Citation
Evaluating the Role of Extinction in a Human Operant Model of Resurgence

Material Information

Title:
Evaluating the Role of Extinction in a Human Operant Model of Resurgence
Creator:
Laureano, Brianna
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (37 p.)

Thesis/Dissertation Information

Degree:
Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Psychology
Committee Chair:
DE LEON,ISER G
Committee Co-Chair:
DALLERY,JESSE
Committee Members:
DEVINE,DARRAGH P

Subjects

Subjects / Keywords:
extinction -- human-operant -- resurgence
Psychology -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Psychology thesis, M.S.

Notes

Abstract:
Operant resurgence of problem behavior may occur following the discontinuation of differential reinforcement of alternative behavior (DRA). One possible explanation is that dense schedules of reinforcement for the alternative response and the ensuing shift in response allocation make it unlikely that the target response fully contacts extinction. We examined this possibility in an analog model of problem behavior with college students. Some participants experienced the conventional ABC resurgence paradigm (A = reinforcement of target behavior, B = differential reinforcement of problem behavior, C = extinction of both responses; the resurgence test) Other participants experienced a four-phase sequence in which extinction of the target response preceded DRA. If resurgence occurs because the target response is not fully extinguished in the conventional paradigm, we would expect less resurgence when extinction is implemented before DRA (i.e., the four-phase participants). The results confirm this, specifically if the target response showed clear evidence of extinction during the initial extinction phase. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (M.S.)--University of Florida, 2018.
Local:
Adviser: DE LEON,ISER G.
Local:
Co-adviser: DALLERY,JESSE.
Statement of Responsibility:
by Brianna Laureano.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
LD1780 2018 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

EVALUATING THE ROLE OF EXTINCTION IN A HUMAN OPERANT MODEL OF RESURGENC E By BRIANNA LAUREANO A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2018

PAGE 2

2018 Brianna Laureano

PAGE 3

To my Mom and Dad, without whom I definitely would have dropped out by now

PAGE 4

4 ACKNOWLEDGMENTS I am deeply thankful for the people in my life professionally and personally who have supported and guided me every step of the way. It has taken a very large village to get me to this point I would like to extend my gratitude to my advis or, Dr. Iser G. DeLeon, for his continued guidance in this program and in this field. I have learned so much from you committee members, Dr. Dallery and Dr. Devine, for their advisement and support throughout this process. Most importantly, I would like to thank my amazing friends and family, without indebted I am to those who have sho wn me love and support extends beyond any project or program. And finally, I am not sure I can truly thank my parents enough. For graduated, thank you from the bottom of my heart, Mom and Dad.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF FIGURES ................................ ................................ ................................ .......... 6 ABSTRACT ................................ ................................ ................................ ..................... 7 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ...... 8 2 METHOD ................................ ................................ ................................ ................ 13 Participants ................................ ................................ ................................ ............. 13 Apparatus ................................ ................................ ................................ ............... 13 Instructions ................................ ................................ ................................ ............. 14 Procedure ................................ ................................ ................................ ............... 15 3 RESULTS ................................ ................................ ................................ ............... 18 4 DISCUSSION ................................ ................................ ................................ ......... 30 APPENDIX POST SESSION SURVEY ................................ ................................ ............................ 34 LIST OF REFERENCES ................................ ................................ ............................... 35 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 37

PAGE 6

6 LIST OF FIGURES Figure page 3 1 Participants who were assigned to the Ext Firs t group and met the extinction criteria. ................................ ................................ ................................ ................ 22 3 2 Participants who were assigned to the Ext First group and did not meet the extinction criteria. ................................ ................................ ................................ 23 3 3 Participants who were assigned to the DRA 24 group. ................................ ....... 24 3 4 Participants who were assigned to the DRA 12 group ................................ ....... 25 3 5 Average target responses per minute across each phase. ................................ 26 3 6 I ndividual target responses per minute in the last min of DRA and the first min of EXT EXT ................................ ................................ ................................ 27 3 7 Average rates of target responding across BL and each min of EXT EXT.. ....... 28 3 8 Average rates of alternative responding across each min of EXT EXT. ............ 29

PAGE 7

7 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EVALUATING THE ROLE OF EXTINCTION IN A HUMAN OPERANT MODEL OF RESURGENC E By Brianna Laureano De cember 2018 Chair: Iser G. DeLeon Major: Psychology Operant r esurgence of problem behavior may occur following the discontinuation of differential reinforcement of alternative behavior (DRA). One possible explanation is that dense schedules of reinforcem ent for the alternative response and the ensuing shift in response allocation make it unlikely that the target response fully contacts extinction. We examined this possibility in an analog model of problem behavior with college students. Some participants experienced the conventional ABC resurgence paradigm (A = reinforcement of target behavior, B = differential reinforcement of problem behavior, C = extinction of both responses; the resurgence test) Other participants experienced a four phase sequence in w hich extinction of the target response preceded DRA. If resurgence occurs because the target response is not fully extinguished in the conventional paradigm, we would expect less resurgence when extinction is implemented before DRA (i.e., the four phase pa rticipants). The results confirm this, specifically if the target response showed clear evidence of extinction during the initial extinction phase

PAGE 8

8 CHAPTER 1 INTRODUCTIO N Resurgence refers to the reemergence of a previously extinguished response w hen reinforcement for an alternative response is discontinued or lessened. (Epstein, 1983; Rawson et al., 1977). This behavioral phenomenon is typically examined in a three phase sequence (e.g. Lieving & L attal, 2003; Marsteller & St. Peter, 2012; Leitenberg et al., 1970). In Phase 1, a target response is reinforced. In Phase 2, the target response is placed on extinction and an alternative response is reinforced. In Phase 3, both behaviors are placed on pr ogrammed extinction. Resurgence is said to occur if the target response reemerges during Phase 3 at a level higher than that observed at the end of Phase 2. Recent interest in resurgence stems partly from its adoption as a model of relapse following succe ssful intervention. For example, the treatment of severe behavior problems of individuals with intellectual disabilities generally begins by identifying the consequences that maintain problem behavior (i.e. a functional assessment). Individuals are then ta ught to emit an appropriate, alternative response to produce that consequence while the target response is placed on extinction. However, if the reinforcement of the alternative response then ceases for any of a variety of potential reasons (e.g. a teacher is attending to other children in the classroom), problem behavior may be expected to resurge (i.e. treatment relapse). Thus, resurgence of severe problem behavior, such as aggression or self injury, could lead to treatment setbacks and may put the client and others at risk for physical harm. Identifying the variables that prevent the resurgence of problem behavior when periods of extinction occur may help mitigate harm that can be caused by treatment integrity errors

PAGE 9

9 Researchers have offered several mecha nisms to account for the resurgence phenomenon. One account involves extinction induced variability (Sweeney and Shahan, 2016). It is possible that the increase in target responding is attributable to the response variability the typically occurs during ex tinction. Thus, some researchers have observed that increases in responding during extinction are not necessarily limited to the previously reinforced response; other available response that were never reinforced may increase during the final extinction ph ase (e.g. Bolvar et al., 2017; Sweeney and Shahan, 2016). Alternatively, Winterbauer, Lucke, and Bouton (2013) suggested a context change hypothesis to explain resurgence, which proposes that resurgence occurs when the organism fails to generalize the lea rning of the extinction contingency of the target response from the treatment context to a context where alternative reinforcement is no longer available. Moreover, resurgence may also be accounted for with behavioral momentum theory (Shahan and Sweeney, 2 011). In Phase 2, reinforcement of the alternative response is paired with extinction of the target response. Reinforcement of the alternative response disrupts the target response but may also increase the density of reinforcement in that context which, a ccording to BMT, could perversely strengthen the target response. In Phase 3, the disruptor is removed when reinforcement for the alternative response is discontinued and the target response resurges. Lattal and St. Peter Pipkin (2009) offered two additio nal reasons why resurgence may occur. First, it is possible that during the third phase, the absence of reinforcement during Phase 2, the context for emitting the alt ernative response is the withholding of

PAGE 10

10 reinforcement for the target response. The absence of reinforcement becomes a reinforcement is withheld for the alternative resp onse in Phase 3, the subject switches over and resurgence is observed. Second, resurgence may occur because the target response does not sufficiently contact extinction during Phase 2 (i.e. the prevention of extinction hypothesis; Rawson et al., 1977). A d ense schedule of reinforcement for the alternative response in Phase 2 results in rapid re allocation of behavior towards the alternative response, such that the organism experiences little exposure to the absence of reinforcement for the target response. Therefore, during the resurgence test, when reinforcement of the alternative response is withheld, the target response reemerges because it was insufficiently extinguished. This final hypothesis has been evaluated in studies with non human subjects (e.g. Cleland, Foster, & Temple, 2000; Leitenberg et al., 1975). Cleland et al. tested the prevention of extinction hypothesis with hens by programming the extinction of a target response prior to implementing DRA, thus creating a four phase sequence. After repe ated exposure to the four phase sequence, a standard three phase sequence without the additional extinction phase was programmed and levels of resurgence were compared between the two sequences. Greater resurgence was observed when exposure to extinction w as not programmed before DRA. Therefore, the results support the hypothesis that DRA may not sufficiently expose non human subjects to the extinction of the target behavior. However, a possible confound of Cleland et al. (2000) is that it was a within subj ect comparison using repeated exposures to each sequence. For each subject,

PAGE 11

11 there were a total of 8 replications of the four phase sequence and only 2 replications of the three phase sequence. The authors also did not control for the length of the conditio ns, which led to differences in exposure to extinction and reinforcement conditions. Research has shown that lower levels of resurgence are associated with either longer durations of extinction of the target behavior or longer durations of alternative rein forcement (e.g., Leitenberg et al., 1975; Exp. 4; Sweeney and Shahan, 2013a; Wacker et al., 2011 ). Finally, while Cleland et al. (2000) adds to the non human resurgence literature, there is currently no research evaluating the prevention of extinction hyp othesis with human subjects. As DRA is a common intervention for the treatment of problem behavior for humans, it is important to determine if DRA is truly extinguishing the target response or if it is simply reallocating responding in favor of the alterna tive. The purpose of the current study was to evaluate the prevention of extinction hypothesis with human subjects in a controlled setting. We also wanted to control for the duration of exposure to extinction and treatment conditions when comparing levels of resurgence to ensure our results were not confounded by extraneous variables such as uneven exposure to DRA. Again, if resurgence is influenced by limited exposure to extinction for the target response, one might expect differences in resurgence levels between the conventional ABC paradigm and an alternative, in which the subject fully experiences extinction of the target response. The latter can be accomplished by arranging extinction of the target response, in isolation, prior to reinforcing the alter native response, thus obviating the putative effects of rapid response reallocation. Then, during the resurgence test, we would expect lower levels of the target response when

PAGE 12

12 extinction is implemented before DRA. We tested this possibility using a compute rized model of resurgence with college students.

PAGE 13

13 CHAPTER 2 METHOD Participants T hirty six undergraduate students participated, ranging in age from 18 to 20 years. Twenty nine participants identified as female and seven identified as male. All participants were enrolled in an introductory psychology course and received course credit for their participation. Participants could also earn points based on performance; at the end of the session, points were exchangeable for $0.01 each. All participan ts were paid in cash and they earned a range of $8.65 to $1 8.78, with an average of $13.26 Apparatus Sessions were conducted individually with each participant. Sessions were held in a relatively barren room containing a table, chair, laptop, and USB mou se. All conditions of the experimental tasks were coded using Visual Basic 2013 Community Edition, which controlled the experiment and recorded the results. Throughout each phase, the computer screen displayed two squares, one blue and one orange, and a cu mulative point score. Bot h squares had areas of 3.61 cm and moved in circular patterns at speeds of 6.25 pixels per sec. Clicking on the blue square was analogous to emitting the target behavior and clicking on the orange square was analogous to emitting t he alternative behavior. Participants could click either of the squares throughout each phase; however, points were only delivered for clicking the blue square during baseline and the orange square during DRA. When points were delivered, the session time paused, and a new screen

PAGE 14

14 collection response was programmed to model a consumption response that occurs in an applied setting when reinforcers are delivered. For instance, when a clinician delivers food contingent upon an appropriate response, the client must pick it up and consume it by physically chewing and swallowing the food item. Since our reinforcers were tokens delivered electronically, clicking to collect the token served as the consummatory response. Once participants collected the point, the session resumed, the moving icons appeared on the screen again, and the session time resumed. The program also included programmed phase changes. At the end of each phase, the session time paused, and instructions appeared on the computer screen saying the current phase had ended and a new one would begin. At the bottom of the new screen was an ico icon, the new phase would begin and the session time would resume Instructions After signing the consent form, participants were given the following instructions: be working. Your task will be completed on this laptop. You will only need to use the mouse provided to earn as many points as possible. For verification, please move the mouse so I can ensure that it is working properly. This study is credit based, so yo u must stay for the entire hour to earn your full research credits. However, we do offer a chance to earn money. For each point you earn, you will also earn $0.01. After I leave the room, the instructions on the screen will be the only ones you receive thr oughout the session. It is up to you to figure out how to earn points and I will not be able to help you earn points at any time during your session. How you respond is completely up to you and you may stop responding at any time. Should you stop respondin g, I just ask that you keep the program on the screen. If you close out of

PAGE 15

15 the program, we lose all the data and it stops your session time completely. As well, you will know when your session is over because the screen will minimize on its own; when this occurs, please come get my attention and I can get you started on the final part of the room until the task was complete Procedure A between subjects design was used to evaluate resurgence with and without exposure to extinction prior to reinforcing an alternative response. Participants were assigned to one of three groups: EXT Before, DRA 24, or DRA 12. All procedures were held const ant across the groups with the following exceptions. First, the EXT Before group was exposed to extinction prior to beginning DRA, and thus completed the experiment in four phases (BL, EXT, DRA, and EXT EXT). By contrast, the DRA 24 and DRA 12 groups exper ienced extinction only as it was embedded into DRA and thus completed the experiment in the conventional three phase resurgence sequence (BL, DRA, EXT EXT). The second difference was the duration of DRA. For the EXT Before and DRA 12 groups, the DRA phase was 12 min. For the DRA 24 group, the DRA phase was 24 min. The rationale for the differing lengths of DRA is described below. Sixteen participants were assigned to EXT Before, ten to DRA 24, and ten to DRA 12. All participants completed the experiment d uring one session. Participants in the EXT Before group completed all 4 phases in a total of 48 minutes. Participants in the DRA 24 and DRA 12 groups completed their sessions in 48 min and 36 min, respectively

PAGE 16

16 Baseline During baseline, participants earned one point on a fixed ratio (FR) 1 schedule when they emitted the target response (i.e., clicked on the blue square). Emitting the alternative response (i.e., clicking on the orange square), resulted in no programmed consequences. All gr oups were exposed to baseline, which was 12 min in duration EXT During EXT, the target response was placed on extinction. Thus, no points were delivered for emitting either response. Only the EXT Before group was exposed to EXT. This phase was 12 min in duration Prior research had shown that repeated exposure to extinction conditions for the target response may lead to lower levels of resurgence (e.g., Sweeney & Shahan, 2013a; Wacker et al., 2011). In the current experiment, extinction of the target res ponse was embedded in the EXT, DRA, and EXT EXT phases. Therefore, if participants were sufficiently contacting the extinction contingency of DRA, they would experience a continuous exposure to extinction, rather than repeated exposure. DRA During DRA, th e target response was placed on extinction. Participants earned points emitting the alternative response on an FR 1 schedule. All groups were exposed to DRA, but the duration of this phase differed across groups as mentioned above. For the EXT Before and DRA 12 groups, the DRA phase was 12 min. For the DRA 24 group, the DRA phase was 24 min. This procedural manipulation was included for two reasons. First, it was important to compare resurgence while controlling for the time that the target response was n ot reinforced. To do so, the DRA phase of the DRA 24 group equaled the total time of the EXT and DRA phases of the EXT Before group and, thus, the entire

PAGE 17

17 experiment lasted 48 min for each group. Second, it was important to compare resurgence while addition ally controlling for the time the alternative response was reinforced. Longer exposure to reinforcement of the alternative response may lead to less resurgence of the target response (Leitenberg et al., 1975). To control for exposure to treatment condition s, the DRA 12 group was exposed to the same length of time in DRA (12 min) as the EXT Before group and, thus, the entire experiment lasted 36 min for this group EXT EXT During EXT EXT, no points were delivered for either response. This phase was the res urgence test for all participants. All groups were exposed to EXT EXT for 12 min Post session. When the program ended, the screen automatically closed, and participants were instructed to gain the attention of the experimenter. Then, the experimente report data on the following information: (a) age, (b) gender, (c) year in college, (d) major, (e) what the participant thought the experiment was designed to determine, (f) what strategies the par ticipant used in deciding which shapes to click during each experiment, (g) any other thoughts the participant had on the study. A copy of post session survey is in Appendix A.

PAGE 18

18 CHAPTER 3 RESULTS This study was an attempt to isolate the effects of prior ex tinction on resurgence; therefore, we conducted separate analyses of the EXT First group for those participants whose responding during EXT was consistent with an extinction effect. An extinction effect was defined as a greater than 80% decrease of target responding in the last 5 min of EXT relative to the last 5 did not meet this criterion. We continued to recruit participants until we had 10 participants in the EXT First group that met the criteria. I n the following discussion, EXT First All will refer to all 16 participants in the group and EXT First Ext Only will refer to the 10 participants in the group that met the extinction criterion. Figure 3 1 shows the individual data for the participants who were assigned to the Ext First group and met the extinction criteria. On average, target responding decreased and remained at low levels during the EXT and EXT EXT phases. Figure 3 2 shows the individual data for the participants who were assigned to the Ext First group and did not meet the extinction criteria. On average, target responding remained high or variable during the EXT and EXT EXT phases. Figure 3 3 shows the individual data for the participants who were assigned to the DRA 24 group. Figure 3 4 shows the individual data for the participants who were assigned to the DRA 12 group. On average, target responding remained high and/or variable across the EXT EXT phase for both DRA 24 and DRA 12 groups. Figure 3 5 shows the average target responses pe r minute across all phases. On average, when reinforcement was available (during BL and DRA conditions), participants in each group engaged in the target response at similar rates; therefore,

PAGE 19

19 differences observed during the resurgence test were not the res ult of a difference in the magnitude of responding across groups. For example, it is not the case that the EXT Only group showed a lower average resurgence than the DRA 24 group because the dless of the contingency in effect. During EXT EXT, both DRA 24 and DR A 12 groups emitted more responses on average (12.93 and 19.37, respectively) than did the EXT First All and EXT First Only groups (6.19 and 1.15, respectively). Thus, as anticipated, p articipants who experienced extinction before DRA showed less resurgence than those who only experienced extinction embedded with DRA. In addition, participants in the EXT First group that appeared sensitive to the extinction contingency of the second phas e, engaged in fewest target responses during the resurgence test. We conducted a one way analysis of variance (ANOVA) to determine if there were statistically significant differences in the average number of target responses during EXT EXT between the EXT First EXT Only, DRA 24, and DRA 12 groups. The one way ANOVA indicated a significant difference across groups at the .05 significance level ( F ( 2, 27) = 8.816, p = .001 ). In the recent resurgence literature, resurgence has been defined as any increase of the target response from the last minute of DRA to the first minute of EXT EXT (e.g., Sweeney & Shahan, 2016). By this definition, all participants demonstrated resurgence of the target behavior. To show the within group responding, Figure 3 6 shows indivi dual participant target responses in the last minute of DRA and the first minute of EXT EXT. The range of target responses in the first minute of EXT EXT were 1 to 83 for the EXT First All group, 1 to 15 for the EXT First EXT Only group, 11 to 53 for the D RA

PAGE 20

20 24 group, and 9 to 40 for the DRA 12 group. The average rate of target responses in the first minute of EXT EXT were 20.9 for the EXT First All group, 7.7 for the EXT First EXT Only group, 25.9 for the DRA 24 group, and 24.1 for the DRA 12 group. A one way ANOVA was calculated to compare the average levels of target responding in the first minute of EXT EXT between the EXT First Ext Only, DRA 24, and DRA 12 groups. There was a statistically significant difference between groups at the .05 significance le vel ( F (2, 27) = 7.926, p = .002 ). Reporting the rate of responding in the first minute or session of a resurgence test is important because it shows how responding increases immediately following the removal of alternative reinforcement. This quick increas e demonstrates how fleeting the extinction effects of the treatment might be. In a clinical context, it also emphasizes that failing to implement treatment may place the client at risk. However, when analyzing resurgence, the initial target rate is not the only important measure to consider. Another important factor to consider if resurgence occurs is how long resurgence effects persist. In other words, after the first minute of the EXT EXT phase, does the target response return to the same near zero rates observed in DRA or does it return to rates similar to BL? Figure 3 7 shows the mean rates of target responding across BL and across each minute of EXT near zero rates by the second minute of EXT EXT. Neither the DRA 24 nor DRA 12 d a similar return to the levels observed during DRA The same group patterns were observed for the alternative response. Figure 3 8 shows the mean rates of alternative responding across each minute of EXT EXT. In the EXT First

PAGE 21

21 group, alternative behavior decreased by the second minute of the phase; however, alternative behavior persisted at moderate rates for the DRA 24 and DRA 12 group s

PAGE 22

22 Figure 3 1 Participants who were assigned to the Ext First group and met the extinction criteria.

PAGE 23

23 Figure 3 2 Participants who were assigned to the Ext First group and did not meet the extinction criteria

PAGE 24

24 Figure 3 3 Participants who were assigned to the DRA 24 group.

PAGE 25

25 Figure 3 4 Participants who were assigned to the DRA 12 group

PAGE 26

26 Figure 3 5 Average target responses per minute across each phase Error bars repre sent standard error of the mean

PAGE 27

27 Figure 3 6 Individual target responses per minute in the last min of DRA and the first min of EXT EXT

PAGE 28

28 Figure 3 7 Average rat es of target responding across BL and each min of EXT EXT. Error bars represent standard error of the mean

PAGE 29

29 Figure 3 8 Average rates of alternative responding across each min of EXT EXT. Error bars represent standard error of the mean

PAGE 30

30 CHAPTER 4 DISCUSSION In the present study, we sought to determine if exposing participants to extinction prior to introducing DRA would lessen resurgence. We observed that, on average, resurgence was lower when target behavior was extinguished prior to implementing DRA. This suggests that the reinforcement contingency of DRA may prevent target behavior from sufficiently contacting the extinction contingency of DRA. Therefore, our results support the prevention of extinction hypothesis. However, recent e xaminations of resurgence in human participants have added an inactive control response to determine whether the effects could be better characterized as extinction induced variability (e.g. Bolvar et al., 2017; Sweeney and Shahan, 2016). If we had includ ed a control response, perhaps we would have observed greater increases in the control response than we observed in the target or alternative responses in the EXT First group. Such a finding would be consistent with the first explanation offered by Lattal and St. Peter Pipkin (2009) that the change in conditions is discriminative for the behavior of switching to something new. In this case, switching to something new would be switching over to a response that is not associated with a history of reinforcemen t. The current study was designed as a simple, proof of concept, operant model to test one theory regarding the mechanisms that contribute to resurgence. Obviously, there are numerous differences between the procedures described and the actual clinical sp aces in which one might be concerned about resurgence. Some important differences include the short experimental sessions, the similarity between the target and the alternative responses, and the brief history of reinforcement of both responses.

PAGE 31

31 Further re search is needed to determine if the present findings would be observed in a clinical population with actual clinical targets. Moreover, implementing extinction in isolation was a mechanism to evaluate the validity of the prevention of extinction hypothes is; it was not intended to serve as a clinical solution or recommendation to mitigate the reemergence of problem behavior. Although implementing extinction before DRA resulted in less resurgence, implementing extinction without a reinforcement component ca n have detrimental effects. First, without a means of accessing the reinforcer, it may simply take longer for behavior to decrease with extinction alone than with other forms of intervention ( Petscher & Bailey, 2008). We examined this in our data by compar ing the average latency, in minutes, to extinction in the EXT phase of the EXT First groups and the DRA phase of the DRA 24 and DRA 12 groups. Extinction was defined as a greater than 80% decrease of target responding relative to the last 5 min of BL. The average latency to extinction for the EXT First All, Extinction First EXT Only, DRA 24, and DRA 12 groups were 4.1, 3.2, 3.8 and 4.4 min, respectively. These differences are only a few minutes and are representative of simple responses with short histories of reinforcement. More research target responses were problem behaviors with extended histories of reinforcement. Second, extinction bursts are more likely if extinctio n is not combined with some other means of accessing the reinforcer (Lerman and Iwata, 1995). We examined this possibility also in our data. For that analysis, an extinction burst was defined as a greater rate of responding in the first 5 min of EXT or EXT EXT relative to the last 5 min of BL. The number of extinction bursts observed for the EXT First All, EXT First EXT

PAGE 32

32 Only, DRA 24, and DRA 12 groups were 5, 1, 1, and 2, respectively. It is also worth noting that of the five extinction bursts observed in t he Ext First All group, four of them were emitted by individuals whose behavior never met the criteria for extinction. Nevertheless, our results showed that resurgence was more pronounced when extinction was embedded within DRA than when extinction was im plemented alone first. This implies that DRA may prohibit clients from sufficiently contacting extinction, even though the withholding of reinforcement is a programmed component of the treatment. Since implementing extinction alone may be detrimental, mor e research should be done to determine clinically acceptable ways to ensure sufficient contact with extinction when implementing DRA. Future studies could evaluate, for example, if implementing lean schedules of reinforcement during DRA would decrease the levels of resurgence observed during EXT EXT. Finally, resurgence is often a rather transient phenomenon. While behavior during the final extinction phase does increase relative to the rate of responding observed during DRA, it also tends to extinguish du ring this final phase. However, resurgence has important clinical implications. Resurgence is a phenomenon that occurs when alternative reinforcement is discontinued. In other words, a change agent (e.g. a caregiver or teacher) has failed to implement trea tment. A caregiver may observe transient resurgences and give up on the intervention because he/she has concluded that it is ineffective. Most importantly, if problem behavior has resurged it is likely to be reinforced, thus directly strengthening the beh avior. Therefore, even though resurgence may be a transient phenomenon, it may have significant effects beyond its

PAGE 33

33 short occurrence. Thus, it is important to continue to evaluate ways to mitigate or lessen resurgence

PAGE 34

34 APPENDIX POST-SESSION SURVEY

PAGE 35

35 LIST OF REFERENCES Bolvar, H. A., Cox, D. J., Barlow, M. A., & Dallery, J. (2017). Evaluating resurgence procedures in a human operant laboratory. Behavioural Processes 140 150 160. Cleland, B. S., Foster, T. M., & Temple, W. (2000). Resurgence: The role of extinction. Behavioural Processes 52 117 129. Epstein, R. (1983). Resurgence of previously reinforced behavior during extinction. Behaviour Analysis Letters 3, 391 397. Lattal, K. A., & St Peter Pipkin, C. (2009). Resurgence of previously reinforced responding: Research and application. The Behavior Analyst Today 10 254. Leitenberg, H., Rawson, R.A., & Bath, K. (1970). Reinforcement of competing behavior during extinction. Science 169, 301 303. Leitenberg, H., Rawson, R. A., & Mulick, J. A. (1975). Extinction and reinforcement of alternative behavi or. Journal of Comparative and Physiological Psychology, 88, 640 652 Lerman, D. C., & Iwata, B. A. (1995). Prevalence of the extinction burst and its attenuation during treatment. Journal of Applied Behavior Analysis 28 93 94. Lieving, G. A., & Lattal, K. A. (2003). Recency, repeatability, and reinforcer retrenchment: An experimental analysis of resurgence. Journal of the Experimental Analysis of Behavior, 80, 217 233. Cuyjet, M. J. (1997). African American men on college campuses: Their needs and percep tions. New Directions for Student Services, 80, 5 16. Marsteller, T. M., & St. Peter, C. C. (2012). Resurgence during treatment challenges. Revista Mexicana de Anlisis de la Conducta, 38, 7 23. Edwards, H. (2000). Crisis in Black athletes on the eve of t he 21st century. Society 37 (3), 9. Petscher, E. S., Rey, C., & Bailey, J. S. (2009). A review of empirical support for differential reinforcement of alternative behavior. Research in Developmental Disabilities, 30, 409 425. Rawson, R. A., Leitenberg, H., Mulick, J. A., & Lefebvre, M. F. (1977). Recovery of extinction responding in rats following discontinuation of reinforcement of alternative behavior: A test of two explanations. Animal Learning and Behavior, 4, 415 420. Shah an, T. A., & Sweeney, M. M. (2011). A model of resurgence based on behavioral momentum theory. Journal of the Experimental Analysis of Behavior, 95, 91 108.

PAGE 36

36 Sweeney, M.M., Shahan, T.A., 2013a. Behavioral momentum and resurgence:Effects of time in extinctio n and repeated resurgence t ests. Learn ing & Behav ior 41, 414 424. Sweeney, M. M., & Shahan, T. A. (2016). Resurgence of target responding does not exceed increases in inactive responding in a forced choice alternative reinforcement procedure in humans. Be havioural Processes 124, 80 92. Wacker, D. P., Harding, J. W., Berg, W. K., Lee, J. F., Schieltz, K. M., Padilla, Y. C., ... & Shahan, T. A. (2011). An evaluation of persistence of treatment effects during long term treatment of destructive behavior. Jour nal of the Experimental Analysis of Behavior, 96 261 282. Winterbauer, N. E., Lucke, S., & Bouton, M. E. (2013). Some factors modulating the strength of resurgence after extinction of an instrumental behavior. Learning and Motivation 44 60 71

PAGE 37

37 BIOGRAPHICAL SKETCH Brianna Laureano received a Bachelor of Science with a major in general psychology from The University of Florida in May 2016 She received a Master of Science in p sychology from The University of Flo rida in December 2018, where she continues as a doctoral student under the advisement of Dr. Iser G. DeLeon