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.


40th Annual Convention; Chicago, IL; 2014

Event Details

Previous Page


Poster Session #292
Sunday, May 25, 2014
7:00 PM–9:00 PM
W375a-d (McCormick Place Convention Center)
1. Responding Under Escalating and Constant Unit Prices: Manipulation of Reinforcement Duration
Area: EAB; Domain: Basic Research
JAMES ALLEN CHASTAIN (Kennedy Krieger Institute), Christopher E. Bullock (Kennedy Krieger Institute), Iser Guillermo DeLeon (Kennedy Krieger Institute)
Abstract: We examined completion of two-digit, computer-based arithmetic problems by undergraduate college students under two conditions. In the Escalating Unit Price condition, fixed-ratio (FR) response requirements to produce the reinforcer, video clips, were systematically increased across blocks of trials while reinforcer duration remained unchanged. In the Constant Unit Price condition, FR response requirements were systematically increased with equivalent increases in duration of video access. Data were recorded on terminal FR schedule values completed prior to participants choosing to terminate a session, total (correct and incorrect) number of problems completed in a session, latency to first response per trial and average rate of problem completion. Results showed that 9 of 12 participants generally completed higher FR values (and consequently more total problems) during Constant Unit Price conditions than during Escalating Unit Price conditions. No within-participant differences in average latency to first response or average response rates were observed across conditions. These data demonstrate the importance of unit price and magnitude of reinforcement values when measuring total responding; and show that measures based on amount of responding, such as progressive ratio schedules, may show sensitivity to differences that more conventional measures such as response rate and latency to respond do not.
2. The Behavioral Economics of Effort: The Unspecified Role of the Sub-Criterion Response
Area: EAB; Domain: Basic Research
CHRISTINA M. NORD (University of North Texas), Jonathan W. Pinkston (University of North Texas)
Abstract: Although response effort is considered a dimension of the cost to obtain reinforcement, little research has examined the economic impact of effort on demand for food. The goal of the present study was to explore the relationship between effort and demand. Three Sprague Dawley rats were trained to press a force transducer under a series of fixed-ratio schedules (1, 10, 18, 32, 56, 100, 180, 320, and 560) under different force requirements (5.6 g, 18 g, and 56 g). Thus, we maintained a constant nominal unit price (responses / food) but varied the minimal response force. Using a force transducer allowed us to measure responses failing to meet the minimal force requirement (i.e. sub-criterion responses), an advantage over prior approaches using weighted levers to manipulate effort. Results showed that demand assessments were reliable at each force requirement. Consistent with prior research, increasing the unit price decreased food consumption, and raising minimum force requirements further reduced demand for food. Additionally, increasing the force requirement increased the number of sub-criterion responses. The increase in sub-criterion responses raises the question of whether previous reports of force-related decreases in food demand result from the force manipulation or from incidental changes in the sub-criterion class.
3. The Near-Miss Effect in Blackjack: Group Play and Lone Play
Area: EAB; Domain: Basic Research
KARL GUNNARSSON (Southern Illinois University Carbondale), Seth W. Whiting (Southern Illinois University), Mark R. Dixon (Southern Illinois University)
Abstract: Previous research in blackjack demonstrated that gamblers report outcomes are closer to wins when the player's total more closely approximates the dealer's total. However, additional comparisons, such as to another player's total or to 21, may affect the prevalence of a near-miss. The current study investigated the presence of a near-miss in blackjack while playing alone and with other players, and examined ratings in relation to the difference of the player's total from 21, the dealer's total, and another player's total. College students played 25 hands of blackjack with the only the dealer and another 25 hands with another player and the dealer and rated how close the outcome was to a win. The results demonstrated the presence of a near miss effect as a function of the numerical distance from the player's total to 21, and the absence of a near miss when the player busts.
4. A Functional Exploration of the Near-Miss Effect in Slot-Machine Gambling
Area: EAB; Domain: Basic Research
JORDAN BELISLE (Southern Illinois University), Rachel Enoch (Southern Illinois University), Mark R. Dixon (Southern Illinois University)
Abstract: The near-miss effect is an event whereby losses that appear close to wins, such as two of the same symbol on a three-reel slot machine, have reinforcing consequences that are similar to wins. The present study provides a functional, rather than a structural, account of the near-miss effect at the behavioral level of analysis. The first experiment demonstrated that the symbols associated with near-misses are conditioned as reinforcers during wins and generalize to near-misses. The second experiment demonstrated that symbols that are never reinforced during play may also obtain reinforcing value. The implications, limitations, and directions for future research are discussed.
5. Structural Characteristics AND Slot Outcomes Impact Subject Evaluations of Slot Machine Likability
Area: EAB; Domain: Basic Research
AMY K. LOUKUS (Southern Illinois University Carbondale), Mark R. Dixon (Southern Illinois University)
Abstract: The current study examined the impact of variables thought to influence gamer choice of slot machine, including structural characteristics (e.g., theme, color scheme, interactivity, number of reels) and displayed outcomes. Fifty-six undergraduate students completed an online survey for course credit in which they rated the likability of 40 slot machines (20 3-reel, 20 5-reel) according to a 5-point Likert scale, and anecdotally identified highly preferred features of their top three favorite machines for each set. Following, a paired-stimulus preference assessment of 10 machines (five 3-reel, five 5-reel machines) allowed for the systematic evaluation of feature preferences across all participants. The group was later presented with the same task but all payouts/machine outcomes were removed. Results suggest that although the structural features more commonly believed to influence machine choice may hold some influence on gamer preference, participants opted to play the machines with winning display patterns more frequently when outcomes were presented and other structural features remained the same. Implications of these results for future research and in the treatment of problem gambling will be discussed.
6. The Role of Conditioned Reinforcement in The Near Miss
Area: EAB; Domain: Basic Research
ROBERT M. SCHIENLE (University of Nevada, Reno), Patrick M. Ghezzi (University of Nevada, Reno), Benjamin N. Witts (St. Cloud State University)
Abstract: The near miss effect in slot machine gambling is generally understood as two out of three winning reel symbols align horizontally at the payline, with the third winning symbol appearing just above or below the payline. Previous slot machine research has shown that the presentation of a near miss increases the duration spent gambling (Côté, Caron, Aubery, Desrochers, & Ladouceur, 2003; Ghezzi, Wilson, & Porter, 2006). The current study aims to extend what is known about the reinforcing properties of the near miss by using an observing response procedure, whereby the participant must make a response to reveal the outcome of a reel spin. University students were recruited via online software and asked to play a simulated slot machine for 125 trials. Results indicate that participants were more likely to make an observing response on near miss reel spins than on winning or full miss reel spins. Therefore, results of this study suggest the presentation of a near miss does, in fact, function as a conditioned reinforcer in slot machine gambling.
7. Persistence of Positive Reinforced Lever Pressing Following Variable vs Fixed Negative Reinforced Water Maze Escape-Training
Area: EAB; Domain: Basic Research
STEPHEN RAY FLORA (Youngstown State University), Sarah Biviano (Youngstown State Universiy)
Abstract: The basis of Eisenbergers Learned Industriousness theory is that reinforcement of high effort for one behavior will result in the generalized persistence of other behaviors (e.g., Eisenberger, Terborg, & Carlson, 1979; Eisenberger, 1990). However, no study has investigated whether negatively reinforced high effort behaviors (e.g., high effort escape) will result in increased generalized persistence for positively reinforced behaviors. In the present study two groups of rats had to escape from a radial arm water maze. The variable escape group, because the submerged platform was placed on a different arm on each trial, had to exert high effort to escape each trial relative to the rats in the continuous escape group that simply had to swim to the same location on every trial. Agreeing with, and extending learned industriousness theory, the rats in the variable escape were more persistent during the bar pressing (shaped and maintained by positive reinforcement food pellets) extinction phase at the end of the study than the continuous escape group. However, inexplicably, a control group receiving no water maze training was the most persistent group and pressed at the highest rates both for reinforcement and during extinction
8. Probability Affects Choices Involving Negative Reinforcers: Experiment 1 – Delay
Area: EAB; Domain: Basic Research
JENNIFER A. BONOW (Blueprints), Patrick M. Ghezzi (University of Nevada, Reno)
Abstract: Examining the variables affecting human choice behavior is particularly important as people regularly engage in risky, harmful, and even life-threatening behaviors such as smoking, gambling, and drinking excessively. The current study was the first in a series investigating the role of the probability of reinforcement in a self-control paradigm utilizing negative reinforcement. This variable is often included in discounting research, but not frequently examined in basic self-control experiments. This experiment examined the effects of varying the delay to, and probability of, reinforcement while holding the magnitude constant. 27 undergraduate students were exposed to a loud white noise while watching a DVD movie. Selecting one of two buttons on the screen resulted in either an immediate or delayed removal of the noise. A parametric examination of different probabilities of reinforcement demonstrated a shift in selecting behavior across participants. The mean percentage of selections of the immediate alternative was 87% of trials when the probability of reinforcement was 1.0 and decreased across reduced probabilities to 36.9% of trials when the probability of reinforcement was .25. Data indicate that probability and delay interact in a self-control paradigm suggesting that the influence of this factor on human choice behavior should be examined more thoroughly.
9. Probability Affects Choices Involving Negative Reinforcers: Experiment 2 – Magnitude
Area: EAB; Domain: Basic Research
Jennifer A. Bonow (Blueprints), LAUREN JONES (University of Nevada, Reno), Patrick M. Ghezzi (University of Nevada, Reno)
Abstract: Examining the variables affecting human choice behavior is particularly important as people regularly engage in risky, harmful, and even life-threatening behaviors such as smoking, gambling, and drinking excessively. The current study was the second in a series investigating the role of the probability of reinforcement in a self-control paradigm, a variable that is often included in discounting research, but neglected in basic self-control experiments. This experiment examined the effects of varying the magnitude and probability of reinforcement while holding the delay to reinforcement constant. 22 undergraduate students were exposed to a loud white noise while watching a DVD movie. Selecting one of two buttons on the screen resulted in either an immediate short- or immediate long-duration removal of the noise. A parametric examination of different probabilities of reinforcement demonstrated a shift in selecting behavior across participants. The mean percentage of selections of the long-duration alternative was 87% of trials when the probability of reinforcement was 1.0 and decreased across reduced probabilities to 31.79% of trials when the probability of reinforcement was .25. Data indicate that probability and magnitude interact in a self-control paradigm suggesting that the influence of this factor on human choice behavior should be examined more thoroughly.
10. Avoidance Performances of Humans Under Negative Reinforcement With the Reduction of Task Disturbing Events
Area: EAB; Domain: Basic Research
DAIKI MISHIMA (Teikyo University), Kaname Mochizuki (Teikyo University)
Abstract: Herrnstein & Hineline (1966) showed the decrease of shock frequency could negatively reinforce rats' lever presses. We tried to replicate their experiments in humans. Participants played a block stacking game in which they could earn points when they stacked up five blocks in three layers on a platform. During the game, the platform was "shaken" according to two values of VT (variable-time) schedules. These shakes crushed all stacked blocks at that time, so it disturbed the participant to get points. When a participant choose a cyan colored block, the VT value changed to large so that the frequency of shakes decrease. The value returned to the smaller one when the next programed shake occurred in the larger value. The task disturbing shakes corresponded to the electric shock and the choice of safety block corresponded to a lever press in rats' experiment. The experiment consisted of three 10 minutes avoidable sessions and unavoidable sessions which appeared alternatively. Five of 12 university students showed increase of avoidance responses in the last avoidable session and two of them were statistically significant. This experiment showed that the decrease of the density of task disturbances could be a reinforcer for human behavior.
11. Demand Assessment for Quantifying the Value of Negative Reinforcers Using a Progressive Ratio Schedule With a Fixed Positive Reinforcer
Area: EAB; Domain: Applied Research
MEGAN KLIEBERT (Marcus Autism Center), Andrea R. Reavis (Marcus Autism Center), Nathan Call (Marcus Autism Center)
Abstract: Several studies have used progressive ratio (PR) schedules to evaluate the efficacy of positive reinforcers (e.g., DeLeon, Frank, Gregory, & Allman, 2009; Penrod, Wallace, & Dyer, 2008, Roane, Lerman, & Vorndran, 2001). Yet there are no direct assessment procedures for quantifying the value of negative reinforcers. The current study sought to determine whether breakpoints when responding on PR schedules for various tasks can be used to quantify the value of those tasks. The study includes one participant. Following a demand latency assessment conducted to identify a hierarchy of demands in terms of aversiveness by examining the latency to problem behavior following the presentation of demands, high and low aversive demands were selected for inclusion in a PR demand assessment during which the participant worked to access a highly-preferred edible item on increasing schedules of reinforcement. Higher breakpoints were observed for the low aversive task and lower breakpoints were observed for the high aversive task. Results indicate this method of quantifying reinforcer value may be ideally suited for the purpose of determining the value of escape/avoidance of various negative reinforcers; therefore, could serve as a means of assisting in developing effective treatments for problem behavior maintained by escape/avoidance. Keywords: demand, escape-maintained problem behavior, negative reinforcer
12. Effects of Early Big Wins and Probabilistic Wins on Progressive-Ratio Breakpoints in Rats
Area: EAB; Domain: Basic Research
CAROLINE WILLIAMS (College of Charleston), Christine Logan Chambers (College of Charleston), Carolyn Kitts (College of Charleston), Chad M. Galuska (College of Charleston)
Abstract: The big-win hypothesis states that gambling will persist longer if a big-win is experienced early in playing. The behavioral hypothesis states that gambling will persist longer is wins are experienced intermittently. The purpose of this series of experiments was to examine the effects of big-win and probabilistic wins on progressive-ratio (PR) breakpoints in rats. Five female Long Evans rats responded on a lever under a PR 5 schedule for a 4-s delivery of a 50% sweetened condensed milk solution. Sessions ended when rats went 5 min without a response or 10 min without earning a reinforcer. Superimposed on this schedule, one condition arranged a big win condition in which a large number (e.g., 12) of dipper presentations were delivered after the second completed ratio. This resulted in a decrease in breakpoints, which subsequent research suggested was not due to satiation. Subsequently, an random-ratio (RR) 10 schedule was superimposed on the PR schedule; this resulted in a significant increase in breakpoints for all of the rats. Parametric manipulations of both the size of the big win and the RR parameter are underway. Overall, the results fail to support the big-win hypothesis and provide support for the behavioral hypothesis.
13. Resurgence in a Peak-Interval Procedure
Area: EAB; Domain: Basic Research
MIRARI ELCORO (Armstrong Atlantic State University), Kennon Andy Lattal (West Virginia University), April H. Graves (Armstrong Atlantic State University), Michael J. Choromanski (Armstrong Atlantic State University), Kimberly R. Wise (Armstrong Atlantic State University)
Abstract: Resurgence is the recurrence of previously reinforced behavior during extinction of another, more recently reinforced behavior. Resurgence provides a framework for deciphering the origin of operant responses, study effects of historical variables, and also understanding various socially relevant behaviors such as creativity and clinical relapse. In the present study, four experimentally nave Long Evans rats were trained using a concurrent variable-interval (VI) 30-s, fixed-interval (FI) 30-s schedule, each programmed on a different lever. After responding stabilized, responding under the VI 30-s schedule was extinguished for 15 sessions. A peak-interval procedure (PIP) then was employed to examine resurgence of the previously VI responding. The PIP consisted of FI 30-s trials (30% of trials) alternating randomly with 300-s peak trials (70% of trials), each separated by 20-s blackouts. Overall response rates revealed resurgence of VI responding during the peak trials. Additional analyses include proportions of lever presses during the resurgence test relative to responses during extinction and analyses focused on responding during peak trials. These findings are compared with similar studies of resurgence and with findings from the literature on behavioral contrast. The present study extended the use of a procedure typically utilized to examine temporal control to the analysis of resurgence.
14. Resurgence of Previously Reinforced Lever Pressing During a DRO Challenge
Area: EAB; Domain: Basic Research
JENNIFER L. HUDNALL (The University of Kansas), David P. Jarmolowicz (The University of Kansas), Michael Sofis (The University of Kansas), Alexandria Darden (The University of Kansas), Shea M. Lemley (The University of Kansas)
Abstract: Resurgence can be defined as the re-emergence of a previously reinforced behavior when a currently reinforced behavior is met with some challenge. Typically, this challenge is to condition of extinction, under which the phenomenon of resurgence has generally been examined. However, a previously reinforced behavior may re-emerge when the current behavior is met with some challenge other than extinction. This experiment examines the phenomenon of resurgence during both extinction and DRO schedules. During phase 1 of this experiment, right lever pressing was maintained on an FR20 schedule of food reinforcement while presses on the left lever were not reinforced. During phase 2, no lever presses on either lever were reinforced. During phase 3, right lever presses were not reinforced, but left lever presses resulted in food delivery. In the final component, phase 4, left lever presses were not reinforced, and right lever presses were maintained on a DRO schedule. Results indicate that, when DRO is a challenge to the maintenance of a current behavior, re-emergence of a previous behavior is both robust and long-lasting. These data add to the emerging literature regarding resurgence under non-extinction conditions.
15. Alternative Reinforcement Rates and the Magnitude of Resurgence
Area: EAB; Domain: Basic Research
SHUN FUJIMAKI (Keio University), Takayuki Sakagami (Keio University)
Abstract: Resurgence is a reappearance of previously extinguished target response following the removal of reinforcement for the alternative response. Shahan and Sweeny (2011) proposed the quantitative model of resurgence based on behavioral momentum theory. The purpose of the present study was to test the prediction from this model, that is, the higher-rate of alternative reinforcement produces greater resurgence. We conducted the experiment using pigeons under a two-component multiple schedule across three-phase. In the acquisition phase, both target responses in Rich and Lean component were reinforced on VI 60-s. In the elimination phase, target responses were extinguished in both components, while alternative responses were reinforced on VI 30-s in Rich component and VI 120-s in Lean component. In the resurgence phase, all reinforcement was withheld. Figure 1 shows proportion of baseline response rates in the elimination and resurgence phases. Pigeon D11 showed greater resurgence in Rich component. Inconsistent with the prediction, however, the magnitude of resurgence was greater in Lean component in B21 and was little difference between components in each pigeon. Thus, there was no systematic relationship between alternative reinforcement rates and resurgence.
16. Repeated Resurgence Across Sessions
Area: EAB; Domain: Basic Research
JAMES E. COOK (West Virginia University), Kennon Andy Lattal (West Virginia University)
Abstract: Resurgence is the reoccurrence of a previously extinguished response when an alternative response no longer produces reinforcement. Resurgence is a replicable process, but few studies have examined the results of repeated resurgence tests. Resurgence tests with nonhuman animals may take weeks to prepare, and the resurgence effect is transitory, lasting only 1-2 sessions, making examinations of resurgence across time difficult. Four pigeons underwent repeated resurgence tests across daily sessions. Each session was divided into 3 phases. In Phase 1, responses on the left key (resurgence key) were reinforced on a fixed-interval (FI) 30-s schedule. In Phase 2, responses on the center key (alternative key) were reinforced on a FI 30-s schedule, and responses on the left key were placed on extinction. In Phase 3 (resurgence test), responses on all keys were placed on extinction. The right key (control key) never provided reinforcement. Phases changed when responding occurred exclusively on the key providing reinforcement for 2.5 min. In the resurgence test, responding occurred on the resurgence key, and little to no responding occurred on the control key across sessions. The resurgence effect decreased but occurred repeatedly across 30 consecutive sessions.
17. Resistance to Change in Choice and Non-choice Contexts in Rats
Area: EAB; Domain: Basic Research
CALEY N. DOUD (College of Charleston), Chad M. Galuska (College of Charleston)
Abstract: Research investigating the factors governing resistance to change in the face of disruption has focused largely on the prevailing reinforcement rate prior to disruption. We investigated the role that response freedom has on behavioral persistence. Rats responded under a two-component multiple schedule that alternated throughout the session. In one component, presses on a single lever produced food pellets according to a variable interval (VI) schedule. In the other component, two levers were available and a concurrent VI VI schedule was arranged. In one group of rats, the scheduled reinforcement rate was higher in the single component. For another group, the scheduled reinforcement rates was higher in the concurrent component. In the critical group, the scheduled reinforcement rate was held constant in the two components of the multiple schedule. After a baseline was obtained, responding was disrupted with prefeeding and extinction. Results-to-date indicate that when reinforcement rate varied, responding in the component with the higher reinforcement rate was more resistant to change. When the reinforcement rate was held constant, the component associated with the lower response rate tended to be more resistant to change. Little support was found for the contention that having multiple response options increases response persistence in the face of disruption.
18. Behavioral Resistance to Change and Ethanol Effects
Area: EAB; Domain: Basic Research
TALITA CUNHA (Universidade de São Paulo - USP), Miriam Garcia-Mijares (Universidade de Sao Paulo)
Abstract: Drugs can cause different effects in organisms, as different performance in a task or stimulus discrimination uncommon. Organism condition before a task is an important factor to behavior, and this design was not investigated under behavioral momentum approach. The goal of the experiment was to investigate the effects of extinction over the behavioral resistance to change of organisms responding under ethanol effects. Four wistar rats self-administered 5g/kg ethanol diluted in agar (ALC) or 15% maltodextrin diluted in agar (MAL), in random days, and were exposed to a multiple schedule VI15VI45. After baseline, six extinction sessions were performed. The data showed negligible differences in response rates between ALC and MAL conditions on baseline or extinction phases. However, resistance was higher for VI15, under both conditions. The results demonstrated that extinction affected the responses controlled by the multiple schedules in accordance to the behavioral momentum theory, and that this effect is maintained after alcohol consumption. As it seems that this is the first experiment to investigate this issue, the effect of other disruptors should be address by other experiments.
19. Pauses in Multiple Extinction Fixed-Interval Reinforcement Schedules With Fixed Durations of the Extinction Component: Implications for Timing
Area: EAB; Domain: Basic Research
KALLIU CARVALHO COUTO (oslo and akershus university college ), Joao Claudio Todorov (Universidade de Brasilia), Lucas Couto de Carvalho (Oslo and Akershus University College)
Abstract: Five rats served in an experiment with multiple extinction fixed-interval schedules of reinforcement. While the fixed-interval schedule was kept constant at 60 s, the fixed extinction period was varied from 10 to 160 s. Pauses during the fixed-interval schedule decreased systematically with increases in the previous extinction period. Pausing was under control of both time to the next primary reinforcement signaled by the discriminative stimulus associated with the fixed-interval schedule and time since the last primary reinforcement, signaled by the duration of the extinction period.
20. Human Timing: A Comparison of Tree Procedures
Area: EAB; Domain: Basic Research
IVÁN BÁRCENAS (Universidad de Guadalajara), Gonzalo Fernandez (Universidad de Guadalajara), Carlos Torres (Universidad de Guadalajara), Luis Hernando Silva Castillo (Universidad de Guadalajara)
Abstract: Time perception is essential for survival and adjustment of an organism to its environment. Timing is a process underlying choice behavior and learning. The way in which organisms organize their behavior according to the temporal properties of the environment has been extensively studied; several procedures used in the study of animal learning have been adapted for human research; specifically, differential reinforcement of low rate (DRL), temporal bisection and peak procedure. However, the data obtained in such procedures in humans has not been systematically compared to determine the functional equivalence of the different procedures and reliability. The present study aims to compare the numerical index provided by these procedures where the timing is evaluated (indexes are quantitative descriptions of accuracy and precision level in timing). In order to find out about the functional similarities among these procedures, fifty Mexican students were evaluated in three different tasks (Temporal bisection, pick procedure and DRL). Comparisons of procedures provide knowledge for understanding the human perception of time and information about reliability of the instruments and tasks used to study of timing.
21. A Neurocomputational Model of Stimulus Class Formation in Down Syndrome
Area: EAB; Domain: Basic Research
ALVARO TORRES CHAVEZ (Universidad Nacional Autonoma de Mexico), Angel Tovar y Romo (Universidad Nacional Autonoma de Mexico), Adriel Ruiz (Universidad Nacional Autonoma de Mexico)
Abstract: The neurophisiological profile of individuals with Down syndrome (DS) sets restrictions on stimulus class formation (i.e., learning to categorize stimuli on the basis of functional equivalence). At the neuronal level, learning is an outcome from the balance between long term potentiation (LTP) and long term depression (LTD); animal models of DS have shown an atypical balance between LTP and LTD, this is probably the basis of difficulties reported in DS populations when they learn and derive stimulus relations of the kind observed in stimulus class formation tasks. We developed a neurocomputational model to analyze the influence of LTP/LTD imbalance on class formation. AB and BC stimulus relations where trained in the model, and then we evaluated the emergence of the transitive AC relations. Initially the model was unable to show emergence of the transitive relations, this is similar to difficulties reported on tests performance of individuals with DS. We proved different training protocols and found that difficulties related to the emergence of transitive AC relations could be overcome when the learning of BC relations did not interfere with the learning of AB relations. From these simulations we derive an appropriate training structure and sequence to be used with people with DS.
22. A Comparison of Match-to-Sample and Respondent Training of the Blocking Effect in Equivalence Classes
Area: EAB; Domain: Basic Research
KRISTOPHER BROWN (Youngstown State University), Michael C. Clayton (Youngstown State University)
Abstract: Basic research has implicated Kamin’s blocking effect in equivalence class formation. Although both match-to-sample and respondent-type training are procedures used to facilitate class formation, only match-to-sample has been studied in research on blocking in class formation. Since respondent-type training may resemble natural learning opportunities more closely, research on the blocking effect in respondent-trained equivalence classes is important for a more coherent explanation of both blocking and class formation. Thus, the purpose of the present study was to compare match-to-sample and respondent-type training for their susceptibility to blocking using undergraduate students as subjects. Initial results indicate that scores on post-tests containing stimuli included in the blocking preparation to be slightly lower than scores on tests containing non-blocked stimuli. Results are discussed in regards to the necessary and sufficient conditions for equivalence class formation and directions for future research.
23. Equivalence Class Formation Without Explicit Reinforcement of Conditional Discriminations
Area: EAB; Domain: Basic Research
HANNA STEINUNN STEINGRIMSDOTTIR (Oslo and Akershus University College), Erik Arntzen (Oslo and Akershus University College )
Abstract: Different procedural arrangements have been used to establish the necessary conditional discriminations needed for testing for equivalence formation. We asked the following research questions: Is it possible to establish conditional discriminations without any programmed consequences? If these conditional discriminations are intact before testing, will the participants form equivalence classes? Sixteen college students participated in the present experiment. The training and testing of equivalence class formation were arranged in a paper-and-pencil format. Each trial was organized as a row and the sample stimulus was always in the left-most column, while the three comparisons were in three columns to right. For each trial, pairs of stimuli (e.g., A1 and B1) always appeared together while the other comparisons (e.g., B2 and B3) appearing together with them changed from trial to trial. The participants were asked to highlight the correct comparison in each row. When one comparison was highlighted for all 36 trials, the participants handed in the training sheet and were given a test sheet with 18 trials. The results from the training showed that 15 participants responded correctly according to the experimenter-defined conditional discriminations while none of the participants responded in accordance with stimulus equivalence on the test (see Figure 1).
24. The Effects of Experimentally Establishing Meaningful Stimuli on Equivalence-Class Outcomes
Area: EAB; Domain: Basic Research
KATERINA THANOPOULOS (University of Massachusetts Lowell), Richard W. Serna (University of Massachusetts Lowell), Ankit Patel (University of Massachusetts Lowell), Nicole Martocchio (University of Massachusetts Lowell), Lanny Fields (Queens College, City University of New York)
Abstract: Understanding the conditions under which stimulus equivalence relations emerge has important theoretical implications for a behavioral definition of symbolic “meaning,” as well as implications for how to better train symbol-referent relations amongst learners with intellectual challenges. Consider the potential stimulus class of non-representational forms, ABCDE. Past research indicates that (1) a high percentage of participants demonstrate equivalence relations if the C stimuli are meaningful (e.g., a set of familiar pictures), and (2) the percentage will be much lower if all the stimuli in the class are non-representational. The present study asked whether experimentally establishing non-representational C stimuli as meaningful in college-student participants would produce results similar to those found when the C stimuli are familiar pictures. Thus, Group 1 received pre-training and testing in which C stimuli became part of a separate equivalence class (i.e., CFG), prior to ABCDE training and equivalence testing. Results were compared against Groups 2 (non-meaningful stimuli) and 3 (picture stimuli). The results to date show that Groups 1 and 3 had relatively higher equivalence success rates than Group 2, indicating that the pre-training for Group 1 established meaning for the C stimuli. Additional data collection is underway to increase group sizes and statistical power.
25. Effects of Different Emotional Expressions on the Relatedness of Stimuli in Equivalence Classes
Area: EAB; Domain: Basic Research
MARIELE CORTEZ (Universidade Federal de Sao Carlos), Julio C. De Rose (Universidade Federal de Sao Carlos)
Abstract: Recent studies using a semantic differential (SD) to verify transfer of meaning among equivalent stimuli showed a difference in evaluations for different emotional expressions (lower deviation for stimuli that were equivalent to happy faces than for stimuli that were equivalent to the angry faces). This study used an SD to verify relatedness as a function of different emotional expressions (happiness, sadness, and fearless). Ten college students formed equivalence classes that consisted of arbitrary stimuli (B, C, D, and E) and meaningful stimuli - faces expressing happiness, sadness, and fearless (A). After showing equivalence class formation, participants evaluated D stimuli with an SD. Using the same device, an untrained control group (n=18) evaluated the faces. Preliminary analysis seemed to indicate lower deviation scores for stimuli that were equivalent to the fear and sad faces than for stimuli that were equivalent to the happy faces. In this case, participants seemed to overestimate D stimuli (e.g., D stimuli equivalent to the happy faces were evaluated as more positive than the happy faces evaluated by the control group). Further analysis is needed to confirm the effects of the different emotional expressions used in this study on the relatedness of stimuli in equivalence classes.
26. Comparing the Effect of Emotions on False Memories Using the Stimulus Equivalence and DRM Paradigms
Area: EAB; Domain: Basic Research
NATALIA AGGIO (Federal University of Sao Carlos, Brazil), Julio C. De Rose (Universidade Federal de Sao Carlos)

The Stimulus equivalence and DRM Paradigms were used to investigate the effect of emotions on false memories. On Phase 1, six college students were taught three four-stimuli equivalence classes - Classes 1, 2 and 3 - and three 12-stimuli equivalence classes - Classes 4, 5 and 6, all with one node. The nodes for Classes 1, 2 and 3 were familiar pictures. Three happy, three neutral and three angry faces were randomly used as the node in Class 4, 5 and 6, respectively. The other stimuli in all classes were nonsense words. On Phase 2 participants first saw three lists composed by nine of the 11 nonsense words from Classes 5, 4 and 6. Next, a distracter task were performed for three minutes and than three lists composed by all stimuli from previously list (Targets), the rest of the nonsense words form Classes 5, 4 and 6 (Critical Distractors) and the nonsense stimuli from Classes 1, 2 and 3 (Non-related distracters) were shown. Participants should indicate witch stimuli were presented on the first lists. Data showed more recognition of critical than non-related distracters on the list composed by stimuli from Neutral class, compared to Happy and Angry classes.

27. Effects of Stimulus Discriminability on the Emergence of Symmetrical Relations in Adult Humans
Area: EAB; Domain: Basic Research
YUSUKE HAYASHI (Penn State Hazleton), James G. Modico (Penn State Hazleton), Christopher T. Russo (Penn State Hazleton), Manish Vaidya (University of North Texas)
Abstract: The goal of the present study was to investigate effects of discriminability of sample and comparison stimuli on the emergence of symmetrical relations. College students were trained on two types of baseline conditional discrimination tasks involving simple (one Japanese character) and complex (two Japanese characters) stimuli. One task involved simple samples and complex comparisons (simple-complex), whereas the other task involved complex samples and simple comparisons (complex-simple). Test trials, run without programmed consequences, presented the symmetry probe of the baseline relations trained. That is, the simple-complex baseline relations presented on the training trials were presented as the complex-simple symmetry relations on the test trials. Similarly, the complex-simple baseline relations were presented as the simple-complex symmetry relations. The results show that the simple-complex baseline relations were acquired faster than the complex-simple baseline relations. The results also show that, relative the acquisition of the baseline relations, the simple-complex symmetrical relations emerged faster than the complex-simple symmetrical relations. In accord with Saunders and Greens (1999) theoretical analysis that untrained discriminations need to be acquired on the stimulus-equivalence test trials, the present results suggest that stimulus discriminability of sample and comparison stimuli on test trials affects the emergence of symmetrical relations.
28. Evaluating the Results of A Functional Analysis of Inappropriate Mealtime Behaviors Using a Trial-by-Trial Analysis
Area: EAB; Domain: Applied Research
LAUREN M. WORCESTER (Gonzaga University), Anjali Barretto (Gonzaga University)
Abstract: While previous studies have demonstrated the effectiveness of the use of functional analysis to determine the reinforcers for inappropriate mealtime behavior, the use of a trial-by-trial analysis has not been applied to clarify functions observed throughout functional analyses of inappropriate mealtime behaviors. The current study utilizes the functional analysis procedures detailed by Piazza, Fisher, Brown, Shore, Patel, Katz, Sevin, Gulotta, and Blakely-Smith (2003) in combination with the trial-by-trial analysis to evaluate the occurrences of behavior during the establishing operation and reinforcement intervals. Results showed that the use of the trial-by-trial analysis helped to clarify the functions of behavior during the functional analysis of inappropriate mealtime behavior. A specific behavioral treatment package aimed at addressing the functions identified throughout the functional analysis was then implemented.
29. A Function-based Intervention for Decreasing Problem Behavior and Facilitating Conditional Requesting
Area: EAB; Domain: Applied Research
MO CHEN (Department of Educational Psychology, University of Minnesota), Joe Reichle (University of Minnesota), Shelley Kreibich (University of Minnesota)
Abstract: The study was aimed to examine the effectiveness of a function-based intervention for decreasing problem behavior and facilitating conditional requesting in a seven-year-old girl with autism using a multiple-probe design across settings. An antecedent-based functional analysis showed that the girls problem behavior was maintained by positive reinforcement in the form of access to preferred items and activities. The tolerance for delayed reinforcement (TFD) procedure was implemented to increase her waiting duration for preferred tangibles with no problem behavior till 5 minutes. Subsequently, a conditional requesting procedure was implemented using menu and communication boards: If she selected a pictorial symbol from her communication board that was also on the menu board, she could have it immediately; If she selected a pictorial symbol from her communication board that was not on the menu board, she had to wait for it with no problem behavior for 5 minutes. Results showed that TFD and conditional requesting procedures effectively reduced the girls problem behavior to a low level, and increased her ability to request items conditionally. Also, intervention outcomes generalized to three other untrained settings and her waiting duration even increased to about 1 hour when probe data were collected during her activity schedules.
30. Functional and Equivalence Class Formation via Specific Consequences in Children With Autism
Area: EAB; Domain: Basic Research
Eugenia Andrea Lee Santos (Universidade Federal do Para  ), KATARINA KATAOKA DIAS (Universidade Federal do Para  ), Romariz Barros (Universidade Federal do Para  )
Abstract: Some previous studies, reporting equivalence or functional class formation with class-specific consequence procedures, found intersubject variability due to loss of reinforcers efficacy. In the present study, we tested (1) the efficacy of introducing variations of events (within a category) as class-specific consequences and (2) the inclusion of such consequences into the equivalence classes. Two children diagnosed with autism participated. They were given identity matching to sample training (stimulus set A) and simple discrimination reversal training (stimulus set B). A variety of videos of a same theme plus a variety of food within the same category (such as sweet or salty) were used as class specific consequences. Probe trials tested for the emergent arbitrary relations AB and Consequence-A relations. Both participants showed 100 percent accuracy in baseline discriminations after four 20-trial training sessions. In probe sessions, both participants showed strong evidence of equivalence class formation, including respective class specific consequences. This study is being continued: new class members will be included via repeated reversal of simple discrimination procedure and tested via matching to sample procedure and vice-versa. We discuss if the distinction between equivalence and functional class formation is merely procedural or reflects two different behavioral phenomena.
31. Transfer of Emotional Meanings to Abstract Stimuli After Simple and Conditional Discrimination Training With Class-Specific Outcome
Area: EAB; Domain: Basic Research
MARCELO VITOR SILVEIRA (Universidade Federal de São Carlos), Julio C. De Rose (Universidade Federal de Sao Carlos)
Abstract: Stimuli class formation implies the development of new types of stimuli control and the extension of psychological functions to members of a class. The present study investigated whether mixed training involving simple discrimination and MTS training with class-specific outcomes was sufficient for the development of two stimulus classes and whether transfer of function phenomenon would occur when the merger behaviors were based in relations among antecedent stimuli and consequences. Five undergraduates were exposed to the training involving stimuli A1, A2, B1, B2, C1 and C2. One signal served to indicate correct matches for class one (Sr1), while other indicated matches for class two (Sr2). After demonstrating the formation of classes A1B1C1 and A2B2C2, all participants learned parings between pictures of human faces expressing emotions (D1 and D2) and stimuli from set A. In further tests with a semantic differential was verified that two, out of five participants evaluated the abstract stimuli from sets B and C as strongly related to stimuli from set D. These are maybe the first evidence of transfer of functions among stimuli classes established by class-specific outcomes.
32. Happy Faces Are More Salient Than Angry Faces in both Equivalence and ERP Experiments
Area: EAB; Domain: Basic Research
RENATO BORTOLOTI (Universidade Federal de Sao Carlos), Naiene Pimentel (Universidade Federal de São Carlos), Julio C. De Rose (Universidade Federal de Sao Carlos)
Abstract: Several studies have shown that different stimuli may be differentially relatable to other stimuli by experimental training. Happy faces, for instance, seem to be more strongly related to their equivalent stimuli than angry faces. Processes that may account for the higher degree of “relatability” of happy faces are not yet known. The current study investigated if the valence of emotional faces has differential impact on P300 waveforms. The P300 is an event-related potential (ERP) component that reflects processes involved in stimulus categorization. Rare pictures of faces expressing happiness (P=0.1) and anger (P=0.1) were mixed with standard pictures of neutral faces (P=0.8) in an oddball paradigm. In some blocks of trials, participants were asked to count happy faces (targets) and ignore both angry (distractors) and neutral (standards) faces; in other blocks, they were asked to count angry faces (targets) and ignore both happy (distractors) and neutral (standards) faces. Reliable differences between the ERPs to happy and angry faces were found only when the happy faces were the targets. The higher salience of happy faces on categorization processes revealed in the current study seems compatible to (and may help to explain) the higher relatability of these stimuli with others through equivalence training.
Keyword(s): poster session



Back to Top
Modifed by Eddie Soh