Association for Behavior Analysis International

The Association for Behavior Analysis International® (ABAI) is a nonprofit membership organization with the mission to contribute to the well-being of society by developing, enhancing, and supporting the growth and vitality of the science of behavior analysis through research, education, and practice.


36th Annual Convention; San Antonio, TX; 2010

Event Details

Previous Page


Poster Session #409
EAB Poster Session 4
Monday, May 31, 2010
12:00 PM–1:30 PM
Exhibit Hall A (CC)
105. The Effects of Testing Procedures on Results of Transfer of Stimulus Function Tests
Area: EAB; Domain: Experimental Analysis
ERICA FEUERBACHER (University of North Texas), Jesus Rosales-Ruiz (University of North Texas)
Abstract: Transfer of stimulus function tests are widely employed to test for stimulus classes and equivalence relations. However, the procedures for conducting such tests vary and the effects of procedural variables on the outcomes of these tests are largely unknown. Using a domestic dog as our subject, we had previously identified two sets of stimuli to which the dog differentially responded (retrieval vs. non-retrieval). In the current study, we conducted two transfer of function tests to test whether the retrieved set of objects functioned as a toy concept. We compared the effects of two different test procedures (successive and simultaneous presentations) on the results of the transfer tests. In the successive presentation test, our results indicated that transfer of function along toy class lines did not occur. Instead, an error analysis indicated that the dog’s responding was consistent with a feature discrimination that did not coincide with the toy vs. non-toy class. However, in the simultaneous presentation procedure, the dog did respond in a way consistent with a toy concept. We discuss results with regard to the care researchers should employ when selecting procedures and drawing conclusions based on those procedures, and we consider why the procedures may produce different results.
106. Further Behavioral Economic Analyses of Choice
Area: EAB; Domain: Applied Behavior Analysis
CASEY J. CLAY (New England Center for Children), Jamie Leigh Lebowitz (New England Center for Children), Jason C. Bourret (New England Center for Children)
Abstract: The current study is an extension of previous research analyzing choice using behavioral economics. In the first study, preference for choosing between a number of identical edibles versus receiving an edible with no choice was examined. In Study 2, consequences involving choice or no choice were arranged contingent upon responding across a range of FR values. Data were plotted as work and demand functions and are discussed in terms of the effects of including choice as part of reinforcement delivery on responding across unit prices.
107. A Simultaneous Approach to Functional Communication Training and Generalization in the Treatment of Destructive Behavior Maintained by Multiple Functions
Area: EAB; Domain: Applied Behavior Analysis
TODD M. OWEN (University of Nebraska Medical Center), Caitlin J. Smith (Munroe-Meyer Institute), Kasey Stephenson (Munroe-Meyer Institute, University of Nebraska Medical Center), Terry S. Falcomata (University of Texas at Austin), Henry S. Roane (SUNY, Upstate Medical University)
Abstract: In the current study, we used a modified functional communication training (FCT) procedure to simultaneously treat multiple functions of destructive behavior (i.e., aggression, self-injury [SIB]) exhibited by an adolescent with a developmental disability. To facilitate discrimination between maintaining reinforcement contingencies, the participant was simultaneously trained to touch a “break” card to escape from academic demands or a “movie” card to gain access to a movie. This treatment produced an immediate reduction in both aggression and SIB and the respective mands occurred at elevated rates. Delays to reinforcement were incorporated into the treatment for each function and were systematically thinned to 15-min delay values. Thus, a 15-min delay was independently associated with each respective reinforcement contingency (i.e., escape, access to movie) and reinforcement intervals were gradually increased to 5-min. In addition, the session length was extended to 2 hours to facilitate generalization of the treatment to the classroom setting. Rates of destructive behavior remained low during delay fading and generalization training. Reliability data were collected on 41% of the sessions resulting in a mean agreement of 98%.
108. Using Equivalence Procedures With Class-Specific Reinforcers and Responses to Teach Math to Young Children
Area: EAB; Domain: Applied Behavior Analysis
RACHEL KOLB (University of North Carolina, Wilmington), Carol Pilgrim (University of North Carolina, Wilmington), Ashley Blackwell (University of North Carolina, Wilmington), William Hogan (University of North Carolina, Wilmington), Whitney Luffman (University of North Carolina, Wilmington), Jenna Peterson (University of North Carolina, Wilmington), Minela Subasic (University of North Carolina, Wilmington)
Abstract: This study used stimulus-equivalence procedures, with compound class-specific consequences and class-specific responses to teach math to young children. Recombinative generalization (Meuller et al, 2000) was tested, where units of trained math stimuli were recombined and probed as novel math problems. A test-train-test-train-test design was used. Experimental participants contacted pretests, training, and post-tests Matched control participants contacted pretests and post-tests only. Five experimental participants completed the first conditional-discrimination training, quantity-to-quantity matching. Bead-counting was the class-specific response trained. Compound class-specific consequences included numerals, spoken words, and printed words. Subsequently, experimental participants demonstrated all emergent relations. Controls did not demonstrate emergent relations on post-tests. Participant KW contacted pretests for subsequent training phases and responded 100% accurately. On recombination probes KW was accurate after the format of the flashcards was changed to match the computerized math stimuli. Four experimental participants contacted AB training, matching quantities to +1 math problems. SH received post-tests and demonstrated all emergent relations. Control participant AJ demonstrated no emergent skills on post-tests. SH received recombination probes and demonstrated no emergence. SH contacted AC training, matching quantities to +2 math problems, then demonstrated all possible emergent relations including equivalence. This study supports Sidman’s (2000) account of equivalence, and addresses teaching efficiency.
109. The Effect of Adding a Common Delay to a Probability Discounting Task
Area: EAB; Domain: Experimental Analysis
SHAWN R. CHARLTON (University of Central Arkansas), Bradley Gossett (University of Central Arkansas)
Abstract: Current discounting research focuses on three forms of discounting: temporal, probability, and social. In general, these literatures show that discounting in each of the three contexts is influenced by similar variables (magnitude, sign, commodity type). However, few studies have investigated the results of combining these contexts. An exception to this is the work of Charlton, Yi, and colleagues (2009) on delayed social discounting. This research suggests that by adding a common delay to a social discounting context, participants become more willing to share an outcome (social discounting decreases) The current study addresses the possibility that including a delay to a probability-discounting task will similarly influence the observed discounting rates. Participants in this study are given a probability-discounting task in which they choose between a small certain amount and or a larger gamble ($50 for certain of a 90% chance of winning $100) at each of three delays 0 days, 182 days, or 365 days. The researchers anticipate that adding the delay to the outcomes’ availability will shift the rate of probability discounting toward s = 1.0. Initial results suggest this to be the case. The results from the full study will be presented and their implications discussed.
110. Feeding History and Genotype Increase Survival of Rats Exposed to Contingencies of Food Restriction and Food-Related Travel
Area: EAB; Domain: Experimental Analysis
W. DAVID PIERCE (University of Alberta), Abdoulaye Diane (University of Alberta), Donald C. Heth (University of Alberta), James C. Russell (Alberta Institute of Human Nutrition), Spencer D. Proctor (Alberta Institute of Human Nutrition)
Abstract: Rats die from activity anorexia (AA) when exposed to time-limited feedings and the opportunity for wheel running. We compared the behavior and survival of juvenile JCR-LA-cp obese-prone (cp/cp) and lean-prone (+/?) rats repeatedly exposed to daily 1.5 h feedings followed by 22.5 h of free access to running wheels. During an adaptation phase, half of the obese-prone rats (n=8) were pair fed the daily average amount of food consumed by freely fed lean-prone animals (n=8) and the remaining obese-prone rats received free access to food (n=8). Next, all animals were exposed to the AA protocol until reaching 75% of free-feeding body weight (starvation criterion) or lasting 15 days (stability criterion). Results showed that obese-prone rats consumed more food and lasted twice as long in AA compared to lean-prone animals, even when similar in initial body weight. Despite running at the same pace, total wheel running of obese-prone rats was three time more than lean-prone animals. The pair fed obese-prone rats consumed more food and had less body weight loss than freely fed obese prone rats. The effects of feeding history and genotype contribute to surviving the contingencies of food restriction and food-related travel.
111. Suppression of Schedule-Induced Drinking in Rats by a Lick-Lever Withdrawal Contingency
Area: EAB; Domain: Experimental Analysis
ANGELES PEREZ-PADILLA (Universidad Nacional de Educacion a Distancia), Ricardo Pellon (Universidade Nacional de Education a Distancia)
Abstract: Six food-deprived rats were exposed to a fixed-interval 30-s food reinforcement schedule and developed stable patterns of operant lever pressing and schedule-induced drinking (Stage A). Animals were matched in three pairs according to their final levels of response rates. Each lick made by any master rat then initiated a resetting 10-s period of lever withdrawal without stopping the fixed-interval timer (Stage B). This led to reductions in licks per minute in all animals, without accompanying decreases in the rate of lever pressing or in reinforcement rate. Drinking and pressing in yoked-control rats, which received food at the same times as those exposed to the lick-dependent lever withdrawal contingency (masters), were not generally altered during Stage B. A rat showed an increase in licks per minute. Initial conditions were reinstated during the last phase of the experiment (Stage A); licking produced no longer any programmed consequence. Schedule-induced drinking of master rats recovered, and no significant changes were observed in the behaviour of control rats or in lever pressing by master rats. These results amplify our knowledge about the procedures capable of punishing schedule-induced polydipsia, and support the view that simply removing the opportunity to press an operant lever on well-trained animals is sufficient to punish adjunctive drinking.
112. Exclusive Preference Does Not Readily Develop When Rats Respond on Concurrent Ratio Schedules of Wheel-Running Reinforcement
Area: EAB; Domain: Experimental Analysis
TERRY W. BELKE (Mount Allison University)
Abstract: Previous investigations of wheel-running reinforcement using concurrent schedules suggest that behavior allocation differs from that observed with more conventional reinforcers. To investigate this difference 16 female Long Evans rats were exposed to concurrent FR 4 FR 4 schedules and the schedule value on the initially preferred alternative was systematically increased across ratios of 6, 8, 12, 16, 24, 32, 40, 56, 72, and 88 responses. The dependent measure was the ratio at which greater than 90% of responses occurred on the unchanged alternative. This procedure was carried out with sucrose solution (.1 ml 7.5% w/v) as the reinforcer on both alternatives and with wheel running (30 s) as the reinforcer on both alternatives within each rat. Results showed that the average ratio at which 90% or more responses were allocated to the unchanged alternative was approximately three fold higher with wheel-running reinforcement. Further analysis showed that although preference shifted with initial increments in the ratio with both reinforcers, changeovers as a percent of total responses decreased with sucrose, but not wheel-running reinforcement. The implications of these results for how wheel running might differ and for theories of choice are discussed.
113. Concurrent Token Production Schedules in Rats: Assessing Sensitivity to the Token Reinforcer Ratio
Area: EAB; Domain: Experimental Analysis
TRAVIS RAY SMITH (Southern Illinois University Carbondale), Blake A. Hutsell (Southern Illinois University Carbondale), Melissa J. Swisher (University of North Texas), Eric A. Jacobs (Southern Illinois University Carbondale)
Abstract: Rats’ lever pressing was maintained by a concurrent schedule of token reinforcement. During token production, a random interval 15 s schedule arranged token deliveries. Each token delivery was probabilistically assigned to either the right or left lever to arrange an overall reinforcer ratio of 9:1 or 3:1, depending upon condition. The location of the rich and lean schedules varied across sessions according to a 31-session pseudo-random binary sequence. Token exchange periods occurred after every token earned. During exchange periods, each token exchange produced access to sweetened condensed milk. Data from both conditions were well described by the generalized matching law and response ratios showed a considerable amount of undermatching.
114. Within-Session Positive Behavioral Contrast as an Animal Model of Pathological Gambling
Area: EAB; Domain: Experimental Analysis
SHAWN SEYEDAIN-ARDABILI (McNeese State University), Benjamin J. Parker (McNeese State University), Cam L. Melville (McNeese State University)
Abstract: Animal models of substance abuse have been used successfully to describe the environmental conditions that control these important behaviors (e.g., Koob, 2000). Animal models of other addictive behaviors such as pathological gambling have not been developed. Positive behavioral contrast is an attractive potential animal model of pathological gambling. One of the central behavioral characteristics of pathological gambling is chasing the bet in which gambling behavior increases following exposure to losses. Similarly, positive behavioral contrast referes to an increase in responding following exposure to reduced rates of reinforcement. The present study asks if positive contrast will be observed when signaled time-outs (losses) replace programmed reinforces during the contrast phase of the procedure. Five male Long-Evans rats pressed levers in a within-session contrast procedure. In the baseline (first half of the session), a multiple variable-ratio 15 bariable ratio 15 schedule delivered reinforcers for lever pressing. In the contrast phase (second half of the session), during the second component of the multiple schedule, 90% of the programmed reinforcers scheduled by the variable ration schedule were replaced by five-second timeouts. Positive behavioral contrast was observed. Implications for an animal model of pathological gambling are presented.
115. Rapid and Slow Changes in Choice Behavior Under Frequently Changing Concurrent-Chains Schedules
Area: EAB; Domain: Experimental Analysis
TAKU ISHII (Keio University), Takayuki Sakagami (Keio University)
Abstract: Key-pecking responses of six pigeons were reinforced with a concurrent-chains schedule operating on left and right response keys. The initial links were two equal variable-interval 8-s schedules. The terminal link for one key (rich key) was fixed-interval (FI) 8-s schedule, and that for the other key (lean key) was changed across sessions according to a pseudorandom binary sequence (PRBS) of FI 8-s and FI 16-s schedules. The choice behavior in the initial links was sensitive to the relative rates of reinforcement in the terminal links; with the FI 16-s terminal link for the lean key, relative rates of key pecking to the rich key increased, and with the FI 8-s terminal links for both of the two keys, the relative rates approached 0.5, being biased in favor of the rich key. Then, we switched the rich and lean key and conducted another PRBS. The results showed that the choice behavior remained sensitive to the relative rates of reinforcement in the terminal links, but the bias in favor of the previously rich key dissipated only gradually. These results may suggest that choice behavior is influenced by both of the short-term and long-term histories of reinforcement.
116. The Effects of Differential Reinforcement of Other Behavior Schedules as a Function of Initial Interval Length: A Translational Study With Rats
Area: EAB; Domain: Experimental Analysis
ALYSSA MARTIN (University of Houston-Clear Lake), Dorothea C. Lerman (University of Houston-Clear Lake), Lindsay Evans (University of Houston-Clear Lake), Jonathan R. Miller (University of Kansas), Lillie Wilson (University of Houston-Clear Lake)
Abstract: Differential reinforcement of other behavior (DRO) is a common intervention for problem behavior. However, few studies have examined the most appropriate way to select the initial DRO interval. In previous basic and applied research, the initial DRO interval was selected arbitrarily or based on the mean interresponse time (IRT). Basing the DRO interval on the IRT is a reasonable approach for ensuring that the individual contacts the contingency with a sufficient degree of regularity during treatment. Nonetheless, the extent to which an individual must contact the contingency to produce a successful treatment outcome remains unclear. Relatively short DRO intervals may be highly effective but impractical to implement in applied settings. In this translational study, rats were exposed to conjoint schedules of fixed-interval 120 s for lever pressing and DRO. The initial DRO interval was established such that either 50% or 25% of the IRTs would meet the reinforcement requirement. Both adjusting and non-adjusting DRO schedules were examined, along with a schedule thinning procedure. Results indicated that responding was sensitive to both initial DRO intervals and the DRO intervals could be successfully thinned beyond 25% for some rats. These results have important implications for arranging DRO in applied settings.
117. Enhancing Equivalence Class Formation in Preschool Children and Individuals With Mental Retardation With an Extensive Pretraining
Area: EAB; Domain: Experimental Analysis
PRISCILA CRESPILHO GRISANTE (Universidade Federal de São Carlos), Julio C. De Rose (Universidade Federal de São Carlos)
Abstract: Part of stimulus equivalence research has shown mixed results mainly when conducted with preschool children or individuals with mental retardation. This variability could be attributed in part to procedural failure to effectively establish stimulus control relations planned by the experimenter. This experiment employed to preschool children and individuals with Down syndrome an extensive pretraining that gradually established one conditional discrimination with familiar stimuli and four conditional discriminations with abstract stimuli interspersed with pretraining reviews in attempt to develop generalized conditional discriminations skills and minimize errors occurrence. Pretraining was followed by the baseline two-choice conditional discriminations training AB and BC and equivalence tests. Finally, a test in which novel stimuli replaced one of the comparisons of baseline conditional discriminations was applied to verify its adequacy to identify sample-S+ or Sample-S- relations (stimulus control tests). The data showed the learning of baseline with just a few errors and the formation of two three-member classes of equivalent stimuli by all participants. Stimulus control tests were not able to identify the programmed sample-comparison relations, so procedural investigations are made necessary. These results indicate that well-planned pretraining tasks can diminishes errors occurrence enhancing the probability of establishment of planned stimulus equivalence classes in these population
118. Aversive Control of Betta Splendens Behavior Using Water Disturbance
Area: EAB; Domain: Experimental Analysis
Hernan Camilo Hurtado Parrado (University of Manitoba), JOSEPH J. PEAR (University of Manitoba), Praepun Khattiyakornjaroon (University of Manitoba), Kimberly Froese (University of Manitoba)
Abstract: A research program at the University of Manitoba currently explores Betta splendens operant behavior under different contingencies. Specifically, because Betta splendens is native to calm water environments, water disturbance in the form of air bubbles and water currents was tested for the first time as having an aversive function. The poster presents the methodology and results of different procedures, that on one hand confirmed the aversive function of water disturbance, and in the other, constitute preliminary experiments on avoidance and escape behavior. Experiment 1 to 4 tested behavioral differences when air bubbles and water currents in different strengths were introduced to the experimental tank. The amount of time the subject spent in different areas of the tank, and the number of times the fish visited or approached the disturbed area, were both calculated. A clear preference of the fish for the non-disturbed area was always observed across several alternations of the location where the water disturbance was introduced. Experiment 5 constitutes a systematic replication of experiments 2 to 4, this time using a different subject and design. Finally, Experiment 6 corresponds to the modification of the procedure by introducing a signal that preceded the presentation of the water currents.
119. Temporal Regulation in Children Using Differential-Reinforcement-of-Low-Rates
Area: EAB; Domain: Experimental Analysis
MÉLISSA GAUCHER (Université du Québec à Montréal), Jacques Forget (University de Quebec a Montreal), Céline Clément (Université de Strasbourg)
Abstract: Temporal regulation is studied in animals and humans with operant procedures, mostly with fixed-interval (FI) schedules or differential-reinforcement-of-low-rates (DRL). Studies with human subjects indicate differences in response patterns depending on the participants’ age. Infants show response patterns similar to those observed in animals, a scalloping pattern, whereas adults show mostly two types of patterns, low-rate or high-rate. Children can produce both, patterns typical of animals and of adults. The main hypothesis is that there is a transition in temporal regulation from nonhuman-like contingency-governed operant behavior to verbally-governed behavior in humans. This study uses DRL 5 seconds and DRL 20 seconds to measure temporal regulation in children aged between 2;6 and 7;3. The response consists in a press on a yellow circle located in the middle of a touch screen. The reinforcer is a 20 seconds cartoon. To evaluate the main hypothesis, language and cognitive development are assessed and children’s collateral responses are observed.
120. The Relation Between Impulsive Choice and Cardiovascular Reactivity
Area: EAB; Domain: Experimental Analysis
JAMES W. DILLER (Eastern Connecticut State University), Connor H. Patros (Eastern Connecticut State University), Michael M. Gale (Eastern Connecticut State University)
Abstract: Impulsive choice has been defined as the selection of a small, immediate reinforcer, to the exclusion of a larger, delayed reinforcer. Various factors, including drugs of abuse, have been shown to modify levels of impulsive choice. Because drugs have physiological effects and can alter impulsivity, there may be a cardiovascular correlate of impulsive choice. Additionally, the spontaneously hypertensive rat, an animal model of attention-deficit hyperactivity disorder, was originally bred for its cardiovascular profile. The present study explores the relation between impulsivity (measured using a computerized delay discounting task) and cardiovascular reactivity (change in heart rate during a serial subtraction task) in a college-based sample. Data were collected from 39 participants (24 female). A linear regression suggests that there is a predictive relation between cardiovascular reactivity and impulsivity, B= -.40, t(38) = -2.67, p < .05, R2 = .16. That is, individuals who exhibited higher degrees of impulsive choice also exhibited greater cardiovascular changes in the serial subtraction task. Evaluating the influence of cardiovascular factors on impulsivity may provide additional information about its genesis, possibly leading to improved screening methods for the behavioral and health-related problems associated with this type of maladaptive choice.
121. Comparison of Operant Behavior by Individual Goldfish and Schooled Zebra Fishes
Area: EAB; Domain: Experimental Analysis
TODD HILLHOUSE (Northern Michigan University), Paul Thomas Andronis (Northern Michigan University)
Abstract: Recent advances in developmental neurobiology have poised zebrafish (Danio rerio) as the preeminent model organisms for the study of relations between neurological development and the emergence of adaptive behavior. The behavior of such animals that live and function in groups is commonly portrayed as being in accord with the same laws as those applicable to the behavior of individual organisms. For example, descriptions of foraging by groups and by individuals are commonly rationalized by the same economic variables, and are assumed to be essentially equivalent. This poster reports some preliminary data for comparing aggregate operant behavior by groups of zebrafish (D. rerio) with simple operant behavior by individual goldfish (Crassus auratus). We examined the extent to which there might indeed be point-to-point correspondence in the control over group versus individual foraging behavior in these two species of fish. The subjects were first exposed to simple schedules of food reinforcement, and aggregate patterns emitted by the zebrafish were compared to patterns emitted by individual goldfish under identical contingency requirements. Subsequent experiments involve foraging choices under multiple concurrent schedules of reinforcement.
122. Instructional Programming for Generalized Sound-Print Relations: Towards Augmentative Communication for Individuals With Low Speech Intelligibility
Area: EAB; Domain: Experimental Analysis
ANNA C. SCHMIDT (University of Kansas), Nancy C. Brady (University of Kansas), Kathryn Saunders (University of Kansas)
Abstract: For individuals with little functional speech, literacy skills can enable effective use of augmentative and alternative communication (AAC). Providing listeners with supplemental alphabetic cues has been shown to substantially increase speech intelligibility (i.e., the listener’s understanding of the speaker). In one technique for providing letter cues, the speaker must abstract the initial sound of a spoken word and point to the corresponding letter on an augmentative device. In the current study, two men with intellectual disabilities, low speech intelligibility, and receptive language age equivalent scores of less than three years completed training designed to teach abstraction of initial sounds. Spoken words were presented by a computer, and participants selected the corresponding letter on the computer screen. Words were trained in pairs, in a two-choice matching-to-sample task. Generalization to additional words with the same initial sounds was assessed. The training resulted in highly accurate letter selection for both participants. At the end of training, participants were also highly accurate with untrained words, could accurately select letters in a six-choice task, and maintained high accuracy for over 70 spoken words. These findings are a first step towards the development and effective use of AAC involving supplemental alphabetic cues for this underserved population.
123. Effects on Stimulus Equivalence Test Performance of Varying the Delay in Delayed Matching-to-Sample
Area: EAB; Domain: Experimental Analysis
CHRISTOFFER EILIFSEN (Akershus University College), Erik Arntzen (Akershus University College), Felix Hognason (Akershus University College)
Abstract: When training prerequisite relations for stimulus equivalence using a matching-to-sample procedure, introducing different delays between the offset of conditional stimuli and the appearance of discriminative stimuli has resulted in systematic variations in test performance as a function of the delay. We have previously obtained results that have indicated lower yields of stimulus equivalence responding in groups of participants exposed to a fixed delay of 3000 ms, compared to a groups exposed to delay values of 100 ms or 12000 ms. The current study is an attempt replicate these findings and, in addition, to allow for a more precise analysis of the effects of different delay values by exposing the same participant to several delays during separate training procedures. Adult participants will be exposed to 0 ms, 2000 ms, 2500 ms, 3000 ms, 3500 ms, and 4000 ms fixed delays during training over six experimental phases, with each of training phases being followed by a corresponding test for stimulus equivalence.



Back to Top
Modifed by Eddie Soh