Association for Behavior Analysis International

The process of resolving ambiguity is both pervasive and central for everyday cognition. In naturalistic settings, people are constantly confronted with words that have different meanings,facial expressions that are equivocal,and entire social situations that can lead to various interpretations. However, emotion in faces and contextual information were investigated separately in the previous studies. Moreover,few research has investigated precisely this issue analyzing of eye movements. Thus, we combined a task performance (verbal response to ambiguous facial expressions which consists of confliction and contextual stimuli) and eye movement information (where the participant fixate when completing the task) to investigate the difference between ambiguous facial expressions of emotion and contextual cues. 18 participants viewed color image sequences of 39 different character displaying facial expressions of happy, sad, angry, disgusted, surprised, and fearful faces (see table1). After seeing each image, the participants were asked to answer emotional category for the character. The result showed that the participants fixation and verbal responses matched correctly only at 25.4 %( see Figure1). In addition to this, when ambiguous stimuli are showed to the participants, their fixation and verbal responses matched correctly only at 23.8 %( see Figure1) .

Power relations may be conceived as contingencies involving four different functions: prescription, regulation, administration and monitoring of interactions (Ribes, 2001). In order to evaluate the role of differential exposition to an authority in obedient behavior, sixteen schoolchildren were assigned to four groups. Each group was exposed to 1, 2, 3, or 4 sessions to an authority condition, in which the experimenter exercised all the power functions mentioned above as participants solved puzzles. After that, in a word search task, all groups were exposed to each of the power functions in the following sequence: prescription, regulation, supervision, and administration, under the assumption that this is the sequence in which the power functions are established in real power interactions. Results showed that participants in Group 1 (one day of authority exposition) emitted more disobedient responses than the rest of the groups, increasing this kind of responses in regulation and prescription phases. These results are discussed in terms of a comparison between them and the results found by Rangel, Vazquez, Pulido & Reyes (unpublished) in which participants were exposed during different number of days to a familiarization condition, in which the experimenter did not exercise any power function in the situation.

Impulsive choice has been found to predict greater alcohol consumption in rats. However, the extent to which alcohol consumption may be modified by experimental manipulation of impulsive choice remains unclear. In the present study, we are examining the effects of an early, prolonged history of exposure to reinforcement delay on subsequent impulsive choice and alcohol consumption. Three groups of Long-Evans rats (21 days old at start of experiment) have been exposed to 100 training sessions (out of 120 total) in which they respond for food reinforcement on a single lever delivered after either no delay (n = 14), a fixed delay (n = 14), or a progressively increasing delay (n = 16). Reinforcement delay for rats in the "progressive-delay" group has been gradually increased across training sessions contingent on criterional performance (i.e., short response latencies and few trial omissions). Following training, between-group differences in impulsive choice and alcohol consumption will be assessed. Additional data to be collected. Anticipated month of study completion is December 2011.

Studies using question-based discounting tasks have produced weak to moderate correlations between discounting of delayed and uncertain rewards in support of single process explanations of discounting (Myerson, Green, Hanson, Holt & Estle, 2003). However, differential magnitude effects on delay and probability discounting are problematic for these theories because they predict manipulations to impact delay and probability discounting similarly. This study examined correlations between discounting of delayed and probabilistic outcomes using an experiential task in which repeated discounting choices were required as part of a computer based skiing game. Results indicated that while both probability and delay discounting versions of the task produced discounting well described by hyperbolic functions, discounting rates on the two tasks were uncorrelated.

Divorce rate within United States have been steadily increasing in the last 20 years. A major factor within divorce is due to financial constraints between partners. The current study examined if temporal discounting patterns change when married partners individually discount hypothetical money to that of when the partners discount together. Varaibles of age, years married, education, and income level are incorporated into the analysis. Initital findings show large differences in discounting patterns when the partners discount individually to that of when paired together. More so, descrepancies within paried discounting are noted. The utility of delay discounting as an assessement tool will be discussed in potentially aiding this increasing problem.

Research has found that the rate at which probabilistic reinforcers are discounted varies as a function of the magnitude of those reinforcers. However, the value of a given reinforcer may also be a function of the availability of other reinforcers. This study manipulated relative reinforcer value by varying background income. Participants in this study were first year psychology students who played an experiential discounting task in which they earned points by making jumps on a ski run. The points available for these jumps were used as an analogue of background income, which was manipulated by changing the number of points available. Discount rates were obtained by providing participants with occasional opportunities to make free run jumps in which they made a choice between two outcomes that varied in size. The probability associated with the larger of these two outcomes varied but its magnitude remained the same. The magnitude of the smaller outcome titrated across trials but its probability remained the same. Data were well described by hyperbolic functions but there were no significant differences in discount rates. Results will be discussed in terms of magnitude effects and energy budget.

Humans and non-humans are often said to lack self-awareness (Goldiamond, 1959, 1962, 1965, 1966; Nisbett & Wilson, 1977). Although there are clearly many ways to interpret such terms as self-awareness, they all seem to involve cases in which individuals own prior behavior is discriminative for their subsequent behavior. In the present research pigeons were trained in a compound, discrete-trial procedure. The first component of a trial was a conventional matching to sample (MTS) component. The second component was a probe in which the pigeons were presented with two further stimuli, responses to one of which were reinforced conditional on correct performance in the prior matching component. Four pigeons responded with an accuracy above 90%, and near 100% on both the MTS and probe components, indicating the pigeons own prior behavior had indeed become discriminative for their subsequent behavior. The procedure appears promising as a means for assisting individuals with developmental disabilities to better come under the discriminative control of the outcomes of their own prior behavior.

The effects of spatial food distribution were evaluated on the producer-scrounger strategies in groups of rats. The experimental design included four groups: Matrix Separate, a group with four members was exposed to four food container arranged in a 60 x 90 cm matrix; Matrix Near, a group with four members was exposed to four containers arranged in a 30 x 30 cm matrix; Irregular Separated, a group with four members was exposed to four food container arranged in a irregular distribution with a minimal separation between containers of 60 cm; Irregular Near, a group with four members was exposed to irregular distribution as a previous group, but a minimal separation between containers of 60 cm. The groups exposed to matrix distribution used producing as predominant strategy to obtain food; the groups exposed to near containers used producing as predominant strategy.

Given the ubiquitous role of time in all schedules of reinforcement, ratio schedules can be equated to interval schedules with linear feedback loops (McDowell & Wixted, 1986). To examine the role of feedback in preference between variable schedules, 4 pigeons were trained on a discrete-trial concurrent-chains arrangement with variable schedules serving as the terminal link alternatives. Baseline consisted of performance on a VR30 VR30 concurrent-chains arrangement with a brief initial link instead of an inter-trial delay (Gibbon et al., 1988). Upon reaching a stable pattern of indifference (50%) in the preference ratios, the birds were transferred to a concurrent-chains schedule with variable ratio and variable interval plus linear feedback terminal links. The linear feedback function was positively sloped to reinforce shorter inter-response times as opposed to standard variable interval schedules which reinforce longer inter-response times. After a return-to-baseline condition, the birds were exposed to a concurrent-chains schedule with VR30 and VI 30 terminal links. Preference was determined by the percentage of choices for variable interval with feedback per 40-trial session. This study expands upon the literature regarding linear feedback and seeks to investigate its role in preference between fully self-regulated (ratio) schedules and variable interval schedules with feedback.

A stimulus comes to be a reinforcer for an arbitrary operant response through imprinting. The present study compared chicks' choice behaviors reinforced by an imprinted stimulus with those reinforced by food under concurrent schedules of reinforcement. Newly hatched chicks were exposed to a moving red cylinder. After that, their preferences for the stimulus were tested in the situation where both the stimulus and a novel stimulus were simultaneously exposed to each chick. Only chicksthat chose the familiar stimulus were used as the subjects for the following experiments. They were divided into the imprinted stimulus (n = 11) or the food (n = 12) groups. According to the group, the chicks' key-peck responses were shaped by using the imprinted stimulus or food as reinforcer, respectively. After that, chicks' responses were under a concurrent variable-interval variable-interval schedule. Five ratios of independently scheduled reinforcement for each reinforcer were utilized: 1:0, 7:3, 1:1, 3:7, and 0:1. Four chicks of the imprinted stimulus group showed matching and further6 chicks did undermatching. However, only3 chicks of the food group showed undermatching, and other chicks of the group did not any matching. These results may reflect some differences of both reinforcers.

Several studies have shown that behavior in a choice situation may be more sensitive to changes in reinforcement rate than to changes in reinforcer magnitude. One variable that may be responsible for differential control of behavior by reinforcement rate and reinforcer magnitude is the onset of the reinforcer. The purpose of the present experiment was to compare multiple and single-onset reinforcement in a choice procedure where the total duration of access to the reinforcer was held constant between the 2 alternatives. Pigeons responded in a concurrent-chain procedure where 10 free-choice trials and 10 forced choice trials were presented within a session. The terminal links lead to 1 or several onsets of the reinforcer. When fixed ratio (FR) 1 schedules were used in the initial and terminal links, choice between single and multiple-onset reinforcement varied between subjects. When an FR 40 terminal link was introduced, choice of the multiple-onset reinforcement became nearly exclusive. This effect was replicated when the keys leading to the multiple and single-onset reinforcement were reversed and when using a differential-reinforcement-of-other-behavior schedule in the terminal link.

Lever pressing was maintained by a concurrent token production schedule in rats. Token deliveries (i.e., steel ball bearings) were assigned probabilistically to either the right or left lever such that the ratio of left to right token deliveries was either 1:6 or 6:1, depending upon condition. The location of the rich lever remained constant within session, but varied across daily sessions according to a pseudorandom binary sequence. Once assigned to a lever, token delivery was arranged by a random interval 15 s schedule. Transition to token exchange was arranged by either a fixed ratio (FR 2 or 4) or random ratio (RR 2 or 3) schedule, depending upon condition. During token exchange, depositing each token in a receptacle produced access to sweetened condensed milk. Session-wide lever press ratios tracked daily changes in the programmed token ratio and were adequately described by the generalized matching law. Considerable undermatching was observed, however. Systematic effects of manipulating the exchange production schedule on rates of lever pressing and sensitivity to the token reinforcer ratio were found in only one of four subjects. Sign tracking elicited by the tokens was also observed.

Pigeons were trained on a concurrent-chains procedure. Terminal links were always fixed-interval (FI) 10-s and FI 20-s during the rapid acquisition condition. Across sessions, the location of the shorter terminal link varied according to a 31-step pseudorandom binary sequence. In a subsequent, 11-session "suddenly equivalent" condition, terminal links were both FI 15-s. To obtain measures of temporal control, occasional "no food" terminal links lasted 60-s and pecks had no effect. In the rapid acquisition condition, initial-link response allocation favored the shorter terminal link and stop times from no food terminal links indicated temporal discrimination. Initial- and terminal-link pecking stabilized within the first half of sessions. Initial-link response allocation never reached indifference in the 11 suddenly equivalent sessions. However, stop times rapidly adjusted to the intermediate FI value. Residual covariation analyses of log initial-link response and log expected immediacy ratios indicated that the mechanisms determining initial- and terminal-link responding were interdependent in the rapid acquisition condition. Residual covariation analyses from the suddenly equivalent condition indicated inter-subject variability. Taken together, these results may indicate that deviations in subjective terminal-link immediacy play a role in choice and timing behavior.

This study investigated the mechanism of change in Functional Analytic Psychotherapy (FAP), whose efficacy has been supported by single-case investigations. The lack of experimental tests instigated the present research and for that reason, this study proposed a single-case experimental design with withdrawal phases (A-B1-C1-B2-C2; A= therapy without planned interventions; B=introduction of FAP; C= withdrawal of FAP). The aim of the study was to evaluate the effects of FAP interventions on the following clinically relevant behaviors: 1) verbose talking, 2) talking with no correspondence with the therapist speech, 3) superficial speech and 4) aggressive verbal responses. The sessions of two clients were recorded (20 sessions each) and coded using the Functional Analytic Psychotherapy Rating Scale. The results showed that the frequency of CRB1 decreased in the phases where FAP was introduced (B1 and B2) and the withdrawal occurred in the phases where FAP was removed (C1 and C2). The CRB2 and CRB3 showed the opposite tendency, i.e. frequency increased in phases B and decreased in phases C. This design proved to be useful for research in psychotherapy and indicated the therapist contingent responding as the most important mechanism of change.

Different research designs can contribute to evidence-based practice, and single case experimental designs are useful for establishing causal relationships in the context of an individual, a typical context in psychotherapy. Single subject experimental designs have been largely used in basic and applied behavior analytic research, but less so in psychotherapy research. The extensive knowledge of those designs is an asset for behavior analysts psychotherapy researchers interested in understanding the process of change and in determining causal relations. To conduct single case quantitative studies, it is necessary to measure repeatedly over time, allowing estimates of variability in the behavior of interest, its level of occurrence and apparent trends. In our research group we were able to develop independent and dependent variable measures that are significant and interfere little in the therapeutic relationship. Different category systems have been created, some individualized, some standardized and some standardized with built-in individualization. Data analyses included measures of frequency, duration and sequential analysis. Initially the design of those studies was descriptive, but recent projects are using single case reversal designs. Samples of results of different repeated measures systems and forms of analysis will be presented, as well as some results obtained with a reversal design.

Computer based digital pictures of items may be an effective format for presenting complex stimuli in preference assessment. In this study we evaluated the correspondence between preference hierarchies generated from paired choice preference assessment method with both a computerized tool and with concrete items. The participants were eight boys and six girls of typical development, ranging from 3 to 5 years old, attending a public preschool participated in Study 1. Nine boys with intellectual disabilities, ranging from nine to thirteen years old, attending a school for children with disabilities and five female special education teachers (T1, T2, T3, T4 and T5) participated in Study 2. In Study 1, the top-ranked List 1 stimulus corresponded for six of the 14 participants, and the top-ranked List 2 stimulus corresponded for ten of the 14 participants. In Study 2, the top-ranked List 1 stimulus corresponded for five of the nine participants, and the top-ranked List 2 stimulus corresponded for seven of the nine participants. The implications of these results for future research on computer based digital pictures of items format for preference assessments is discussed.

Motivating operations are antecedent variables that alter the value of consequent events as well as some dimension of the corresponding response class. The interactions between2 motivating operation conditions and3 reinforcer classes were evaluated using a superordinate multielement design. Two individuals with developmental disabilities participated. Participants were exposed to a motivating operation condition followed by a reinforcer assessment, in which the reinforcer response requirements progressively increased following contact with the terminal reinforcer. The motivating operation conditions were pre-session access until rejection and pre-session restriction for 24 hours. Reinforcers were stimuli representative of primary, conditioned, and token reinforcers. The results showed that the manipulation of motivating operations produced reliable changes in the effectiveness of each reinforcer. For one of the participants, the effectiveness of the motivating operation appeared to be influenced by the reinforcer class.

Basic behavioral literature suggest that an organisms preference for reinforcement schedules that require response repetition or variability is inversely related to the response requirements. For example, Rodrigues et al. (1995) found that preference for a Lag contingency decreased as the degree of variation required for reinforcement increased. Children with autism typically engage in repetitive motor and vocal behaviors. It is unknown, however, whether these individuals prefer repetitive behavioral patterns, or whether they engage in these patterns due to low reinforcement of other behaviors or because varied behaviors do not exist in their repertoire and thus cannot be selected through reinforcement. In the current study, repetitive and variable behavior was first brought under stimulus control by providing reinforcement for variable responses on a Lag 3/FR 3 schedule in the presence of one stimulus and providing reinforcement for repeated responses in the presence of a second stimulus. Once discriminated responding was observed in the multiple schedule, participants preference for varied and repeated schedules were assessed in a concurrent chain procedure. Results indicated that participants showed a preference for the repeated component of the multiple schedule.

The current research aims to determine whether preschool children understand the task instructions and can demonstrate their opinions on how a resource should be distributed in a forced choice paradigm, as in the Sharing Game (Kennelly & Fantino, 2007). In a within-subjects design,2 experiments were conducted involving repeated-trials over10 opportunities in which 17 male and female preschoolers made choices to distribute resources between themselves and a passive other. The task involved simplified verbal instructions and was visually presented in a table-top format with cardboard, metal coins, and2 stuffed animals,1 in each side of the cardboard. The results showed that for the boys, there was a higher percentage of competitive choices while, for the girls, the results showed a higher percentage of optimal choices. When confronted with the option of being egalitarian or altruistic, the results did not show gender difference. Both boys and girls tended to be more egalitarian than altruistic. Overall, the children's results indicated that the experimental set up was adequate to yield choice performance, but the extent to which each component of the instructional package (verbal instructions and presentation format) can be a relevant contextual variable is still under scrutiny.

Interval methods of data collection are commonly used in the field of applied behavior analysis even though discontinuous measures, by definition, only sample the occurrences of a target behavior, and therefore, may be inaccurate. Several studies have evaluated false positives and false negatives in regards to interval methods of data collection using simulated data. However, none of these studies has evaluated false positives and negatives using nonsimulated behaviors. In addition, prior studies have not evaluated whether intervals methods of data collection obscure or produce trending that is evident in continuous records. The purpose of the present study was to evaluate the sensitivity of partial interval recording (PIR) and momentary time sampling (MTS) for detecting changes in duration and frequency events with data recorded from soccer games. Three separate experiments were conducted to evaluate continuous event data for (a) false positives produced by interval methods, (b) false negatives produced by interval methods, and (c) trends that may become evident when data are converted to different interval sizes of PIR and MTS. The results from the study indicate that simulated data provide a good model for predicting the sensitivity of interval methods for detecting changes in nonsimulated behavior.