Hiroto Okouchi

Abstract Three pigeon dyads were exposed to a two‐component multiple schedule comprised of two tandem variable‐interval 30‐s interresponse time (IRT) > 3‐s schedules in the presence of different stimuli. Pecks to keys by both pigeons of a dyad occurring within 500 ms of one another were required for reinforcement under one tandem schedule (the coordination component), and such coordinated responses were not required under the other (the control component). The terminal link of each schedule ensured that the reinforced coordination response was an IRT > 3 s. Rates of coordinated IRTs > 3 s and total rates of coordinated responses (composed of IRTs > 3 s and IRTs ≤ 3 s) were higher in the coordination components than in either of two different control components in which coordination was not required for reinforcement. This difference in coordinated responses in the presence and absence of the coordination requirement under stimulus control transitorily deteriorated and then was reestablished when the relation between the stimulus and the coordination contingency or its absence was reversed. The results show coordinated responding to function as a discriminated social operant.

Abstract Although hypothetical rewards have been used almost exclusively in human discounting studies, investigations of their validity are limited. The present experiment compared the discounting of monetary reward value by probability across conditions in which the rewards were real, potentially real, and hypothetical. Twenty‐four undergraduates choose between an uncertain large reward and a certain small reward 60 times (trials). In the real and hypothetical reward conditions, the participants made choices with real and hypothetical money, respectively, in every trial. In the potentially real condition, they did so with real money in randomly selected three of the 60 trials and with hypothetical money in the remainder. The log₁₀‐transformed h values of a hyperbolic probability‐discount function and the values of the area under the curve with an ordinal transformation of odds against were higher and lower, respectively, in the potentially real and in the hypothetical reward conditions than in the real reward condition, demonstrating that the probability discounting of hypothetical monetary rewards was larger than that of real rewards. These results suggest that future studies are required to identify why the hypothetical reward procedure overestimates the discounting rates of real rewards.

The experiment examined effects of instructions on human schedule performance when providers of the instructions were persons other than the experimenter. Pressing a key by one member of each pair of eight undergraduates produced points on a multiple fixed-interval (FI) FI schedule. Another member of the pair was exposed to a multiple fixed-ratio (FR) differential-reinforcement-of-low-rate (DRL) schedule. After every session, verbal descriptions of how to increase points written by another member of the pair were presented. The final response rates under the multiple FI FI schedule were indistinguishable between the two components in each of four participants although they each received verbal descriptions indicating that the way to increase points with the key was to respond rapidly in one component and to respond slowly in another. Such weak effects of instructions given from persons other than the experimenter were contrasted with stronger effects of instructions given by the experimenter, that is, the final response rates were higher with the instruction to respond rapidly than with the instruction to respond slowly for three of the four participants. This differential effect across the instruction providers was not found in each of four participants who were exposed to the multiple FR DRL schedule.

The present experiment demonstrated a procedure analyzing the behavior of controllers by measuring undergraduates' responses that produced consequences to other undergraduates. One member of each pair of 42 undergraduates, referred to as a learner, was asked to earn points exchangeable for money and told that touching a square shown on the screen of the display monitor may or may not change points. Unlike standard operant experiments, the point change was not determined by any computer program but by responses of another experimentally na < ve undergraduate, referred to as an instructor. Response rates for 15 of 21 learners were higher than those for their yoked partners, who got and lost points independent of their responding. For 17 of 21 instructors, responses to an alternative (S+ key) that produced points to the learners were more than responses to another alternative (S- key) that resulted in the learners' point loss. Each of 16 instructors pressed the S+ key more frequently after the learner responded than after the instructor pressed the S+ or S- key. The mean interval between the learner's response and the instructor's S+ response was shorter than 1 s for each of 14 pairs of instructors and learners. Consistent with the behavioral principle of positive reinforcement, these results suggest that the instructors generally provided positive reinforcers depending on and immediately following the learners' responses.

The present study examined whether bidirectional response-response relations could be established without direct reinforcement. In AB training for 12 undergraduates, higher rates of touches to a white circle on the monitor screen (A1) produced two stimuli (B1 and B2) on half of the trials, whereas lower rates (A2) produced the same effect on the other half. Choosing one of the two stimuli was reinforced according to the preceding responding (A1B1 and A2B2). In BC training, touching a stimulus (B1 or B2) produced three white circles lined up horizontally on the screen, after which one of two different response sequences to the circles (left-center-right, C1 or C2, and right-center-left, C2 or C1) were reinforced, depending on the stimulus presented (B1C1 and B2C2). After AB and BC relations were established, 11 of 12 participants showed the emergence of untrained relations (BA, CB, AC, and CA) throughout five test sessions, and the remaining participant showed it in the first four test sessions. These response-response relations were replicated with five other undergraduates and different trained relations.

Two experiments were conducted to assess stimulus control and generalization of remote behavioral history effects with humans. Undergraduates first responded frequently under a fixed-ratio (FR) schedule in the presence of one line length (16 mm or 31 mm) and infrequently on a tandem FR 1 differential-reinforcement-of-low-rate (DRL) schedule when a second line length (31 mm or 16 mm) was present. Next, an FR 1 schedule in effect in the presence of either stimulus produced comparable response rates between the stimuli. Finally, a tandem FR 1 fixed-interval (FI) schedule was in effect under those same stimuli (Experiment 1) or under 12 line lengths ranging from 7 to 40 mm (Experiment 2). In both experiments, responses under the tandem FR 1 FI schedule were frequent in the presence of stimuli previously correlated with the FR schedule and infrequent in the presence of stimuli previously correlated with the tandem FR 1 DRL schedule. Short-lived but systematic generalization gradients were obtained in Experiment 2. These results show that previously established rates of behavior that disappear when the establishing contingencies are changed can subsequently not only reappear when the contingencies change, but are controlled by and generalize across antecedent stimuli.

Undergraduates were exposed to a series of reinforcement schedules: first, to a fixed-ratio (FR) schedule in the presence of one stimulus and to a differential-reinforcement-of-low-rate (DILL) schedule in the presence of another (multiple FR DRL training), then to a fixed-interval (FI) schedule in the presence of a third stimulus (FI baseline), next to the FI schedule under the stimuli previously correlated with the FR and DRL schedules (multiple FI FI testing), and, finally, to a single session of the multiple FR DRL schedule again (multiple FR DRL testing). Response rates during the multiple FI FI schedule were higher under the former FR stimulus than under the former DRL stimulus. This effect of remote histories was prolonged when either the number of FI-baseline sessions was small or zero, or the time interval between the multiple FR DRL training and the multiple FI FI testing was short. Response rates under these two stimuli converged with continued exposure to the multiple FI FI schedule in most cases, but quickly differentiated when the schedule returned to the multiple FR DRL.

The present experiment examined whether a response class was acquired by humans with delayed reinforcement. Eight white circles were presented oil a computer touch screen. If the undergraduates touched two of the eight circles in a specified sequence (i.e., touching first the upper-left circle. then the bottom-left circle), then the touches initiated an unsignaled resetting delay culminating in point delivery. Participants experienced one of three different delays (0 s, 10 s, or 30 s). Rates of the target two-response sequence were higher with delayed reinforcement than with no reinforcement. Terminal rates of the target sequence decreased and postreinforcement pauses increased as a Function of delay duration. Other undergraduates exposed to yoked schedules of response-independent point deliveries failed to acquire the sequence. The results demonstrate that a response class was acquired with delayed reinforcement, extending the generality of this phenomenon found with nonhumans animals to humans.

Study objective: Effects of naming training and comparison-as-node training on the establishment of stimulus equivalence were examined. Design: Group comparison and within-subject comparison designs were used. Participants: 13 undergraduates participated. Intervention: In Phase 1, all participants experienced matching-to-sample training under the linear-series training structure, followed by a test examining whether three 5-member equivalent classes had been established. In Phase 2, the participants experienced the same matching-to-sample training and equivalent class test as Phase 1, but with different stimulus sets. Prior to this, participants were given naming training, comparison-as-node training, or no additional training. Measures: Percentage of correct responses and response latency. Results: No participant established stimulus equivalence in Phase 1. In Phase 2, participants who had received naming training or comparison-as-node training established stimulus equivalence, whereas participants without such additional training did not. Conclusion: Prior training on comparison-as-node, as well as on naming, contributed to the establishment of stimulus equivalence.

One group of undergraduates responded under a fixed-ratio (FR) 25 schedule and a second group responded under a differential-reinforcement-of-low-rate (DRL) 5-s schedule (first history phase). Both groups of subjects were then exposed to a differential-reinforcement-of-other-behavior (DRO) 5-s schedule (second history phase), and finally to fixed-interval (Fl) 5-s (Experiment 1), variable-interval (VI) 5-s (Experiment 2), extinction, FR, and DRL (Experiment 3) schedules (history testing phase). Response rates under the extinction and the FR schedule in the history-testing phase were higher for subjects with a history of the FR schedule than for subjects with a DRL history. Such a systematic difference by remote histories was not observed when the FI, VI, and DRL schedules were in effect during testing.

Study objective: The present study examined effects of a remote history of a DRL schedule on interresponse times (IRTs) under a VI schedule with human subjects. Design: Double-history, single-history, and no history conditions were compared across subjects. Setting: The experiment was conducted individually in a room. Subjects: Two male and eleven female undergraduates with no experience of reinforcement experiments, 19 to 22 years old, participated. Procedure: In the doublehistory condition, DRL 1-s with LH 1-s, DRL 5-s with LH 5-s, and VI 5-s schedules were in effect in that order. The single-history subjects were exposed only to the last two schedules, whereas the no-history subjects experienced only the VI 5-s schedule. Results: For all subjects in the double-history condition, IRTs longer than 1s and equal to or shorter than 2s, which had been reinforced differentially under the DRL 1-s with LH 1-s, emitted more frequently during the last five VI 5-s sessions than during the first five. Such a systematic difference was not observed in the single-history and no-history conditions. Conclusion: These results suggest that a schedule history affected remotely on the VI responding, and that the IRT was a useful measure for examining the remote history effects.

The present article discusses how events outside a subject's skin and not accessible to another subject but to an experimenter may contribute to experimental analyses of private events. Of 16 undergraduates, 8, referred to as instructors, first learned conditional discriminations (i.e., B1C1, B2C2, B3C3, and B4C4) in a standard matching-to-sample (MTS) task with the stimuli Bs as the samples and the stimuli Cs as the correct comparisons. Then the other 8 subjects, learners, were exposed to modified MTS trials in which responses of the learners were reinforced or punished not by the experimenter but by the instructors. Conditional discriminations to be established were A1C1, A2C2, A3C3, and A4C4, in which the sample stimuli As were presented simultaneously with the stimuli Bs so that the instructors could not see the As but only the Bs. For 2 of the 8 pairs, the learners learned the AC conditional discriminations from the instructors who were not accessible to the stimuli As. Functionally, private events have been defined by their accessibility rather than their structure (e.g., Skinner, 1953). In this context, therefore, the stimuli As may be characterized as a kind of private events within the present 2 pairs of instructors and learners.