IT should be notified now!

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.

Donate to SABA Capital Campaign
Portal Access Behavior Analysis Training Directory Contact the Hotline View Frequently Asked Question
ABAI Facebook Page Follow us on Twitter LinkedIn LinkedIn

42nd Annual Convention; Downtown Chicago, IL; 2016

Event Details

Previous Page


Poster Session #471
Tuesday, May 31, 2016
12:00 PM–2:00 PM
Riverside Exhibit Hall, Hyatt Regency, Purple East
Chair: John Bai (University of Auckland)
1. Brave the Pain but Savor the Pleasure? Empirical-Normative Discrepancies in Preferences for Single Outcomes of Losses and Sequences of Gains
Area: TPC; Domain: Basic Research
PRZEMYSLAW SYLWESTER MARCOWSKI (SWPS University of Social Sciences and Humanities), Wojciech Bialaszek (SWPS University of Social Sciences and Humanities), Pawel Ostaszewski (SWPS University of Social Sciences and Humanities)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: According to the standard microeconomic theory, postponing and dividing losses and receiving integrated gains as soon as possible should be the rational course of action. We analyzed the behavior of 197 undergraduate students to investigate the impact of the form of gains and losses (single-package or sequenced) on delay discounting. Particularly, we aimed to determine whether there is indeed a preference for sequenced losses and single-package gains – as per the normative theory. To test our hypotheses we used a dynamic multiple-staircase discounting procedure. Participants chose between alternatives consisting of sequences with constant value and adjusting immediate option – where the immediate option increased or decreased each time it was chosen for losses and gains, respectively. Discounting rates were then calculated as areas under the discounting curve. Interestingly, we found that sequenced payoffs were discounted considerably less steeply compared to their single-package equivalents for both gains and losses. This illustrates the preference for immediate single packages of losses and delayed sequences of gains – which seems to contradict the normative microeconomic theory in both domains. We therefore propose that, at least to some degree, sequenced losses behaviorally act as more aversive, while sequenced gains act as more rewarding.
2. Super-Resurgence? Investigating ABC Super-Resurgence Effects
Domain: Basic Research
CHRISTOPHER OHEARN (West Virginia University), Tyler Nighbor (West Virginia University), Stephanie L. Kincaid (Marcus Autism Center), Kennon Andy Lattal (West Virginia University)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: Super-resurgence is a combination of renewal and resurgence procedures developed by Kincaid et al. (2015). Nighbor et al. (2015) used a concurrent schedule with ABC super-resurgence procedure on one key and an AAA control procedure on the other key. The AAA control procedure produced a larger resurgence effect than the ABC procedure. The current investigation replicated the ABC super-resurgence procedure using three naïve pigeons in a single schedule in the absence of the concurrent AAA control procedure. A resurgence effect was found for all subjects in the C component, but a larger resurgence effect was found following a return to the A context. In a second experiment, the procedure of Nighbor et al. (2015) will be replicated in an attempt to further disambiguate their findings.
3. A Menstrual Cycle Phase-Effect on Loss Aversion: An Initial Investigation Using a Concurrent-Operants Method
Domain: Basic Research
MARCIA VENTURA (Brigham Young University), Diego Flores (Brigham Young University), Frank Robertson (Brigham Young University), Michael Seeley (Brigham Young University), Savannah Keenan (Brigham Young University), Venice Jardine (Brigham Young University), Jordan Sgro (Brigham Young University), Harold L. Miller Jr. (Brigham Young University)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: Normal hormonal fluctuations produce measurable, differential outcomes in experiments with female participants. However, the effects of circulating gonadal steroid hormones and the menstrual cycle on human decision-making, specifically, loss aversion, remain undetermined. We used the SubSearch Game to examine loss aversion to determine if womens loss-averse behavior in a monetary gain/loss procedure varied as a function of the menstrual cycle. Twenty-five college-age, regularly-cycling females participated in 12 sessions wherein they played the SubSearch videogame. Each session corresponded to either menses onset, ovulation or the mid-luteal phase. The SubSearch Game involves a concurrent-operants method in which the player uses a mouse to move a submarine icon to retrieve underwater objects. The screen is divided vertically in half. The player can switch between the half-screens at any point. Occasionally, according to concurrent variable-interval variable-interval (VIVI) schedules, the retrieval of an object results in the delivery of points via an on-screen counter and which are exchanged for money following the session. Retrieval may also produce the loss of points. Each session consisted of four 9-min components in which the reinforcer ratio varied, as did the background color on the screen. Punishers were delivered in half of the components.
4. The Effects of Conventional Extinction and Variable Time Schedules on Differential Reinforcement of Low-Rate Behavior Responding
Domain: Basic Research
CHRISTIAN YENSEN (West Virginia University), Tyler Nighbor (West Virginia University), Alex Cutlip (West Virginia University), Kennon Andy Lattal (West Virginia University)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: Differential responding of low-rate behavior (DRL) schedules are constructed so that only responses separated by t seconds or more from the previous response receive reinforcement. Low rate behavior has been found to be more resistant to extinction. Outside of removing reinforcement entirely, extinction can also be accomplished by providing response-independent reinforcement. The purpose of the present experiment was to evaluate the effects of conventional extinction and variable-time (VT) extinction on DRL maintained responding. In the VT extinction component, the length of the VT was yoked to reinforcement rates in the preceding baseline. For 3 of 4 subjects, responding extinguished faster in conventional extinction component than VT component. For 2 subjects, Inter-response time distributions were very similar in VT component to DRL 15-s even though the contingency changed. A confound of experiment 1 was that the reinforcement rates in the VT component varied across subjects, making inter-subjective comparisons difficult. Experiment 2 will involve a return the DRL baseline and will attempt to eliminate the primary confound in experiment 1 by equating the VT reinforcement rates across subjects.
5. Persistence of Behavior During Differential Reinforcement
Domain: Basic Research
KAREN SLUTER (University of Waikato), Therese Mary Foster (University of Waikato), James McEwan (The University of Waikato), Timothy Edwards (University of Waikato)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: Adding reinforcers into a context, whether contingent on behavior or not, typically results in increased persistence of behavior in extinction. Increases in reinforcers occur when differential reinforcement of alternative behavior is used to reduce problem behaviors. Whilst there is often success at reducing problem behaviours, research also suggests that persistence of the problem behaviour may actually increase due to the increased reinforcement in the context. Training the alternative behavior in a separate context does not increase the reinforcers in the target context and can prevent the increased persistence of the problem behavior when the alternative behaviour is introduced. An analogue of this procedure, with domestic hens, confirmed that training the analogue alternative behavior in a separate context resulted in the analogue of the problem behavior being less persistent during extinction than when the alternative had been trained in the same context (i.e., under traditional differential reinforcement conditions). It is however, often unfeasible to implement extinction completely in applied contexts, so investigation as to whether such differences in the persistence of the problem behavior would be seen with other disruptors, such as alternative sources of reinforcement, were carried out.
6. The Effects of Lag Schedules and Multiple Response Alternatives on Response Resurgence
Domain: Basic Research
ASHLEY BAGWELL (University of Texas at Austin), Joel Eric Ringdahl (University of Georgia), Terry S. Falcomata (The University of Texas at Austin)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: The mitigation of response resurgence is a topic which has garnered recent attention due to its importance in a clinical setting. The present study examined the mitigation of response resurgence in a human operant study using a computer program to teach multiple response alternatives using a Lag 3 schedule of reinforcement. In the first of three phases, a target response was trained. In the second phase the target response was placed on extinction and trials alternated between the single alternative component and the multiple alternative component. In the multiple alternative component, responses were reinforced on a Lag 3 schedule of reinforcement. The rate of reinforcement obtained during the multiple alternative component was used to determine the rate of reinforcer delivery in the single alternative component. In the final phase, all responses were placed on extinction. Of the six undergraduate students who participated in this study, three came under the control of the programmed contingencies. All three of these participants demonstrated higher rates of resurgence in the component which simulated a single response alternative when compared to the component which simulated multiple response alternatives. Potential clinical implications and areas for future study are discussed.
7. Not Quite the Same: Immediate and Impending Threats Exert Different Levels of Aversive Control in Humans
Domain: Basic Research
OWEN JAMES ADAMS (University of North Texas), Vanessa Lopez (University of North Texas), Thomas Wright (University of North Texas), Sandy Magee (University of North Texas), David M. Richman (Texas Tech University), Simon Dymond (Swansea University), Michael W. Schlund (University of North Texas)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: In this investigation, we used an approach-avoidance paradigm to examine how different methods of presenting cues correlated with aversive stimulus presentation (or threat) modulate control by positive reinforcement. Using a between groups design, we examined the effects of impending or immediate threat on avoidance behavior. Two versions of a novel approach-avoidance choice task were used. On a trial, both tasks presented a monetary reward alongside a discriminative stimulus (CS+ threat) that signaled the current probability of a money loss (range= 0-1.0). Pressing an approach button produced the reward or probabilistic loss, while pressing an avoidance button prevented loss. Each trial of the Immediate Threat task presented a CS threat level during the choice period. In contrast, each trial of the Impending Threat task began with the lowest CS threat level which increased one level every 2 s until the choice period. Results showed impending compared to immediate threat was associated with increased avoidance to low threats and decreased avoidance to higher threats. Choice reaction times also showed a decreasing trend under impending threat while reaction times showed an increasing trend under immediate threat. These findings suggest aversive control and avoidance in humans is modulated by how threats are encountered.
8. Effects of Delayed Reinforcement and Response-Independent Food on Resitance to Change
Domain: Basic Research
FLÁVIA FERREIRA (Universidade de Brasilia), Josele Abreu Rodrigues (Universidade de Brasilia), Carlos Renato Xavier Cançado (Universidade de Brasilia, Brazil), Raquel Moreira Aló (Universidade de Brasília, Brazil)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: Two experiments with four rats each were conducted to investigate the resistance to change of responding maintained under conditions of response-independent food and delayed reinforcement. In both experiments, lever pressing was maintained under a three-component multiple schedule in baseline. Variable interfood intervals programmed the same rate of food in each component. In components 100% and 10%, the percentage of response-dependent and immediate food was 100 and 10, respectively. In Component Delay, a tandem variable-interval (VI) fixed-time (FT) schedule (Experiment 1) or a tandem VI differential-reinforcement-of-other-behavior (DRO) schedule (Experiment 2) was in effect. Across baseline sessions, the delay value (i.e., FT or DRO) was yoked to that obtained in Component 10%. Responding was disrupted by extinction tests in both experiments. Responding generally was more resistant to change in Component 10% then in Component 100% in both experiments. In Experiment 1, resistance to change in Component Delay was assistematic. In Experiment 2, for three rats, resistance to change in Component Delay was similar to that in Component 10%. These results indicate that resistance to change can be affected by conditions in which the response-reinforcer relation is altered but reinforcement rate is equated across multiple-schedule components.
9. Is Bigger Better? Effects of Increasing Reinforcer Magnitude on Human Approach-Avoidance
Domain: Basic Research
KAYKAY MCELWRATH (University of North Texas), Thomas Wright (University of North Texas), Vanessa Lopez (University of North Texas), Owen James Adams (University of North Texas), Sandy Magee (University of North Texas), David M. Richman (Texas Tech University), Simon Dymond (Swansea University), Michael W. Schlund (University of North Texas)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: Every individual has a tolerance level for environmental threat and aversive stimulation that aids self-preservation. As threat intensity escalates, a tipping point is reached whereby behavior switches from being under the control of positive reinforcement (approach) to negative reinforcement (avoidance). Approach-avoidance conflict paradigms are commonly used to understand the competition between appetitive (e.g., food) and aversive (e.g., shock) contingencies. In this investigation, we examined the effects of increasing the magnitude of positive reinforcement on human choice to approach or avoid (N=5). We developed a novel approach-avoidance task where a monetary reward appeared in the presence of a conditioned stimulus (CS+ threat) that signaled increasing probability of a money loss. Across trials, reward was fixed while CS threat level varied unpredictably. Approach produced the reward or probabilistic loss, while avoidance prevented loss. Results showed increasing the CS threat level produced the desired switch from approach to avoidance. However, increasing the magnitude of positive reinforcement for approach only marginally increased approach responding. These results provide important insights into the competition for control over behavior that can occur between appetitive and aversive contingencies and highlights the disproportional control often exerted by aversive events.
10. Resistance to Change of Operant Variability: A Parametric Analysis
Domain: Basic Research
Raquel Moreira Aló (Universidade de Brasília, Brazil), JOSELE ABREU RODRIGUES (Universidade de Brasilia), Carlos Renato Xavier Cançado (Universidade de Brasilia, Brazil), Adam H. Doughty (College of Charleston)
Discussant: Rogelio Escobar (National Autonomous University of Mexico)
Abstract: Resistance to prefeeding was studied under three different variability requirements. Across three baseline conditions, four lever-press sequences by rats were maintained on a two-component multiple schedule. In the VAR component, a threshold contingency of 0.1 was in effect. In the REP component, only one sequence was reinforced. In the YOKE component, no variability requirement was in effect. Reinforcement probabilities were equated across multiple schedule components in each baseline condition. In baseline, U values were greater in the VAR than the REP and YOKE components, and similar between the latter two. In the prefeeding test after the multiple VAR YOKE baseline, resistance of U values was greater in the VAR than in the YOKE component. In the prefeeding test after the multiple REP YOKE baseline, resistance was similar between schedule components. Finally, in the prefeeding test after the REP VAR baseline, resistance of U values was greater in the VAR than in the REP component. Thus, baseline variability levels were directly related to the behavioral persistence, regardless of whether this level was required (VAR and REP) or allowed (YOKE) by reinforcement contingencies in each schedule component.
11. Avoidance Behavior in the Madagascar Hissing Cockroach
Domain: Basic Research
PAUL THOMAS THOMAS ANDRONIS (Northern Michigan University), Collin Hahn (Northern Michigan University), Morghan Minnick (Northern Michigan University), Leslie Smith (Northern Michigan University)
Discussant: Harold Miller, Jr. (Brigham Young University)
Abstract: The Madagascar Hissing Cockroach (MHC) (G. portentosa) has been studied extensively by ethologists, with a principal focus on several innate behaviors, like mating behavior and the eponymous alarm call they emit when startled or in distress. These animals have the potential of becoming an advantageous animal in operant research, particularly at a time when vertebrate laboratories are becoming prohibitively expensive and the regulatory environment increasingly onerous. With their simple nervous systems, they might also become excellent subjects for coordinated behavioral-neurobiological inquiries. While there seems to be a growing interest in this species among operant researchers, there is relatively little lab lore nor relevant behavioral information about the MHC. The present study represents an initial foray into simple avoidance behavior by these animals. The experiment involves simple choice in a T-maze between escape arms with either light or an electromagnetic field.
12. Reinforcement Rate and Resurgence in the Within-Session Procedure
Domain: Basic Research
SHUN FUJIMAKI (Keio University; JSPS), Takayuki Sakagami (Keio University)
Discussant: Harold Miller, Jr. (Brigham Young University)
Abstract: The present experiment examined the relation between alternative rates of reinforcement and resurgence by using the within-session procedure. Each session consisted of the following three phases: In the Acquisition phase, lever-pressing response of rats was reinforced on a variable-interval (VI) 20-s schedule until 40 reinforcers were delivered. In the Elimination phase, this target response was eliminated while the alternative response to the other side lever was reinforced according to either VI 5-s, 20-s, or 80-s. This phase lasted for until 40 reinforcers were delivered and the number of target response decreased less than 3 for each of the last three 30-sec bins. The Resurgence phase ended after 10-min during which all reinforcers were withheld. Rats were exposed to each of three alternative reinforcement conditions (i.e., VI 5-s, VI 20-s, or VI 80-s) for five times, but the order of each condition was randomly assigned for each session. Although all rats showed resurgence in all conditions (Figure 1, left panel), the magnitude of resurgence increased as a function of the reinforcement rates (Figure 1, right panel). These findings were consistent with previous studies and supported the prediction of models of resurgence based on behavioral momentum theory.
13. Behavioral Effects of Delayed Timeouts From Reinforcement
Domain: Basic Research
THOMAS P. BYRNE (Massachusetts College of Liberal Arts), Alan D. Poling (Western Michigan University)
Discussant: Harold Miller, Jr. (Brigham Young University)
Abstract: Timeouts are sometimes used in applied settings to reduce target responses, and in some circumstances delays are unavoidably imposed between the onset of a timeout and the offset of the response that produces it. The present study examined the effects of signaled and unsignaled timeouts in rats exposed to concurrent fixed-ratio 1 fixed-ratio 1 schedules of food delivery, where each response on one lever, the location of which changed across conditions, produced both food and a delayed 10-s timeout. Delays of 0-38 s were examined. During longer delays, multiple reinforcer deliveries were available. Delayed timeouts often, but not always, substantially reduced the number of responses emitted on the lever that produced timeouts relative to the number emitted on the lever that did not produce timeouts. In general, larger effects were observed with signaled timeouts. These results demonstrate that delayed timeouts, like other delayed consequences, can affect behavior, albeit less strongly than immediate consequences.
14. Waiting for Resurgence: Resurgence Following Delayed Reinforcement
Domain: Basic Research
TYLER NIGHBOR (West Virginia University), Christian Yensen (West Virginia University), Kennon Andy Lattal (West Virginia University)
Discussant: Harold Miller, Jr. (Brigham Young University)
Abstract: Resurgence is the recurrence of a previously reinforced operant when another operant is placed on extinction. Although conventional extinction consistently produces resurgence, the question of whether local periods of extinction may produce resurgence has been less frequently addressed (e.g., Lieving & Lattal. 2004). The purpose of the current investigation was to evaluate if local periods of extinction in the form of signaled delays produced resurgence. In experiment 1, three pigeons served as subjects. During the initial phase, key-pecking on two keys was reinforced under alternating VI 60-s schedules. Following, during an alternative reinforcement phase, key-pecking was extinguished on one of the two keys and reinforced under a tandem VT 10-s FI 60-s schedule on the other key (technically concurrent tandem [VT 10-s FI 60-s] extinction). During the resurgence test, rather than conventional extinction, the tandem schedule from the previous phase was converted to a chain VI 10-s FT 60-s schedule (technically concurrent chain [VI 10-s FT 60-s] extinction, and the key went dark for the remainder of the 60s. Resurgence was found for 3 of 3 pigeons, showing that resurgence may occur following fixed delays, or another form of local extinction.
15. Effects of Aversive and Appetitive Stimuli on Conditioned Place in Madagascar Hissing Cockroaches
Area: AAB; Domain: Basic Research
LINDA MUCKEY (Southern Illinois University Carbondale), Matthew L. Johnson (Southern Illinois University Carbondale), Ashley Shayter (Southern Illinois University Carbondale), Mark R. Dixon (Southern Illinois University Carbondale)
Discussant: Harold Miller, Jr. (Brigham Young University)
Abstract: The following study was conducted in attempt to address the effects of competing aversive and appetitive stimuli on the conditioned place paradigm. Madagascar Hissing Cockroaches were utilized as an animal model for the extension and further understanding of this concept. The experimental apparatus was sectioned in half with coarse sandpaper on one side and fine sandpaper on the other side acting as the tactile discriminative stimuli. Fans underneath the apparatus directed air through holes in the apparatus floor, which would then act as the aversive stimulus. The amount of time spent in each half acted as the dependent variable for considering place preference or avoidance. The aversive stimulus was presented simultaneously with the appetitive stimulus in a specified half of the apparatus. An initial preference assessment was conducted for determining the subjects’ relative food preferences, later to be used as the appetitive stimuli. Based on preference assessment data, appetitive stimuli were classified as high preference or low preference. Stimuli within each preference level were tested in an effort to determine if presentation of high preference or low preference stimuli would exert more control over conditioned place.
15a. Further Analyses of Response-Reinforcer Dependency and Resistance to Change
Domain: Basic Research
Carlos Renato Xavier Cançado (Universidade de Brasilia, Brazil), FLÁVIA FERREIRA (Universidade de Brasilia)
Abstract: The effects of different percentages of response-dependent food on resistance to change were investigated in two experiments with rats. In Experiment 1, percentages of 100, 50 and 10 of response-dependent food were in effect in a three-component multiple schedule in baseline. In components 50% and 10%, dependent and independent food were programmed dependently. Responding was disrupted by a variable time 30 s in each component. In Experiment 2, the effects on resistance to change of how food was programmed were investigated by using a three-component multiple schedule. In one component, 100% of the food was response-dependent and in the other two, 10% of the food was response-dependent. In components 10%-I and 10%-D, respectively, dependent and independent food were programmed independently and dependently. Responding was disrupted by extinction tests. In Experiment 1, responding generally was more resistant to change in Component 10% than in components 50% and 100%. In Experiment 2, resistance to change was similar in components 10%-I and 10%-D, and greater in these components than in Component 100%. These results indicate that resistance to change is affected by the response-reinforcer dependency and that the effect is not specific to how the response-dependency was programmed and how responding was disrupted.
17. Gotta Bad Feeling: Sustained Fear Responses to Conditioned Aversive Stimuli but Not Appetitive or Neutral Stimuli
Domain: Basic Research
THOMAS WRIGHT (University of North Texas), Zach Wingfield (University of North Texas), Aidan Bennawy (University of North Texas), Tilija Stanojevic (University of North Texas), KayKay McElwrath (University of North Texas), David M. Richman (Texas Tech University), Simon Dymond (Swansea University), Michael W. Schlund (University of North Texas)
Discussant: Harold Miller, Jr. (Brigham Young University)
Abstract: Fear conditioning procedures are widely used to study fear and anxiety. During fear conditioning (FC), a neutral stimulus is paired with delivery of an aversive stimulus, such as electric shock or money loss. Over trials, the neutral stimulus becomes a conditioned aversive stimulus (CS+) that elicits fear indexed by an increase in autonomic responses, in particular, skin-conductance responses (SCRs). A second neutral stimulus never paired with an aversive stimulus (CS-) is used as a control. In this investigation, we used a within-subject design (N=20) to examine two questions about FC and SCRs: (1) Can SCRs be maintained to a 12 s CS+? (2) Do equivalent SCRs occur to a CS paired with aversive money loss and a CS paired with money gain? We found that a 12 s CS+ presentation produced a sustained SCR, but a 12 s CS- presentation and 12 s CS paired with money gain did not. These findings suggest SCRs can be maintained for a substantial time period, which may index anxiety rather than fear, and SCRs were restricted to stimuli paired with an aversive stimulus.
18. Consistent and Inconsistent Treatment Integrity Failures During Differential Reinforcement of Alternative Behavior
Area: EDC; Domain: Applied Research
GABRIELLE MESCHES (West Virginia University), Lucie Romano (West Virginia University), Claire C. St. Peter (West Virginia University)
Discussant: Harold Miller, Jr. (Brigham Young University)
Abstract: Treatment integrity failures are commonly studied by measuring response rates when errors are consistently made during intervention implementation. For example, St. Peter Pipkin and Vollmer (2010) programmed commission errors by specifying a constant probability of reinforcement for problem behavior during an intervention based on differential reinforcement of alternative behavior (DRA). Consistent commission errors have detrimental effects on DRA. The consistency of these errors may not always be the case in real-life situations. An intervention agent may implement treatment perfectly for a couple of days and then the next day or two implement treatment with low integrity. Effects of inconsistent treatment integrity have not yet been compared to those of consistent errors, but emerging evidence suggests that periodic exposure to interventions implemented with high integrity may reduce detrimental effects of low-integrity implementation. The current study uses nonclinical participants engaging in arbitrarily selected responses to evaluate consistent and inconsistent treatment integrity failures when commission errors occur during DRA applied to an arbitrary response. Results support emerging evidence by showing that inconsistent integrity failures are less detrimental to treatment effects than consistent integrity failures.
19. Amount of Instruction Information in the Solution of the Tower of London Task
Area: VRB; Domain: Basic Research
ROSALINDA ARROYO (Universidad Nacional Autónoma de México), Maria Luisa Cepeda Islas (FES Iztacala UNAM), Diana Moreno Rodriguez (FES Iztacala Universidad Nacional Autonoma de Mexico), Hortensia Hickman (Universidad Nacional Autonoma de México, FES-Iztacala), Maria Bautista (UNAM, FESI)
Discussant: Harold Miller, Jr. (Brigham Young University)
Abstract: In order to evaluate the effect of the amount of information provided before and during the solution of the Tower of London task. This study included three groups: General Instructions group, which presented global aspects of the task; Specific instructions group, which also described conditions of response and its relationship with feedback; and Precise instructions group, these included the previous aspects of the other groups but added to each trial the number of moves required. After training, all groups performed two tests, both without feedback or additional instructions, Test 1 only changed the number of movements required and Test 2 only color stimuli was changed. At the end of each phase they were asked: 1) What did you do to solve the task? and 2) How would you explain to another what to do to resolve it? The results show that all groups exceed by at least 40% of the trials the number of moves, however the general instructions produced lower latency. Verbal reports showed a relationship between the quality of the report and the type of group. These findings are discussed in light of the evidence in the area with matching to sample and reinforcement schedules.
20. Stimulus Clarity and Negative Conjugate Reinforcement
Domain: Basic Research
LAUREN JONES (University of Nevada, Reno), Daylee E. Brock (University of Nevada, Reno), Teal McAllister (University of Nevada, Reno), Patrick M. Ghezzi (University of Nevada, Reno)
Discussant: Harold Miller, Jr. (Brigham Young University)
Abstract: Schedules receiving little attention are schedules of covariation, specifically including conjugate reinforcement. Though far less studied, researchers have examined conjugate reinforcement from both a basic and an applied point of view. The present study follows a recent trend in toward basic research with humans by examining conjugate reinforcement as a function of the change in the clarity of a stimulus when responding falls below a certain rate. A six component MULT schedule was used. Each component was associated with a self-selected visual display(s) that diminished in clarity at different rates when responding fell below a pre-determined response rate. The question of interest was how these different rates of diminished clarity affects responding. The findings revealed a functional relation between the rate of diminished stimulus clarity and the rate of responding: an increase in the rate at which the stimulus diminished produced a corresponding increase in the rate of responses that prevented the stimulus from diminishing. The implications of these data and their relation to an understudied phenomenon, negative conjugate reinforcement, are addressed.


Modifed by Eddie Soh