|
| EAB Sunday Poster Session |
| Sunday, May 28, 2023 |
| 1:00 PM–3:00 PM |
| Convention Center, Exhibit Hall F |
| Chair: Sally L. Huskinson (University of Mississippi Medical Center) |
|
| 96. Conditional Relations: Modification in the Behavior of the Individual and the Properties of Objects |
| Area: EAB; Domain: Applied Research |
| DINORAH ARELY ESCUDERO (Universidad Veracruzana), Betsabe Ruiz Lopez (Universidad Veracruzana), Agustin Daniel Gomez Fuentes (Universidad Veracruzana) |
| Discussant: Casey Marsh (University of South Florida) |
| Abstract: The term creativity has been used to designate novel and original processes. the psychological phenomenon from behavior analysis, as creative behavior, emerges as a conditional relationship transforms the properties of objects and the individual himself. the present study analyzes the conditional relationship from a field logic, between an individual and a stimulus object. An observation and continuous recording system in used with molar units of measurement supported by parameters of time and space, and in language as a convectional system. Two artists from Veracruz participated in the study in the process of creating a handwork of art. the results of the present study suggest that psychological phenomenon, as original behavior, emerges as a functional contact of different levels, made possible by historical and situational dispositional factors and in language as a means of conventional contact. the term creativity is not a psychological phenomenon, but a term used to value acts or products as original that imply conditional interactive behaviors between the individual and the object of stimulus, the product of result can be socially valued as creative behavior. |
| |
|
| 97. Preschool Education: Acquisition of Life Skills for Environmental Protection |
| Area: EAB; Domain: Applied Research |
| MINERVA PEREZ JUAREZ (Universidad Veracruzana, Mexico), Leiny Johanna Borgwardt (Universidad Veracruzana) |
| Discussant: Sally L. Huskinson (University of Mississippi Medical Center) |
| Abstract: In Mexico, educational excellence has focused on improving curriculum and student performance evaluation. However, its application shows limitations related to the concept of competence. An alternative is educational planning centered on the learner and supported by Behavior Analysis (Ribes, 2008, 2018), which includes the domains of knowledge, and the psychological aspects considered in their learning: the levels of functional aptitude and abilities that constitute the competencies, as well as the modes of language involved. The purpose of this study was to analyze the effect of a Teaching-Learning Unit on the acquisition of competencies for environmental protection based on meeting achievement criteria at different levels of functional aptitude. Five children of both sexes from the third grade of preschool participated. A within-subject design with baseline, an intervention phase, and a follow-up phase was used. The results suggest that life skills are displayed more frequently when speaking-listening modes are used and in the five levels of the taxonomy of functions. However, competencies emerge more frequently in the first three levels of the taxonomy of functions. The results are discussed based on the logical categories of behavior theory. |
| |
|
| 98. Reducing Racial Through the Stimulus Equivalence Paradigm |
| Area: EAB; Domain: Basic Research |
| DENISE APARECIDA PASSARELLI (Universidade Federal de São Carlos), Táhcita Medrado Mizael (Federal University of São Carlos (UFSCar - Brazil)), Julio C. De Rose (Universidade Federal de Sao Carlos) |
| Discussant: Casey Marsh (University of South Florida) |
| Abstract: Racial Prejudice can be defined as negative evaluations of stimuli that symbolize an ethnicity. Mizael et al. (2016), using a procedure based on the Stimulus Equivalence Paradigm, reduced the frequency that children associating black faces with a negative symbol. There was a reduction in the differences between ratings of black and white faces on a Likert scale. We replicated this procedure and included the Doll Test to investigate generalization. Nine participants (5 Males, 8.2 years) that demonstrated negative evaluations of black faces were selected to participate in the intervention. Participants were trained to relate a positive symbol (A1) to an abstract stimulus (B1) and then B1 with black faces (C1). The symmetry and equivalence relations were tested. Eight participants formed an equivalence class between black faces and the positive symbol. At the group level, after the intervention, we had a reduction frequency that participants related black faces with negative symbols and black dolls with negative attributes. In addition, before the intervention, white faces were significantly more positively evaluated than black faces; after the intervention, there was no significant difference. However, at an individual level, we found significative results only in the reduction frequency that black faces were related to negative symbols. The results replicated the findings of Mizael et al. (2016) but indicated the challenge of obtaining significant findings at the individual level in a more variated set of measures. |
| |
|
| 99. Do Fixed or Variable Schedules Better Protect Against Effects of Integrity Errors? |
| Area: EAB; Domain: Basic Research |
| CELIA ROSE CELONA (Salve Regina University ), Stephanie Hope Jones (Salve Regina University) |
| Discussant: Sally L. Huskinson (University of Mississippi Medical Center) |
| Abstract: Reduced-integrity implementation of reinforcement-based procedures results in degraded therapeutic outcomes. The reinforcement schedules associated with the procedures impact the extent to which errors are detrimental. However, previous evaluations with manipulations of reinforcement schedules across similar procedures only compared fixed schedules. To address this gap in the literature, we assessed the extent to which errors were detrimental when occurring during the same treatment, but with either fixed or variable reinforcement schedules programmed for alternative sources of reinforcement. We hope that these results will inform arrangements of interventions that are differentially robust when faced with implementer error. |
| |
|
| 100. Text Messages as an Intervention Tool to Increase School Attendance of Teenagers |
| Area: EAB; Domain: Applied Research |
| BRUNA MARIA BARBOSA DA SILVA FRANÇA (Universidade de Brasilia), Roberta Freitas-Lemos (Virginia Tech), Elenice Seixas Hanna (Universidade de Brasilia) |
| Discussant: Casey Marsh (University of South Florida) |
| Abstract: A growing number of studies have been conducted on the effects of sending text messages as a research tool aimed at educational issues. In most of these study designs, the messages are sent to the parents. The present study investigated the effects on school attendance of sending text messages directly to the students. In a series of three experiments to reach the public in situations of socioeconomic vulnerability (n = 24; n = 7; n = 9) , 16 to 19 year old students received text messages during 15 school days. Eligible students had low school attendance (<75%) in the baseline period and were enrolled in the first year of high school. Fifteen messages that reported possible immediate and delayed consequences related to school attendance were developed. Messages were sent at the end of the school day in a randomized order to each participant. Results showed that school attendance rates increased significantly in Study 1. These results were replicated in Study 2, but not in Study 3. We discusses different school procedures that may partially explain the variability of the results between the studies. Text messages might be a useful tool to influence school attendance. Further studies should explore the combination of text messages with other school policies to improve the problematic issue of low school attendance. |
| |
|
| 101. Examining the Magnitude Effect in Rats |
| Area: EAB; Domain: Basic Research |
| EMILY SAETEURN (University of Alaska Anchorage), Gwen Lupfer (University of Alaska Anchorage), Mychal Machado (University of Alaska Anchorage) |
| Discussant: Sally L. Huskinson (University of Mississippi Medical Center) |
| Abstract: The magnitude effect in discounting tasks refers to the tendency to wait more patiently when the reinforcer value is high (e.g., $100 today vs $150 tomorrow elicits more patient responding than $1 today vs $1.50 tomorrow). In the current project, we sought to determine whether flavor preference would impact the presence of a magnitude effect in rats. Following nonhuman-adapted paired-stimulus (PS) and multiple-stimulus without replacement (MSWO) preference assessments to determine flavor preference, each rat was exposed to a series of delay-discounting tasks. Each task was 60-min and involved the rat choosing between a single, immediate pellet or three pellets delivered after an increasing delay. The percentage of lever presses allocated to the larger delayed reinforcer were averaged across 5 days for each subject for her most and least-preferred flavors. Area under the curve (AUC) was subsequently calculated. Results showed when preferences were measured by the PS and MSWO, AUCs were higher for subjects’ highest preferred and lowest preferred flavors, respectively. These data might suggest that the PS is superior to the MSWO for assessing rat flavor preference, and that the PS should be used to further investigate the magnitude effect in nonhuman animals. |
| |
|
| 102. Effects of Ethanol and Nicotine Co-administration on Risky Choice in Rats |
| Area: EAB; Domain: Basic Research |
| ERIN E. WYLIE (West Virginia University), Karen G. Anderson (West Virginia University) |
| Discussant: Casey Marsh (University of South Florida) |
| Abstract: Individuals often engage in the comorbid use of alcohol and nicotine, leading to further-escalated health consequences than the use of either drug in isolation. Given these poor health outcomes from combined administration of alcohol and nicotine, some of which may be due to increased risky choice, more basic research is needed to better characterize drug/dose interactions on risky choice. Probability-discounting procedures involve a series of discrete choices between a smaller, certain reinforcer and a larger, increasingly uncertain reinforcer. Such procedures allow for the assessment of risky choice across studies and facilitate the understanding of drug-related increases in risky choice. The present study investigated effects of several doses of acute ethanol alone (Experiment 1) and in combination with several doses of nicotine (Experiment 2) using eight Sprague-Dawley rats. Effects of these drugs on risky choice will be evaluated using both visual and statistical analysis. Preliminary results for Experiment 1 indicate dose-dependent increases in risky choice following ethanol administration. Such altered decision-making patterns may be better understood through further investigation into combined effects of these drugs on risky choice. |
| |
|
| 103. Conditioned Reinforcement by Stimuli Correlated and Uncorrelated With a Schedule of Food Reinforcement |
| Area: EAB; Domain: Basic Research |
| LILLIAN LOUISE SKIBA-THAYER (West Virginia University), Michael Perone (West Virginia University) |
| Discussant: Sally L. Huskinson (University of Mississippi Medical Center) |
| Abstract: A conditioned reinforcer derives its function from a relation to an established reinforcer. Observing response procedures have been used to evaluate the function of stimuli positively correlated (S+), negatively correlated (S-), and uncorrelated (S1, S2) with established reinforcers such as food (with animals) or money (with humans). Verbal instructions have been shown to alter the extent to which correlated and uncorrelated stimuli reinforce responding in human subjects (Perone & Kaminski, 1992). The present experiment is concerned with the reinforcing function of stimuli with subjects not susceptible to instructional control: pigeons. The pigeons earn food reinforcers by pecking a key on a compound schedule: A variable-interval 60-s schedule alternates irregularly with extinction. On each side of the food key is an observing key. Observing pecks produce, intermittently, brief displays of the stimulus correlated or uncorrelated with the ongoing component of the food schedule. By manipulating the consequences of the two observing keys, we compare the reinforcing function of the correlated and uncorrelated stimuli: (a) S+ and S- vs. S1 and S2, (b) S+ vs. S1, and (c) S- vs. S2. Absolute and relative observing rates provide a measure of the extent to which these stimuli function as conditioned reinforcers. |
| |
|
| 104. Baseline Variability Affects Single-Case Intervention Effect: Bayesian Simulation and Field Studies |
| Area: EAB; Domain: Applied Research |
| MAKOTO SUZUKI (Faculty of Health Sciences, Tokyo Kasei University; Faculty of Systems Design, Tokyo Metropolitan University), Satoshi Tanaka (Laboratory of Psychology, Hamamatsu University School of Medicine), Kazuo Saito (Faculty of Health Sciences, Tokyo Kasei University), Kilchoon Cho (Faculty of Health Sciences, Tokyo Kasei University), Naoki Iso (Faculty of Health Sciences, Tokyo Kasei University), Takuhiro Okabe (Faculty of Health Sciences, Tokyo Kasei University), Jun'ichi Yamamoto (Tokyo Metropolitan University, Faculty of Systems Design) |
| Discussant: Casey Marsh (University of South Florida) |
| Abstract: Exploring how the accuracy of data comparison between baseline and intervention phases is affected by baseline-data variability and changes in slope and level after treatment may contribute to understanding the effectiveness of intervention for a single patient. The simulation study investigated the relationship between the Bayesian model’s data-comparison accuracy, baseline-data variability, and changes in level and slope. The probability of significant difference between the state and forecast values based on the simulation was fitted to a relation model. Simulation results showed that baseline-data variability and changes in level and slope significantly affect the data-comparison accuracy. The indices of changes in level and slope needed for 95% probability of significant changes from the relation model were 1.43 ± 0.31 and 2.76 ± 2.48, respectively. The field study investigated the intervention effects for actual field data using the relationship derived from the simulation study. Actual changes in level and slope for 64.7% and 41.2% studies were greater than the estimated changes in level and slope. These results imply that baseline-data variability affects the data-comparison accuracy using a Bayesian model and provides information on the extent of changes in the level and slope needed for 95% probability of significant change for field data. |
| |
|
| 105. Surrogate Delay Discounting Rate Depends on Observation of Different Health Behavior |
| Area: EAB; Domain: Basic Research |
| RYAN HIGGINBOTHAM (University of Florida), Jesse Dallery (University of Florida) |
| Discussant: Sally L. Huskinson (University of Mississippi Medical Center) |
| Abstract: Delay discounting refers to a reward's loss of subjective value as a function of delay to its receipt. Research has linked higher rates of delay discounting to numerous health-related behaviors including cigarette smoking. Observing impulsive or self-controlled inter temporal choices such as smoking cigarettes or engaging in healthy behavior may influence someone's choices on a hypothetical delay discounting task and thus their discounting rate. If observation of intertemporal choice influences discounting, it should not matter whether that observation is from a first- or third-person perspective. We showed participants (N=65) two videos: one of a woman smoking and one of another woman eating an apple. Following each video, participants completed a delay discounting task as a surrogate decision-maker for each woman (i.e., the choices were made from the perspectives of the women in the videos). Discounting rates were consistently higher from the perspective of the woman observed smoking suggesting that choices on delay discounting tasks may be influenced by observation of intertemporal choice. |
| |
|
| 106. Choice, Timing, Motivation, and Memory in the FMR1 KO Rat |
| Area: EAB; Domain: Basic Research |
| Bryana A Thieret (St. Lawrence University), William DeCoteau (St. Lawrence University), ADAM E. FOX (St. Lawrence University) |
| Discussant: Casey Marsh (University of South Florida) |
| Abstract: One gene that has been linked to Autism Spectrum Disorder (ASD) and developmental disorders more generally, is FMR1. The FMR1 gene is linked most closely with Fragile X syndrome, but loss of expression of the gene is thought to be one of the most common genetic causes of ASD and developmental disability broadly. In a series of experiments, FMR1 Knockout (KO) rats were exposed to temporal bisection, delay discounting, progressive ratio, and delay matching to sample tasks to characterize complex operant behavior in the preclinical model. We found no differences between FMR1 KO rats and controls on the temporal bisection or progressive ratio tasks, but FMR1 KO choice behavior was less sensitive to delay changes in the discounting task, and small differences emerged in the delay matching to sample task. These findings have implications for the clinical relevance of the FMR1 KO model. |
| |
|
| 107. Development of a Discounting Task for Choice Between Qualitatively Different Food Rewards |
| Area: EAB; Domain: Basic Research |
| HAYLEE DOWNEY (Virginia Tech (FBRI)), Jeffrey S. Stein (Virginia Tech (FBRI)) |
| Discussant: Brandon Patrick Miller (University of Kansas) |
| Abstract: Delay discounting refers to the tendency for delayed outcomes to be devalued. Delay discounting is commonly measured using choices for monetary outcomes, but more complex choices for qualitatively different, non-monetary outcomes may more accurately model the decision-making evident in lifestyle-related disease. To address this, we adapted the 6-trial adjusting delay discounting task to incorporate tradeoffs between food immediacy and healthiness. Specifically, 41 participants with healthy weight or obesity (data collection ongoing; additional data to be collected) chose between a less preferred fast-food meal that was available immediately and a more preferred home-cooked meal that was available after a delay (i.e., travel and preparation time). Across trials, the delay to the home-cooked meal was adjusted based on prior choices until reaching an indifference delay. Longer indifference delays reflect less discounting (i.e., greater willingness to wait). The median indifference delays were 2.94 h (IQR = 2.53) and 2.03 h (IQR =1.67) for the groups with healthy weight and obesity, respectively. Additional, more common delay discounting tasks were also examined and will be discussed. In future studies, if appropriate, we will further validate the meal task using other measures of eating behaviors. |
| |
|
| 108. COVID-19-Related Financial Scarcity Is Associated With Greater Delay Discounting but Not Probability Discounting |
| Area: EAB; Domain: Applied Research |
| HAYLEE DOWNEY (Virginia Tech (FBRI)), Roberta Freitas-Lemos (Virginia Tech (FBRI)), Kelsey Curran (Virginia Tech (FBRI)), Elena Serrano (Virginia Tech), George Davis (Virginia Tech), Jeffrey S. Stein (Virginia Tech (FBRI)) |
| Discussant: Sara Peck (USU; NECC; WNE) |
| Abstract: Prior studies indicate that scarcity may increase delay discounting, the devaluation of delayed rewards. Evidence on the effect of COVID-19 on delay discounting is mixed. Additionally, no study has examined the effect of COVID-19-related scarcity on probability discounting, the devaluation of probabilistic rewards. To examine the relationship between COVID-19-related scarcity, delay discounting, and probability discounting, we recruited 1012 participants with low income on Amazon Mechanical Turk in April 2020. Participants completed measures of delay and probability discounting of food and money, COVID-19-related financial impacts, and food security. Using linear regression, we found that compared to those with no COVID-19-related financial impacts, those with severe COVID-19-related financial impacts had greater delay discounting of money and greater delay discounting of a grocery gift card. Greater food insecurity in the past month was associated with greater delay discounting of a grocery gift card but not delay discounting of money. COVID-19-related financial impact was not associated with probability discounting. Combined with laboratory experiments, this study provides additional evidence for the idea that feelings of scarcity may increase delay discounting. However, because this study was observational, no assumptions of causality can be made about the specific effect of COVID-19 on delay discounting. |
| |
|
| 109. An Evaluation of the Distribution of Follow-Up Sessions on Skill Maintenance |
| Area: EAB; Domain: Applied Research |
| MOLLY MATTES (Western Michigan Universtiy ), Sacha T. Pence (Western Michigan University) |
| Discussant: Brandon Patrick Miller (University of Kansas) |
| Abstract: For behavior change to be effective, it not only needs to occur across persons and settings, but it needs to maintain over time. The distribution of follow-up sessions may influence maintenance of learned skills. Follow-up sessions can be conducted using an equal distribution where the number of days between follow-up sessions stays consistent. Follow-up sessions can also be conducted using a progressive distribution where the number of days between follow-up sessions gradually increase. The experimenters used a multielement design to evaluate follow-up session distribution (control, equal distribution, and progressively increasing distribution) on the maintenance of an arbitrary labeling task taught to nine college-aged students. Following acquisition, the participants completed virtual follow-up sessions across at least 30 days. Both follow-up session distributions conditions were equally effective on skill maintenance for six participants. The progressively increasing distribution condition was associated with higher maintenance for two participants and the equal distribution condition had higher maintenance with one participant. Across all participants, some distribution of opportunities to practice the skill led to higher levels of maintenance when compared to the control condition during which no opportunities to practice were provided. |
| |
|
| 110. Methods for Calculating AUC with Discounting Data |
| Area: EAB; Domain: Basic Research |
| KATILYN MARIE ASHLEY TREEM (Georgia Southern University), Charles Casey Joel Frye (Portland, Oregon), Jonathan E. Friedel (Georgia Southern University) |
| Discussant: Sara Peck (USU; NECC; WNE) |
| Abstract: Delay discounting describes how an outcome loses value as the delay to the outcome increases. Area-under-the-curve (AUC) is an atheoretical measure of discounting. One problem with AUC is that it is not always calculated consistently (Borges et al., 2016). For example, if there is no data point obtained when a participant does not have to wait, then that data point should be interpolated. It is not reliably reported in the literature whether people interpolate that data point. Additionally, there is not a standard method for calculating the AUClog when there is an indifference point when the delay equals zero (Borges et al., 2016). Here new methods are proposed are to calculate all forms of AUC with an associated R package (a free data analysis tool) to do those calculations. The R package provides a consistent, common ground for researchers to accurately compute all versions of AUC.
References
Borges, A. M., Kuang, J., Milhorn, H., & Yi, R. (2016). An alternative approach to calculating area-under-the-curve (AUC) in delay discounting research. Journal of the Experimental Analysis of Behavior, 106(2), 145–155. https://doi.org/10.1002/jeab.219 |
| |
|
| 111. The Effect of Social Skills Instructions on Peer Interactions for Preschoolers With Social Withdrawal |
| Area: EAB; Domain: Applied Research |
| XIN DONG (Nicholls State University), Julie L. Thompson (Texas A&M University) |
| Discussant: Brandon Patrick Miller (University of Kansas) |
| Abstract: Developing social and competence in early childhood is one of the most important aspects of social development with long lasting outcomes (McCabe & Altamura, 2011). Children with social withdrawal often experience peer rejection and victimization (Oh et al. 2008). Children with social withdrawal also have less social interactions with peers which reduce the opportunity for them to practice play skills, communication skills and social skills with their peers. This study replicated a previous study on the effect of social skills instructions on peer interactions (Storey et al., 1994). A multiple probe across participants design was used to evaluate the effect of the social skills intervention package including social skills instruction with peer-mediated component and reinforcements on the social interactions of three preschoolers with social withdrawal. All participants showed increase in social initiations after the intervention was introduced. Slight changes in social responses were observed. |
| |
|
| 112. The Impact of Visual Stimuli in a Behavioral Variability Task |
| Area: EAB; Domain: Basic Research |
| JOCELYN HANSSON (Eastern Michigan University ), James T. Todd (Eastern Michigan University), Claudia Drossel (Eastern Michigan University), Thomas J. Waltz (Eastern Michigan University) |
| Discussant: Sara Peck (USU; NECC; WNE) |
| Abstract: Research and theory suggest behavioral variability is important for learning and problem-solving (Neuringer, 2002). Studies have used simple computer programs, where participants earn points based on their behavior, to study human behavioral variability (e.g., Hansson & Neuringer, 2018). The present study aimed to understand methodological differences between studies on this topic by investigating whether visual stimuli affected participant variability. Additionally, participant verbal behavior was assessed. In the present study, 24 participants used two on-screen buttons to move a circle across the screen in 5 steps, after which a point was awarded depending on the participants’ behavior. Analyses compared levels of observed behavioral variability between two visual conditions: one where participants could see a pyramid depicting all possible paths they could move their avatar, and one where they saw an open field. Analyses also compared two contingencies: one where points were awarded for variable behavior (VAR) and one where points were awarded irrespective of variability (YOKE). Results showed that contingency, but not visual condition affected variability levels; participants who were reinforced for varying engaged in more variable behavior. Participants’ post-game self-reported strategy use are also explored. Implications and directions for future research are discussed. |
| |
|
| 113. A Cluster Analysis of Variables Maintaining Internet Use |
| Area: EAB; Domain: Basic Research |
| EZEKIEL TORRES (California State University East Bay ), Rinisha Naidu (California State University-East Bay), Salma Garcia (California State University East Bay), Liz Kyonka (California State University - East Bay) |
| Discussant: Brandon Patrick Miller (University of Kansas) |
| Abstract: Characterizing typical patterns of internet use and determining their relationships to dependent internet use in a general population may help to identify behavioral markers of internet addiction. A nationally representative sample of 326 Americans completed the Young Diagnostic Questionnaire (YDQ), which assesses internet dependence, and the preliminary Internet Consequences Questionnaire (ICQ-P), which identifies variables potentially maintaining internet use. Hierarchical cluster analysis with ICQ-P scores for 6 categories of reinforcement identified five distinct subgroups of internet users: Procrastinators (N = 83) endorsed escape from demand and private events more than other categories. All-around moderate users (N = 99) had neither the highest nor lowest scores for any category of reinforcement. Moderate social users (N = 54) had relatively low endorsement of all categories except attention. Intensive users (N = 54) had relatively high endorsement of all categories. Asocial users (N = 56) endorsed attention and social avoidance less than other categories. YDQ scores differed across the five clusters, indicating that internet dependence was least likely in moderate social users and most likely in intensive users. Results suggest that people who report deriving the most reinforcement from using the internet may be at greater risk for internet addiction. |
| |
|
| 114. A Comparison of Demand for Sleep Among Undergraduate Students Based on Shift Work Status |
| Area: EAB; Domain: Basic Research |
| CRYSTAL YUET WEN LIM (Eastern Michigan University), Kayla Rinna (Eastern Michigan University), Thomas J. Waltz (Eastern Michigan University), Claudia Drossel (Eastern Michigan University) |
| Discussant: Sara Peck (USU; NECC; WNE) |
| Abstract: Individuals working outside of the typical 8 am to 5 pm work hours, otherwise known as “shift workers” tend to have inconsistent sleep schedules which may impact quality of sleep. This study is an exploratory analysis of data collected from a series of hypothetical purchase tasks for sleep among college undergraduates. Undergraduate participants (n = 88) were asked how much they would pay in exchange for guaranteed hour(s) of restful sleep with the incentive of coverage at work and without losing pay on their next paycheck. Comparisons across those who worked shift hours (n = 54) with non-shift workers (n = 34) found overall demand was higher among shift workers, as demonstrated by greater expenditure (Omax, Pmax), a greater area under the curve, and greater alpha values. Our findings suggest that demand for restful sleep was greater among shift workers than non-shift workers, as shift workers were more likely to purchase hours of sleep even when prices increased. |
| |
|
| 115. Probability Learning in Two Versions of an Iterated Monty Hall Problem |
| Area: EAB; Domain: Basic Research |
| TRYPHENA ARTHUR (California State University, East Bay), Ashley Martinez (California State University, East Bay), Ezekiel Torres (California State University, East Bay), Liz Kyonka (California State University - East Bay) |
| Discussant: Brandon Patrick Miller (University of Kansas) |
| Abstract: We investigated effects of inflating probability of winning on switching in the Monty Hall problem. Undergraduate students participated for course credit. On each trial, a participant chose one of three boxes on a computer screen. Next, a box was removed and the remaining two boxes turned green. The participant could stay with the box they chose or switch to the other available box. The optimal strategy was to switch every trial. Choosing the preselected prize box won a point. An enhanced version of the task with an inflated probability of winning for switching should produce higher rates of switching. The probability of winning if the participant switched was 2/3 in the standard phase and 9/10 in the enhanced phase. The order of phases was counterbalanced across participants. Most participants learned to switch over 100 trials. Compared to participants who started with the standard phase, participants who started with the enhanced phase switched more at the end of the first phase, but became less likely to switch in the standard phase. The timing of the introduction of the enhanced version affected performance such that initial exposure to the standard task increased the strategic benefits of experiencing the enhanced task. |
| |
|
| 116. Beyond Lab Lore: Studying the Learning That Occurs During Pre-Experimental Training |
| Area: EAB; Domain: Basic Research |
| OSMAR LOPEZ (University of Nevada, Reno), Matthew Lewon (University of Nevada, Reno), Christina M. Peters (Northwest Behavioral Associates), Melanie S Stites (University of Nevada, Reno), Emily Danielle Spurlock (University of Nevada, Reno), Linda J. Parrott Hayes (University of Nevada, Reno) |
| Discussant: Sara Peck (USU; NECC; WNE) |
| Abstract: Studies of operant behavior with nonhumans typically involve pre-experimental training. Part of this involves getting a specified response to occur so it can be reinforced, but there is additional learning that must occur before studies of operant behavior can begin. Pre-experimental training is rarely described in detail in published studies, and this training has been described as “lab lore,” or the idiosyncratic procedures developed within individual laboratories. This is unfortunate because a) variability in pre-experimental training (e.g., in “hand-shaping”) within and across laboratories may compromise the generalizability of results and b) what happens during pre-experimental training may allow for the study of important learning processes that are prerequisite to subsequent learning. We present data from several iterations of fully automated pre-experimental training procedures developed in our laboratory for mice as a first step toward systematically describing the learning that occurs before experiments begin. Results show the impact of different training contingencies, reinforcer types (pellets vs. liquid), operanda (lever vs. nose poke), and motivating operations (food vs. water satiation/deprivation). We also attempt to identify the characteristics of animals that do and do not ultimately learn to engage in the specified operant response during these procedures. |
| |
|
| 117. Effects of Multiple Response Options on Target Responding in a Resurgence Procedure |
| Area: EAB; Domain: Basic Research |
| KAYLA BRACCIO (UNCW), Emma Auten (University of North Carolina Wilmington), Elizabeth Paige Thuman (University of North Carolina Wilmington), Emily L. Baxter (University of North Carolina Wilmington) |
| Discussant: Brandon Patrick Miller (University of Kansas) |
| Abstract: Resurgence is the recurrence of a previously reinforced target behavior following a decrease/removal of reinforcement of an alternative behavior. Typically, studies examining resurgence use 2-3 responses in the procedure: a target, alternative, and inactive control response. Across extinction, is expected that the target behavior will decrease across time. It has been observed in human operant studies, however, that target responding persists. The goal of the current study was to test a human operant resurgence procedure that included several inactive control responses, to determine if extinction of the target behavior can be observed following resurgence. Throughout the study, multiple responses were available for participants to respond on. In Phase 1, points were provided for responding on one of the multiple available activities (i.e., target response). In Phase 2, points were provided for responding to a different activity (i.e., alternative response). In Phase 3, all behaviors were placed on extinction. Engagement with all activities was measured across all sessions. It is hypothesized that resurgence of the target behavior would occur initially, followed by complete extinction of the target behavior. Results and implications of human operant resurgence procedures will be discussed. |
| |
|
| 118. Generalized Same-Different Learning With Pigeons in a Slot-Machine Gambling Analogue |
| Area: EAB; Domain: Basic Research |
| MICHAEL ALLEN LOSI (Reed College), Manish Vaidya (University of North Texas), Timothy D. Hackenberg (Reed College) |
| Discussant: Sara Peck (USU; NECC; WNE) |
| Abstract: As part of a larger project developing a laboratory model of slot-machine gambling, we trained 12 pigeons to discriminate stimuli preceding a win (trials ending in food reinforcement) from stimuli preceding a loss (no reinforcement). The stimuli in each trial consisted of a sequence of 3 colored boxes, presented in succession in a horizontal row, designed to mimic spinners in a slot-machine gambling device. We first trained a series of conditional discriminations based on sameness/difference: responses on one key were differentially reinforced with food if all 3 stimuli were the same (S+) and on a second key if the stimuli were different (S-). The pigeons readily learned these same-different conditional discriminations with 30 different colors in baseline (mean accuracy = 91%, range = 86%-96%). We then tested for generalized responding by interweaving 30 novel stimuli (new colors) with the training stimuli. Mean accuracy on these probe trials was 86% (range = 78%-94%), significantly higher than expected by chance, demonstrating generalized same-different responding in a novel relational learning task. Having thus established conditional discriminative control based on similarity/difference, the pigeons subsequently served in an experiment examining the effects of “near misses” in a slot-machine gambling analogue. |
| |
|
| 163. No Evidence That “Near Misses” Increase Gambling in a Token-Based Laboratory Model With Pigeons |
| Area: EAB; Domain: Basic Research |
| MICHAEL ALLEN LOSI (Reed College), Timothy D. Hackenberg (Reed College) |
| Discussant: Sally L. Huskinson (University of Mississippi Medical Center) |
| Abstract: In slot machine gambling devices, “near misses” are stimulus configurations that approximate win trials, and presumed to function as conditioned reinforcers through stimulus generalization with win trials. Although plausible, empirical support for the near-miss effect is shaky at best. The present experiment explored the near-miss effect in a token-based laboratory model of slot-machine gambling. Eleven pigeons chose between (a) earning tokens, (b) producing exchange periods in which the tokens could be exchanged for food reinforcers, and (c) gambling a token for a chance to win more tokens. Each gamble choice cost 1 token for a 30% chance to win 3 tokens. The outcome of a gamble choice was preceded by a sequence of three colored stimuli, designed to mimic the spinners in a slot-machine gambling apparatus. Wins were signaled by 3 matching stimuli, all the same color (e.g., red-red-red), and losses by 3 non-matching stimuli, all different colors (e.g., red-green-white). In some conditions, 50% of the loss trials losses were near misses, signaled by stimulus sequences containing two matching stimuli. Gambling choices were no more frequent when loss trials included near misses than when they did not, calling into question the conditioned reinforcing functions of near misses. |
| |
| |