responses as a method for

increasing desired behavior in a multi-response environment.

The present studies extended previous research in two ways. One, by explicitly

evaluating use of response restriction as a means of increasing a targeted, socially

desirable, response in a naturally occurring, multi-response environment. Two, by

explicitly evaluating the use of contingent access to behavior to increase socially

important behavior in these same naturally occurring, unrestricted environments.














CHAPTER 2
STUDY 1: AN EVALUATION OF A BRIEF RESPONSE-RESTRICTION
PROCEDURE IN A NATURALLY OCCURRING MULTI-RESPONSE
ENVIRONMENT

Previous research has demonstrated that the restriction of available responses can

result in increases in other responses even in the absence of a contingency (Bernstein &

Ebbesen, 1978; Green & Striefel, 1988; Hanley et al., 2000; Lyons & Cheney, 1984).

However, the effects of response restriction in multi-response environments have varied

across subjects and no single rule has been predictive of the outcome of these restrictions

(Green & Striefel, 1988; Lyons & Cheney, 1984). In Study 1, participants were exposed

to a brief response-restriction analysis, consisting of a short series of response

restrictions. The outcome of these brief response-restriction analyses were used to

identify responses that, when restricted, may result in increases in targeted socially

desirable responses as well as to evaluate the reliability of the effects of response

restriction on response reallocation.

Method

Participants and Setting

Participants were 4 individuals who had been referred by parents and caregivers

due to their lack of independent engagement in academic and other socially desirable

behavior. Kyle was a typically developing 13-year-old male. John, was an 8-year-old

male diagnosed with autism. Sarah was a typically developing 14-year-old female. The

final participant, Beth was a typically developing 7-year-old female. Each session was

5 minutes in duration and sessions were conducted 1 to 3 times per week. Sessions for









Kyle, Sarah, and Beth were all conducted in their respective homes. Sessions for these

participants began in the room of their choice, however, participant's movement

throughout their home was not restricted. For John, sessions were conducted in a play

area at his school, in which a computer, a CD player, a television with a VCR and a

number of movies, as well as a wide variety of games, toys, and other activities were

available. For all participants, the materials necessary for the targeted academic

responses were made available in the room in which the session began.

Response Recording and Reliability

Observers collected data using palm-top computers on duration of time spent

engaging in various activities. Engagement was defined as orienting toward and, when

appropriate, being in contact with materials necessary for given activities. Data

collectors scored the occurrence of any identifiable response, if no identifiable response

was observed the data collectors scored nothing. A second observer collected data

independently on 37.8% of all sessions across all studies. Interobserver agreement was

calculated by dividing each session into 10-second intervals, dividing the smaller

duration of recorded responses by the larger across observers, averaging these scores

across the session, and converting them to a percentage. Overall agreement scores for all

participants across all measures averaged 97.3% (range 80.3% to 100%).

Procedure

The primary investigator met with the caregivers of each participant to identify

responses to increase. For Kyle, Sarah, and Beth, age appropriate reading and

mathematics activities were identified by caregivers as responses that were considered

important for the children and which rarely happened at home. For John, working on









mathematics worksheets and reading were identified as important responses that did not

occur unless he was explicitly prompted in an academic setting.

Each participant completed two to four brief restriction analyses. These analyses

consisted of a series of five 5-minute sessions. The first 5 minutes were a free-operant

period. Participants were instructed that they may do anything that they like. Following

the free-operant session, the activity that was engaged in for the greatest duration was

restricted and participants were told that they may do anything that they like other than

the restricted activity. In each subsequent session, the activity engaged in for the longest

duration in the previous session was also restricted and subjects were instructed that they

may do anything that they like other than the restricted activities. Responses were

restricted by instructing participants that they may not engage in these response and no

participants attempted to engage in the responses that they were told were restricted.

Two brief-restriction analyses were conducted with Kyle, three with Sarah and Beth, and

four with John. At the time the brief restriction analyses were conducted, the duration of

participation of Kyle, Sarah, and Beth in this series of studies was unclear. For this

reason fewer brief restriction analyses were conducted with these participants to allow

them to move on to the treatment studies in the series (Studies 2 and 3) more quickly.

Only one brief restriction analysis was conducted per day for all participants.

Results and Discussion

Figure 1 shows the outcomes of the brief restriction analyses for Kyle. In the top

panel, during the free-operant phase, Kyle spent his time playing video games.

Following restriction of video games (Restriction 1), Kyle shot pool. When access to

video games and shooting pool were restricted (Restriction 2), Kyle watched television.

When television was also restricted (Restriction 3), Kyle spent his time reading.






23


Following the restriction of reading as well (Restriction 4), he allocated his time to a

mathematics workbook. The bottom panel shows the results of the second restriction























Brief Restriction Assessment 1


Free Operant
60 -


50 -Playing
Video
40 Games


Restriction 1


Playing
Pool


Restriction 2


Watching
Television


Restriction 3


Reading


Restriction 4


Math


5 10 15 20 25

Brief Restriction Assessment 2


Free Operant Restriction 1 Restriction 2 Restriction 3
60 ] /| -* -*


Restriction 4


Hair
Twirling


5 10 15 20


Minutes





Figure 1. Seconds of responding during each minute of each session during brief

response restriction for Kyle; sessions are denoted by phase change lines.









analysis for Kyle. In the free-operant phase, Kyle spent all of his time doing math.

Following restriction of math (Restriction 1), Kyle allocated his time to reading. When

both mathematics and reading were restricted (Restriction 2), Kyle played video games.

Following the restriction of mathematics, reading, and video games (Restriction 3) Kyle

sat on the floor, engaging in no particular identifiable response. Restriction 4 was

identical to Restriction 3 as Kyle had emitted no identifiable response to be restricted. In

Restriction 4, he remained seated on the floor and spent the session twirling his hair.

Figure 2 depicts the results of the brief restriction analyses for John. Data from

the first analysis are plotted in the top panel. During the free-operant phase, John played

computer games. Restriction of computer games (Restriction 1) resulted in John playing

on a therapy ball. When computer games and the therapy ball were restricted (Restriction

2), he played with a handheld electronic game. When the handheld game was also

restricted in Restriction 3, John watched a video. When access to videos was also

restricted in (Restriction 4), John initially allocated his responding to mathematics

worksheets during the first 2 minutes of the session, and primarily engaged in reading

during the final 3 minutes.

The second panel shows the results of John's second brief restriction analysis. In

the free-operant phase, he played computer games. With access to the computer

restricted in Restriction 1, John watched videos. When both access to the computer and

videos were restricted (Restriction 2), John played on a therapy ball. When access to

computer, videos, and the therapy ball were all restricted (Restriction 3), he played with a

handheld electronic game. When the handheld game was added into the restriction in

















Brief Restriction Assessment 1

Free Operant Restriction 1 Restriction 2 Restriction 3 Restriction 4
60 / / th

Playing Playing Playing an Watching /
50 ona on a Electronic Videos
Computer Therapy Ball Game
40


30

20


10

0 1 Reading

5 10 15 20 25
S10 15 20 25


Brief Restricton Assessment 2
Free Operant Restriction 1 Restrcton2 Restriction 3


Restriction 4






/-


10 15 20 25

Brief Restriction Assessment 3
Restriction 1 Restriction 2 Restriction 3 Restriction 4


10 15 20 25

Brief Restriction Assessment 4
ion 1 Restriction 2 Restriction 3 Restriction 4


Minutes


20 25


Figure 2. Seconds of responding during each minute of each session during brief

response restriction for John; sessions are denoted by phase change lines.









Restriction 4, John spent time working on mathematics during the first 3 minutes and

reading during the last 2.

The third panel depicts the results from John's third brief-restriction analyses.

John played on the computer during the free-operant phase. When access to the

computer was restricted in Restriction 1, he watched a video. When access to the

computer and movies were restricted in Restriction 2, John played with a handheld

electronic game. When the electronic game was also restricted in Restriction 3, he spent

the majority of the first 3 minutes playing on a therapy ball and the last 2 reading. As

playing on the therapy ball was the most probable response in the third restriction phase,

this response was added to the restriction during Restriction 4. During the first minute of

the Restriction 4, John spent time reading and doing mathematics and during the final 4

minutes John only worked on mathematics.

The bottom panel shows the results of the final brief restriction analysis for John.

John played on the computer during the free-operant condition. When the computer was

restricted in Restriction 1, John watched a video. With restricted access to the computer

and movies in Restriction 2, he played with a handheld electronic game. When access to

the computer, videos, and handheld game were all restricted in Restriction 3, John played

on the therapy ball. When the therapy ball was also restricted in Restriction 4, he read

during the first 2 minutes of the session but this decreased during the third minute and he

sat quietly on the floor for the final two minutes.

Figure 3 depicts the results for Sarah. The top panel shows results of the first

brief restriction analysis. In the initial free-operant phase, Sarah allocated most of her

time to watching television. With television restricted in Restriction 1, Sarah watched a

















Brief Restriction Assessment 1


Free Operant
60



5 Watching
Television
40


30


20


10


0


Free Operant
60 -


0 -


-+--+-I


Restriction 1



Watching
Videos


Restriction 2



Reading


Restriction 3


Drawing


Restriction 4



Weaving


5 10 15 20 25


Brief Restriction Assessment 2


10 15 20 25


Brief Restriction Assessment 3


Restriction 1
x-x-x-x


Playing
with
Clay














-+-+--+-4


Restriction 2


Coloring


Restriction 3
x-x

Playing
with a
Toy
Phone














-+ -+--+--


10 15

Minutes


Restriction 4
+-+---+

+ Readng
a
Magazine


20 25


Figure 3. Seconds of responding during each minute of each session during brief

response restriction for Sarah; sessions are denoted by phase change lines.


-
+-+-E-+-H


I









video. With both television and movies restricted in Restriction 2, Sarah allocated 100%

of her time to reading. With reading also restricted in Restriction 3, she spent all of her

time drawing. With television, videos, reading, and sketching restricted in Restriction 4,

Sarah's responding was allocated to weaving.

The second panel shows the results for the second brief-restriction analysis. Sarah

allocated all of her time to weaving in the free-operant phase. With weaving restricted in

Restriction 1, Sarah watched television. When weaving and television were restricted in

Restriction 2, she watched a video. When weaving, television, and movies were

restricted in Restriction 3, Sarah spent her time blowing bubbles. With bubbles also

restricted in Restriction 4, she listened to a CD player for the majority of the first 3

minutes and this response decreased during the final 2 minutes.

The bottom panel depicts the results for the third brief-restriction analysis for

Sarah. In the free-operant phase, she spent most of her time watching television. With

television restricted in Restriction 1, Sarah played with clay. With restriction of

television and clay in Restriction 2, Sarah colored some pictures with markers. With

restriction of television, clay, and coloring in Restriction 3, Sarah sat on her bed without

engaging in any particular response for the first half of the session and played with a toy

phone for the second half. With playing with the toy phone added to the restricted

responses in Restriction 3, Sarah spent her time reading a magazine.

Figure 4 shows the results for the brief restriction analysis for Beth. In the top

panel, the data for the first analysis are depicted. In the free-operant phase, Beth watched

television. In Restriction 1, television was restricted and Beth worked on mathematics.

In Restriction 2, with television and mathematics restricted, she played with dolls. With
















Brief Restriction Assessment 1


Free Operant
60


50 Watching
Television

40


30


20


10


0 -I--UU-lK--5
5


Restriction 1


SMath


Restriction 2 Restriction 3 Restriction 4


10


15 20 25


Brief Restriction Assessment 2


Free Operant Restriction 1
60 t-t


Restriction 2 Restriction 3


5 10 15 20 25


Brief Restriction Assessment 3


Restriction 1 Restriction 2
-o--O-o--O +--+-+-+--+

Working
on a
Report


Restriction 3
-e--- I


Restriction 4


Playing
with
Chalk


10 15 20 25

Minutes


Figure 4. Seconds of responding during each minute of each session during brief

response restriction for Beth; sessions are denoted by phase change lines.


Free Operant
--D -- --


50


40


30


20


10


0 -


+---++---+-----









television, mathematics, and dolls restricted in Restriction 3, Beth initially spent some

time writing in her diary, then played with cards, and ended the session playing with a

ball. As playing with cards was the most likely response in Restriction 3, this response

was included in the restriction for Restriction 4. In Restriction 4, Beth spent the first

minute playing with a ball and the majority of the rest of session writing in her diary.

The middle panel of figure 4 displays the results of the second brief-restriction

analysis for Beth. When television was restricted in Restriction 1, Beth spent the initial

portion of the session counting money in her purse. During the end of the condition, Beth

spent her time reading and playing with a ball. As money counting was the most

probable response during Restriction 1, both television and money counting were

restricted in the second. In Restriction 2, Beth spent her time playing with clay. When

clay was added to the restriction in Restriction 3, she worked on mathematics for the

majority of the condition. With mathematics included in the restriction in Restriction 4,

Beth spent most of her time reading.

The bottom panel of figure 4 depicts the results of the third brief-restriction

assessment. During the free-operant phase, Beth worked on mathematics. When

mathematics was restricted in Restriction 1, Beth watched television. With television and

mathematics restricted in Restriction 2, Beth worked on a report for her social-studies

class. When access to working on the report was also restricted in Restriction 3, Beth

played with a ball. With television, mathematics, working on a social-studies report, and

playing with a ball restricted in Restriction 4, she allocated her time to drawing with

chalk.









The brief restriction analyses produced samples of free-operant responding and

responding under various response restrictions. For all subjects, one of the targeted

responses (reading) was observed following the restriction of certain responses and, for

all subjects other than Sarah, both of the targeted responses (reading and mathematics)

were observed during response-restriction phases. Although there was variability in

effects of sequential restriction across replications for all subjects, the fact that, during

some brief-restriction analyses, certain responses had to be restricted before targeted

responses occurred suggests that these responses may also be competing with the targeted

responses outside of experimental sessions. Also, the fact that targeted responses were

observed in the absence of experimenter imposed contingencies suggests that these

contingencies may not be necessary to produce targeted responding. The brief restriction

analyses for Kyle suggested that playing video games, playing pool, and watching

television may compete with reading and that playing video games, playing pool,

watching television, and reading may compete with doing mathematics. For John, they

suggested that playing on the computer, watching movies, playing a handheld electronic

game, and playing on a therapy ball may compete with both reading and mathematics.

For Sarah, watching television and movies were identified as responses that may compete

with reading. For Beth, brief restriction analyses suggested that watching television,

playing with money, playing with clay, and doing mathematics may compete with

reading and also that watching television, playing with money, and playing with clay may

compete with doing mathematics. In Study 2, a response-restriction treatment was

implemented for each of these participants in which responses identified as possibly

competing with targeted responses in brief restriction analyses were restricted for greater









periods of time in order to determine if restriction of these responses in the absence of an

experimenter imposed contingency was sufficient to produce increases in targeted

responding in naturally occurring, complex, multi-response environments.














CHAPTER 3
STUDY 2: AN EVALUATION OF RESPONSE RESTRICTION AS A METHOD FOR
INCREASING SOCIALLY DESIRABLE RESPONSES IN A NATURALLY
OCCURRING MULTI-RESPONSE ENVIRONMENT

In Study 2, the efficacy of response restriction as a treatment method for

increasing the likelihood of socially desirable targeted responses in naturally occurring

multi-response environments was examined. The utility of the brief restriction analyses

conducted in Study 1 in identifying responses that compete with targeted behavior was

evaluated by restricting access to these responses for greater periods of time and

examining response reallocation. When restriction of these responses was insufficient to

produce reliable increases in targeted responding, the restriction of additional responses

was evaluated.

Method

The participants and settings were identical to those in Study 1. During free-

operant conditions, no restrictions were placed on responding. During the first response-

restriction phase for all participants, those responses identified during the brief restriction

analyses as potentially competing with one of the targeted responses were restricted. For

Kyle, the restricted responses were playing video games, playing pool, and watching

television. For John, the restricted responses were playing on the computer, watching

movies, playing on a Leap Pad, and playing on a therapy ball. For Sarah, the restricted

responses were watching television and movies. For Beth, the restricted responses were

watching television, playing with money, and playing with clay. If restriction of these

responses was not sufficient to produce increases in targeted responding, additional









restriction phases were implemented in which the most probable response observed in the

previous restriction phase was also restricted. As in Study 1, responses were restricted by

instructing participants that they may not engage in these responses. No participants

attempted to engage in restricted responses. A reversal design was used in which free-

operant conditions were alternated with each response-restriction condition. Data were

collected on duration of observed responses and plotted in five-minute units. Data were

collected during 15- to 30-minute periods, on one to three days per week depending upon

the availability of the participants.

Results and Discussion

Figure 5 shows the results for the response-restriction treatment for Kyle. During

the free-operant condition, reading and mathematics occurred briefly, but mainly he

watched television and played video games. During the first exposure to the restriction

condition (R1), Kyle allocated time toward mathematics during 100% of all sessions. In

a brief reversal to free operant, he watched television exclusively. In the second exposure

to the R1, he spent all of his time drawing. In another reversal to free operant, he began

by engaging in mathematics, but ended the condition by watching television. In the third

exposure to R1, he again spent 100% of his time drawing. Thus, the second restriction

condition (R2) involved restriction of drawing in addition to restriction of the previously

mentioned preferred activities. This modification resulted in exclusive responding to

mathematics. When mathematics was restricted in addition to other preferred activities

(during R3), he allocated all of his time toward reading.

Figure 6 displays the results for the response-restriction treatment for John. In the

first free-operant condition, John exclusively spent his time playing computer games. In






























FO R1 FO R1 FO R1 FO R2 FO R2 FO R3 FO R3
300 A XX 3 *_____ --- 1 0 W
S- I 1-- Playing
Video
- 250 Games
-I- Playing
O Pool
200
-A- Watching
Television
150 + Math

S100 I -- Reading

) 50 A- Drawing


10 20 30 40 50 60 70 80 90


Sessions


Figure 5. Seconds of responding during the response-restriction treatment for Kyle. FO
signifies free-operant phases. During R1 phases, access to video games, pool
and television was restricted. During R2, video games, pool, television, and
drawing were restricted. During R3, video games, pool, television, drawing,
and math were restricted.


























300 4""

0 Playing
2~50 on a Math
SComputer
S200

150

-e
100

0 50 Readin

0 ------------ A ------
5 10 15 20 25 30 35

Sessions

Figure 6. Seconds of responding during the response-restriction treatment for John. FO
signifies free-operant phases. During R1 phases, access to the computer,
videos, electronic game, and therapy ball was restricted. During R2 the
computer, videos, electronic game, therapy ball, and books were restricted.









the first restriction phase (R1), reading initially increased but then decreased toward the

end of the phase as he allocated more time to doing mathematics. In the following free-

operant phase, John played computer games for the entire condition. In a second

exposure to R1, reading again increased at the beginning of the condition but decreased

over time. In the subsequent free-operant phase, John spent all of his time playing

computer games. In the second restriction condition (R2), reading was restricted in

addition to the responses restricted in R1 and mathematics initially increased but did not

maintain. In another reversal to free operant, John played on the computer during 100%

of all sessions. In a second exposure to R2, there was an initial increase in mathematics

but a downward trend across the condition. As with Kyle, reliable increases in targeted

responses were observed following response restriction, however, for John, increases in

targeted responses did not maintain across restriction conditions. Nonetheless, in the

final R2 condition, he spent over 16 of the 20 total minutes working on mathematics.

Figure 7 shows the results of the restriction treatment for Sarah. In the first free-

operant phase, most of her time was spent watching television. In the first restriction

phase (R1), playing with a singing toy, playing a handheld video game, playing with clay,

and playing with cards were all observed, with the most likely response being playing

with the singing toy. In a return to free operant, Sarah spent nearly all of her time

watching television. In the following R1 phase, Sarah primarily worked on crafts. In

another reversal to free operant, she spent 100% of her time watching television. In a

return to R1, the most probable response was again doing crafts and in the subsequent

free-operant phase she, again, spent all of her time watching television. During the

























--Watching Television

FO R1 FO R1 FO R1 FO R2 FO R2 FO R3 FO R3 -U--MagazineReading
300 --- -. _mm -
Si-- Playing an Electronic
Game
250
S250 Playing with Clay

S 200- U Playing with Dolls

150 --Listening to a CD
So Player
-I-- Playing with a
. 100 Singing toy
Reading
0 50
S- Playing with Cards
V) 0 IT I,,, II\ ,I I I I
Doing Crafts
1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91
--Listening to a Radio
Sessions


Figure 7. Seconds of responding during the response-restriction treatment for Sarah. FO
signifies free-operant phases. During R1 phases, access to television and
videos was restricted. During R2, television, videos, and crafts were
restricted. During R3, television, videos, crafts, and the radio were restricted.









second restriction condition (R2), television, videos, and crafts were restricted and Sarah

exclusively spent her time listening to music. In the next free-operant phase, Sarah

watched television during 100% of all sessions. In a return to R2, she again exclusively

listened to music. In the following free-operant condition, she spent all of her time

watching television. In the third restriction condition (R3), access to a radio was

restricted along with those responses restricted in R2 and Sarah spent 100% of her time

reading a magazine. In the subsequent free-operant phase, she again exclusively watched

television. In a return to R3, Sarah, again, read a magazine during 100% of all sessions.

For Sarah, restriction of certain responses reliably increased certain other responses.

While the response restrictions did not produce increases in the initially targeted

academic-related reading and mathematics activities for Sarah, her caregivers indicated

that they preferred her spending her time on the activities that were produced through

response restriction (working on crafts, listening to music, and reading a magazine) to her

watching television (her most common response during free-operant periods).

Figure 8 depicts the results for the response-restriction treatment for Beth. During

the first restriction condition (R1), her responding was variable, with playing with dolls,

reading, and mathematics all observed. During the first free-operant condition, Beth

predominantly watched television. In a return to R1, she spent all of her time doing

math. In the following free-operant phase, Beth allocated all of her time to watching

television. In the next exposure to R1, Beth primarily worked on crafts during the first

four sessions and mathematics in the final two sessions. In a return to free-operant,

Beth's responding was variable, with reading, doing crafts, watching television, and

playing a handheld video game all occurring. In the following restriction phase, Beth








41























R1 FO R1 FO R1 FO R1 FO R1 FO R1 FO R1
300
300 ---Watchng Television

1 250 Playmg an Electronic
Game
2 1 1 t-A-Domg Crafts

200 -Readmg

S- I 1- Math
150
o -4- Playing with Dolls
S 100 Writing in a Diary

O I -0-Playing with a Ball
U 50
S- Watching Movies

o0 Irrrr -DD DI I--
0 10 20 30 40 50 60 70 80 90

Sessions


Figure 8. Seconds of responding during the response-restriction treatment for Beth. FO
signifies free-operant phases. During R1 phases, access to television,
counting money, and playing with clay was restricted.









spent most of her time reading. In the subsequent free-operant phase, she exclusively

watched television. In the next response-restriction phase, Beth allocated most of her

time to math. During the return to free operant, she again spent 100% of her time

watching television. In the next R1 phase, Beth spent most of her time watching a movie.

In the following free-operant phase, she exclusively watched television. In the final

response-restriction phase, Beth worked on math during the beginning of the condition

and worked on crafts during the end of the condition. For Beth, while no reliable effects

of response restriction were observed, increases in targeted behavior were observed in

various restriction phases.

Response restriction produced increases in targeted responding in naturally

occurring, multi-response environments, for three of four participants. However, these

increases were only reliable and sustained for Kyle. For John, response restriction

produced reliable initial increases in targeted behavior, but these increases did not

maintain. For Sarah, response restriction did not produce increases in targeted responses

but did produce reliable increases in other desirable responses. In Beth's case, response

restriction had variable effects, producing increases in targeted responding in some

phases, but not reliably. In order to evaluate an alternative treatment for Kyle and to

identify a method for producing more desirable treatment outcomes for John, Sarah, and

Beth, the effects of contingent access to responses on targeted behavior in naturally

occurring, multi-response environments for each participant were examined in Study 3.














CHAPTER 4
STUDY 3: AN EVALUATION OF CONTINGENT ACCESS TO RESPONSES AS A
METHOD FOR INCREASING SOCIALLY DESIRABLE RESPONSES IN A
NATURALLY OCCURRING MULTI-RESPONSE ENVIRONMENT


Study 2 showed that response restriction alone reliably produced sustained

increases in targeted responding for one of four participants. In study 3, the effects of

contingent access to responses on targeted behavior in multi-response environments were

examined for all four participants. Initially, the effects of contingent access to multiple

responses were compared to either free-operant or response-restriction control conditions.

Following this, the effects of contingent access to a single high probability response were

examined.

Method

The participants and settings were identical to those in Studies 1 and 2. A single

response was targeted for increase for each participant, reading for Kyle, math for John,

and reading for Sarah and Beth. For Kyle and John, contingent responses were those that

were likely to be observed under free-operant conditions and those observed to be

probable following response restriction in Study 2, but not those which, when all were

restricted, resulted in an increase in the targeted responses. This ensured that the

contingent responses had a high likelihood of occurrence and that their restriction alone

would not result in increases in targeted behavior. Because reliable increases in targeted

behavior were not observed for Sarah and Beth, the same responses used in the initial

response-restriction phases in Study 2 were used as initial contingent responses in Study









3. These responses were found to be likely under free-operant conditions or following

restriction of probable responses, but their restriction did not produce increases in

targeted behavior. For Kyle, the initial contingent responses were playing video games,

watching television, playing pool, and drawing. For John, the initial contingent responses

were playing on the computer, watching movies, and playing an electronic game. For

Sarah, the initial contingent responses were watching television and movies. For Beth,

the initial contingent responses were watching television, counting money, and playing

with clay. For all participants, following the evaluation of a multiple-response

contingency, the effects of contingent access to a single response were examined. For

Kyle, the single contingent response used in this analysis was playing video games. For

John, it was playing on a computer. For both Sarah and Beth it was watching television.

The single contingent responses were those engaged in during periods in which access to

contingent responses was delivered in the multiple-response contingency phases. During

single-response contingency conditions, only the one contingent response was restricted

and all other responses were concurrently available with the instrumental response.

Free-operant and response-restriction phases were conducted in a manner

identical to those in Studies 1 and 2 except that participants were told which responses

were restricted and that they were free to engage in targeted responses, other non-

restricted responses, or nothing at all. For all participants other than John, during all

response-contingency phases, access to contingent responses was restricted and each

minute spent engaging in the targeted response produced a token exchangeable for one

minute of access to restricted responses following the day's observation period. Poker

chips were used as tokens for all participants. Observation periods lasted between 20 and









30 minutes. At the initiation of each response-contingency observation period,

participants were told that they could earn tokens that they could exchange for access to

restricted responses by doing targeted responses, and that they were free to choose to do

the targeted response, another activity, or nothing at all. John's initial response-

contingency phase was conducted in a manner identical to those of the other participants.

Following this phase, for John, a 21.6 by 27.9 cm plastic board was constructed with 20

Velcro squares arranged in four rows of five on which John could place earned tokens.

In this phase, tokens were poker chips with Velcro squares attached to one side. In all

subsequent response-contingency phases for John, no tokens were used and each minute

of engagement in targeted behavior produced one minute of access to restricted responses

immediately. Session time was stopped during periods in which access to restricted

responses was provided.

Results and Discussion

Figure 9 shows the data for the response-contingency analysis for Kyle. In phase

1, with access to playing video games, watching television, playing pool, and drawing

contingent upon reading, Kyle spent all of his time reading. In the following free-operant

condition, no reading was observed. Kyle spent 100% of his time during all sessions

reading during a return to contingent access to video games, watching television, playing

pool, and drawing. In the next free-operant phase, no reading occurred. We noted that

during all token-exchange periods Kyle exclusively engaged in playing video games.

When access to playing video games alone was contingent upon reading and Kyle spent

all of his time reading. In the following video game restriction phase, reading was not

observed. When access to playing video games was again contingent upon reading in the





























VG, TV, Pool,
Draw Cont


VG,TV,Pool,
FO Draw Cont FO


VG VG
Cont Rest


10 15 20 25 30 35 40

Sessions


Figure 9. Seconds of responding during the contingency treatment for Kyle. FO signifies
free-operant phases. The abbreviations Cont. and Rest. signify contingency
and response restriction conditions respectively.




































Computer, video, Computer, video, Conputer, Coputer, video,
andElectronic andElectronic MVdeo,and andElectronec
GameCont GameDirect Electromnc GameDirect
NewTokenBoard Cont GameRest Cont


Computer Conputer Computer Conputer
Rest Cont Rest Cont


5 10 15 20 25 30 35 40 45 50 55 60 65 70 75



Sessions




Figure 10. Seconds of responding during the contingency treatment for John. FO

signifies free-operant phases. The abbreviations Cont. and Rest. signify

contingency and response restriction conditions respectively.


Computer,
Video,and
Electronic
GameRest


Computer,
Video, and
Electronic
GameContt









subsequent phase, no reading was observed during the first four sessions, then responding

increased for the remainder of the phase.

Figure 10 depicts John's results. In the first restriction phase, John spent no time

doing math. In the first contingency phase, doing math increased and was observed at

variable levels. In the subsequent phase, John was provided with a board on which he

could place his tokens and his doing math remained variable during the first half of the

phase but was eliminated during the final three five-minute periods. In the next phase,

the tokens were removed and each minute of doing worksheets produced immediate

access to restricted responses and John spent nearly all of his time on math. In a return to

the restriction phase, no math was observed. In the following contingency phase, John

spent almost all of his time doing math. As was the case with Kyle, John exclusively

engaged in a single response during all periods in which contingent access to responses

was permitted. For John, this response was playing on a computer. In phase 7, the

effects of computer restriction alone were examined and John spent no time working on

math. When access to the computer was made contingent upon doing math in the

following phase, doing math increased in the third five-minute period and John

exclusively engaged in this response for the remainder of the phase. In a return to

computer restriction, John spent no time was spent on math. In the final contingent-

computer phase, John allocated all of his time to doing math.

Figure 11 depicts the results for Sarah. During the first phase, access to watching

television and videos was contingent upon reading and high levels of reading were

observed. When watching television and movies were restricted with no contingency in

the following phase, no reading was observed. In a return to contingent television and


























TV TV TV TV
andVideo andVideo andVideo andVideo TV
Cont Rest Cont Rest Cont


TV TV
Rest Cont


10 20 30 40 50

Sessions


Figure 11. Seconds of responding during the contingency treatment for Sarah. FO
signifies free-operant phases. The abbreviations Cont. and Rest. signify
contingency and response restriction conditions respectively.









videos, Sarah spent all of her time reading. In a second restriction phase, no reading was

observed. During all periods in which access to restricted responses was delivered during

contingent access to responses phases Sarah exclusively watched television. In phase 5,

contingent access to watching television alone resulted in Sarah spending all of her time

reading. When access to watching television was restricted, with no contingency, in the

following phase, reading occurred during the first four five-minute periods and then did

not occur for the remainder of the phase. In a return to contingent television, reading was

not observed during the first seven five-minute periods, but occurred during all

subsequent sessions of the phase. In a second television-restriction condition, no reading

occurred. In a final contingent-television condition, Sarah spent all of her time reading.

Figure 12 depicts the data for the response-contingency analysis for Beth. In the

first television, counting money, and clay restriction, no reading was observed. When

these responses were contingent upon reading, in the following condition, high levels of

reading were observed. In a return to restriction, reading again decreased to zero levels.

When the contingency was reinstated, Beth spent nearly all of her time reading. As was

the case with Sarah, Beth exclusively watched television during token-exchange periods

in which access to previously restricted responses was permitted. In phase 5, access to

watching television was contingent upon reading and high levels of reading maintained.

When access to watching television was restricted in the following phase, time spent

reading decreased. In a return to contingent television, reading initially increased during

the first four sessions, decreased to zero levels during the following eight sessions and

then increased to high levels for the remainder of the phase (12 sessions). In a return to























TV, Money, TV, Money,
Clay Cont Clay Cont
TV, Money, TV, Money, TV
Clay Rest Clay Rest Cont
300 o I ~ **__ Ir .


TV


TV TV
Rest Cont


10 20 30 40 50 60 70 80 90 100


Sessions




Figure 12. Seconds of responding during the contingency treatment for Beth. FO

signifies free-operant phases. The abbreviations Cont. and Rest. signify

contingency and response restriction conditions respectively.









restriction, reading was not observed for the entire phase. In another reversal to

contingent television, reading increased and occurred at variable levels.

For all participants, contingent access to multiple responses produced reliable

increases in targeted behavior in naturally occurring multi-response environments. When

responses used as contingent responses were restricted in the response-restriction analysis

(study 2), no increase in reading was observed suggesting that the response contingency

was responsible for the increase in, and maintenance of, targeted behavior during the

response-contingency analyses. Another finding common to all participants was that,

during periods in which access to restricted responses was permitted in response-

contingency phases, responding was allocated exclusively to a single behavior (playing

video games for Kyle, playing on a computer for John, and watching television for both

Sarah and Beth). While there was more variability in instrumental responding than was

observed when multiple contingent responses were used, contingent access to this single

response resulted in increases in targeted behavior for all participants even though

multiple alternative responses were simultaneously available. These increases were not

observed when the contingent responses were restricted but no contingency was in effect.

This finding suggests that contingent access to a single high-probability response can

serve as reinforcement for socially significant behavior in a complex, naturally occurring

environment even with concurrently available access to multiple other potentially

competing responses.














CHAPTER 5
GENERAL DISCUSSION

With the advent of preference assessment methodology, the empirical pre-treatment

identification of reinforcers for use in clinical application has become standard in the

field of applied behavior analysis (Hanley, Iwata, Lindberg, & Conners, 2003) and the

utility of these methods in identifying reinforcers that are effective components of

behavioral treatments has been demonstrated numerous times. Typical empirical

reinforcement identification methods involve a relatively simple response being exposed

to contingent access to a number of items or activities in an environment in which

potentially competing responses are largely unavailable. The effectiveness of reinforcers

identified in these preparations in application depends upon their generality extending

across all of the variables that differ between the assessment and the application

environments. Research on the response-deprivation hypothesis suggests an alternative

method of reinforcer identification. Baseline levels of responding are measured in the

treatment environment and access to a consequence response is provided contingent on

an instrumental response such that the ratio of instrumental to consequence responses is

greater than that observed during baseline. The response-deprivation hypothesis has been

shown to effectively identify reinforcers and mitigate to some degree concerns regarding

the circularity of post hoc reinforcer identification (Timberlake & Farmer-Dougan, 1991).

Furthermore, as baseline levels of responding are measured in the environment in which

the reinforcement contingency will be put into place, there is less risk that the generality

of the identified reinforcers will not extend into the treatment context. The use of the









response-deprivation hypothesis for reinforcer identification would seem to have

particular promise for use in naturally occurring environments in which access to many

responses is available and in which the multiple sources of reinforcement maintaining

these responses may compete with reinforcers identified in more restricted environments.

Much research in this area, however, has occurred in the context of restricted operant

preparations in which only a limited number of responses are available. The present

series of studies extends prior work on response restriction, the Premack principle, and

the response-deprivation hypothesis by explicitly examining the use of response

restriction and contingent access to responses to increase socially desirable behavior in

complex, naturally occurring, multi-response environments.

In Studies 1 and 2, the effects of response restriction alone were examined.

Previous work has demonstrated that the restriction of access to behavior, in the absence

of contingencies, can result in an increase in other behavior in both restricted- (Hanley et

al., 2000) and multi-response (Bernstein & Ebbesen, 1978; Lyons & Cheney, 1984;

Green & Striefel, 1998) environments. In previous studies, however, such increases in

behavior have been idiosyncratic across subjects, with no single rule adequately

describing response reallocation following the restriction of a response. The results of

the present studies replicated this finding. The brief response restriction analyses

conducted in Study 1 produced results that were both variable across participants and

across replications within participant.

Under conditions of response restriction in Study 2, with no response contingency

in place, increases in targeted responses were observed for three of four subjects (Kyle,

John, and Beth). For Kyle, reliable and sustained increases were observed in both









reading and math. For John, reliable initial increases were observed in both reading and

math, however, these increases were not sustained over extended periods of time. For

Beth, increases in both math and reading, among other responses, were observed

following restriction of competing responses, however, these increases were not reliable

or predictable. One potentially important finding was the emergence of reasonably

desirable responses that were not initially targeted for increase. For Kyle, an increase

was observed in drawing following restriction of television and video games. For Beth,

working on craft projects increased with restriction of television, counting money, and

playing with clay. For Sarah, when television and movies were restricted, increases were

observed in working on crafts, listening to music, and reading magazines. These findings

suggest that the restriction of perhaps a less desired response (e.g., watching television)

may result in the serendipitous increase in a more desired response (e.g., drawing) in the

absence of imposed contingencies. Although the particular outcome of response

restriction was idiosyncratic across participants, these data imply response restriction as a

potentially effective method for use in multi-response environments. For example, at

time when school work is not necessarily required, response restriction might be used to

produce a greater diversity of leisure activities.

When effective, response restriction may provide a treatment option for clinicians

or parents that is both relatively easy to implement and not prone to failures in treatment

integrity (i.e., response restriction does not require counting of responses, timing, or the

delivery of stimuli leaving relatively little room for error). The general methods of

Studies 1 and 2 may serve as an initial guide for clinicians attempting to make use of

response restriction as a treatment, Kyle's outcome is an example of the most favorable









result. Although the brief analyses of Study 1 produced variable outcomes, they were

successful in identifying some responses that may have been competing with targeted

behavior. This may provide a starting point for a more extended analysis which might

identify further responses that may be competing with target behavior.

The factors responsible for changes in responding under response restriction remain

to be identified. It may be that responses that increase in frequency are members of the

same operant class as those restricted and might be considered members of a response-

class hierarchy (Harding, Wacker, Berg, Barretto, Winborn, & Gardner, 2001; Lalli,

Mace, Wohm, & Livezy, 1995; Richman, Wacker, Asmus, Casey, & Andelman, 1999).

Alternatively, it may be that responses that increase in frequency are either automatically

or socially reinforced and that restricted responses produced more potent reinforcement

and thus competed with them. A third possibility is that some responses may never

increase in frequency as a function of restriction of alternative responses alone (i.e., those

that are not in a participant's repertoire or those that are not automatically or socially

reinforced). Sources of control that may produce variability in behavior under conditions

of response restriction have also not yet been identified. It is possible, for example, that

fluctuations in motivating operations, the presence or absence of relevant discriminative

stimuli, or changing contingencies on any number of responses may affect the outcome of

restriction of access to responses in multi-operant environments.

In Study 3, the effects of contingent access to responses on socially important

behavior in naturally occurring, unrestricted environments were examined. Both the

Premack principle and the response-deprivation hypothesis would predict increases in

instrumental responding given the contingencies arranged in Study 3 (i.e., more probable









responses were made contingent upon the occurrence of less probable responses in a

manner that would make the ratio of instrumental to consequence responses greater than

that observed in free-operant phases). However, both the Premack principle and the

response-deprivation hypothesis are silent with regard to any possible effects of several

alternative sources of reinforcement being available. In Study 3, contingent access to

multiple high-probability responses and a single high-probability response arranged such

that the ratio of instrumental to consequence responses was greater than that observed

during free-operant periods produced reinforcement effects for all participants. This

finding further extends the generality of the Premack principle and the response-

deprivation hypothesis, showing reinforcement effects for socially desirable behavior in

complex, naturally occurring, unrestricted environments.

The effectiveness of the methods used in Study 3 suggests some promise for

general application in multi-response settings. The initial consequence responses were

chosen based on the outcomes of Studies 1 and 2, but it may not have been necessary to

conduct both studies to identify effective contingent responses. The single consequence

response effective as a reinforcer for Kyle was playing video games, playing on a

computer for John, and watching television for Sarah and Beth. The results from the

free-operant phases of the brief response restriction analyses, conducted in Study 1,

indicated that the most probable responses for Kyle were video game playing and math,

playing on the computer for all assessments for John, watching television during two

assessments and math during one for Sarah, and watching television during two

assessments and weaving during one for Beth. When the data obtained during free-

operant periods in Study 1 are aggregated (i.e., the durations of each response in each of









the free-operant periods are added), the most likely free-operant response observed in

Study 1 is the same response that was demonstrated to maintain targeted behavior in

Study 3 for all subjects other than Kyle. For Kyle, only two brief restriction analyses

were conducted and playing video games was the most likely response in one of the two.

Whether or not playing video games would have emerged as the most likely response in

further brief free-operant periods is unknown, but seems likely based on subsequent data.

These data, for three of four participants, suggest that effective contingent responses for

use as reinforcers may be identified through relatively brief free-operant observation

periods in the targeted environment. Furthermore, the most probable response observed

in the initial free-operant phases in Study 2 for all participants was the response shown to

serve as an effective reinforcer in Study 3, again suggesting that a pre-treatment measure

of free-operant responding in a particular environment may be useful in identifying

responses to use as reinforcers.

Research is needed in numerous areas to determine the conditions under which

response restriction and contingencies will be most effective. First, because the current

studies and previous research have found response restriction to produce idiosyncratic

results, research is needed to examine the influences of potentially relevant variables

including motivating operations, response effort, the presence or absence of

discriminative stimuli, and contingencies for alternative responses under conditions of

response restriction. Second, because both the Premack principle and the response-

deprivation hypothesis predict the obtained reinforcement effect in Study 3, evaluations

are needed using a less-probable response with the ratio of instrumental to consequence

responses being greater than that observed during free-operant periods (the two models






59


make different predictions). Third, future research is needed to directly compare the

effectiveness of Premack/response-deprivation approaches to identifying reinforcers to

those based on relatively standard preference assessment methods (e.g., DeLeon & Iwata,

1996; Fisher, Piazza, Bowman, Hagopian, Owens, & Slevin, 1992; Pace, Ivancic,

Edwards, Iwata, & Page, 1985; Roane, Vollmer, Ringdahl, & Marcus, 1998; Windsor,

Piche, & Locke, 1994). The outcome of such a comparison would speak directly to the

applied utility of using contingent access to behavior observed in a treatment

environment as reinforcement.












LIST OF REFERENCES


Allison, J. & Timberlake, W. (1974). Instrumental and contingent saccharin licking in
rats: Response deprivation and reinforcement. Learning and Motivation, 5, 231-
247.

Amari, A., Grace, N. C., & Fisher, W. W. (1995). Achieving and maintaining compliance
with the ketogenic diet. Journal ofApplied Behavior Analysis, 28, 341-342.

Bernstein, D. J. (1998). Establishment of a laboratory for continuous observation of
human behavior. In K. A. Lattal & M. Perone (Eds.), Handbook of research
methods in human operant behavior (pp. 509-539). New York: Plenum Press.

Bernstein, D. J. & Ebbesen, E. B. (1978). Reinforcement and substitution in humans: A
multiple-response analysis. Journal of the Experimental Analysis ofBehavior, 30,
243-253.

Bernstein, D. J. & Michael, R. L. (1990). The utility of verbal and behavioral assessments
of value. Journal of the Experimental Analysis ofBehavior, 54, 173-184.

Catania, A. C. (1998). Learning. Upper Saddle River, NJ: Prentice Hall.

Charlop, M. H., Kurtz, P. F., & Casey, F. G. (1990). Using aberrant behaviors as
reinforcers for autistic children. Journal ofApplied Behavior Analysis, 23, 163-
181.

DeLeon, I. G. & Iwata, B. A. (1996). Evaluation of a multiple-stimulus presentation
format for assessing reinforcer preferences. Journal ofAppliedBehavior Analysis,
29, 519-532.

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.

Green, G. & Striefel, S. (1988). Response restriction and substitution with autistic
children. Journal of the Experimental Analysis ofBehavior, 50, 21-32.

Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of problem
behavior: A review. Journal ofApplied Behavior Analysis, 36, 147-185.

Hanley, G. P., Iwata, B. A., Thompson, R. H., & Lindberg, J. S. (2000). A component
analysis of "stereotypy as reinforcement" for alternative behavior. Journal of
Applied Behavior Analysis, 33, 285-297.











Harding, J. W., Wacker, D. P., Berg, W. K., Barretto, A., Winborn, L., & Gardner, A.
(2001). Analysis of response class hierarchies with attention-maintained problem
behaviors. Journal ofApplied Behavior Analysis, 34, 61-64.

Homme, L. E., de Baca, P. C., Devine, J. V., Steinhorst, R., & Rickert, E. J. (1963). Use
of the Premack principle in controlling the behavior of nursery school children.
Journal of the Experimental Analysis ofBehavior, 6, 544.

Konarski, E. A., Jr. Johnson, M. R., Crowell, C. R., & Whitman, T. L. (1980). Response
deprivation and reinforcement in applied settings: A preliminary analysis. Journal
ofApplied Behavior Analysis, 13, 595-609.

Lalli, J. S., Mace, F. C., Wohn, T., & Livezy, K. (1995). Identification and modification
of a response-class hierarchy. Journal ofApplied Behavior Analysis, 28, 551-559.

Laraway, S., Snycerski, S., Michael, J., & Poling, A. (2003). Motivating operations and
terms to describe them: Some further refinements. Journal ofApplied Behavior
Analysis, 36, 407-414.

Lyons, C. A. & Cheney, C. D. (1984). Time reallocation in a multiresponse environment:
Effects of restricting response classes. Journal of the Experimental Analysis of
Behavior, 41, 279-289.

Meehl P. E. (1950). On the circularity of the law of effect. Psychological Bulletin, 47, 52-
75.

Mitchell, W. S. & Stoffelmayr, B. E. (1973). Application of the Premack principle to the
behavioral control of extremely inactive schizophrenics. Journal of Applied
Behavior Analysis, 6, 419-423.

Osborne, J. G. (1969). Free-time as a reinforcer in the management of classroom
behavior. Journal ofApplied Behavior Analysis, 2, 113-118.

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 ofApplied Behavior Analysis, 18, 249-255.

Postman, L. (1947). The history and present status of the law of effect. Psychological
Bulletin, 44, 489-563.

Premack, D. (1959). Toward empirical behavior laws: 1. Positive reinforcement.
Psychological Review, 66, 219-233.










Richman, D. M., Wacker, D. P., Asmus, J. M., Casey, S. D., & Andelman, M. (1999).
Further analysis of problem behavior in response class hierarchies. Journal of
Applied Behavior Analysis, 32, 269-283.

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.

Sidman, M. (1960). Tactics of scientific research. New York: Basic Books.

Skinner, B. F. (1935). The generic nature of the concepts of stimulus and response.
Journal of General Psychology, 12, 40-65.

Skinner, B. F. (1953). Science and human behavior. New York: Macmillan.

Timberlake, W. & Allison, J. (1974). Response deprivation: An empirical approach to
instrumental performance. Psychological Review, 81, 146-164.

Timberlake, W. & Farmer-Dougan, V. A. (1991). Reinforcement in applied settings:
Figuring out ahead of time what will work. Psychological Bulletin, 110, 379-391.

Windsor, J., Piche, L. M., & Locke, P. A. (1994). Preference testing: A comparison of
two presentation methods. Research in Developmental Disabilities, 15, 493-455.















BIOGRAPHICAL SKETCH

Jason Bourret was born in Milwaukee, Wisconsin. He entered the University of

Florida in 1995 and graduated in 1999 with a degree in psychology. During his

undergraduate studies, he served as a volunteer in a basic behavior analysis research lab

working on choice research and two applied behavior analysis labs working primarily on

research focusing on the treatment of severe behavior problems in individuals diagnosed

with developmental disabilities. In 1999, he enrolled in the behavior analysis program at

the University of Florida to pursue a doctoral degree in psychology. He worked as a

graduate research assistant under the direction of Dr. Timothy Vollmer. Over the course

of his graduate career at the University of Florida, he has been involved in research on

verbal behavior, quantitative analysis of behavior, response allocation in multi-response

environments, descriptive analysis of behavior, and schedules of response-contingent and

fixed-time reinforcement delivery. Expecting graduation in December, 2005, he plans on

continuing his career in behavior analysis through conducting behavior analytic research

with implications for application in clinical settings.