Six subjects formed two three-member equivalence classes following conditional dis­crimination training in a matching-to-sample procedure using nonsense syllables as stimulii. For three subjects, a stimulus from each equivalence class was then given a distinct dis­criminative function. For the remaining three subjects, a stimulus from each equivalence class was given a conditioned reinforcing function. Six additional subjects served as con­trols. Three received the discriminative control training and three the conditioned reinfor­cement training, but none received the conditional discrimination training required to form the equivalence class. Following training, testing was done using an element of the equivalence classes related symmetrically and transitively to the element used in the dis­criminative or reinforcement training. For the equivalence subjects, but not the control subjects, discriminative control and conditioned reinforcement effects transferred to other members of the equivalence classes. This transfer of discriminative and reinforcing func­tions across members of equivalence classes may provide a more complete account of some types of generalization and maintenance. In addition, stimulus equivalence and related phenomena may help explain the control exerted by symbolic stimuli such as in rule governed behavior

The effects of reinforced pretraining on subsequent rule discovery were examined with college students as subjects. Levels of behavioral stereotypy observed during reinforced and non-contingent pretraining were compared. During pretraining subjects received reinforcement if they pressed two keys in a particular sequence. During the problem session pressing each key four times was a necessary condition for reinforcement, but each problem had additional different requirements for reinforcement. Subjects were asked to solve the problems by discovering the rule that determined whether or not they received reinforcement. Levels of stereotyped responding during pretraining were equivalent for contingently and non-contingently trained subjects. During the problem session contingently pretrained, non-contingently pretrained, and naive subjects required equal numbers of trials to solve problems and solved the same number of problems. The results suggest that behavioral stereotypy observed in this experimental preparation may be due to repeated exposure to the task. Differences between the results observed in this study and that of Schwartz (1982) and implications for the use of reinforcement procedures in applied settings are discussed.

These papers were presented at a symposium held at the 93rd annual convention of the American Psychological Association, August, 1985, Los Angeles, CA. The papers, in order, address the following topics: the history of operant research with human subjects with particular attention to the type and quantity of research conducted since the mid-1950s; interpretation versus experimentation in the experimental analysis of human behavior (EAHB) with particular emphasis on rule-governed behavior; the historical and current roles of children as subjects in operant research; the relation between basic laboratory research with humans and applied behavior analysis; and the need for future research concentrating on complex human behavior. Each of these papers is discussed in the final section of the proceedings.

In a behavioral view, the purposes of science are primarily prediction and control. To the extent that a scientist embraces both of these as a unified and generally applicable criterion for science, certain philosophical and theoretical practices are counterproductive, including mentalism in both its metaphysical and metatheoretical forms. It is possible and often worthwhile to recast some mentalistic talk into an issue of behavior-behavior relations. When behavior-behavior relations are approached non-mechanistically, however, analysis cannot stop at the level of the relations themselves. Several analytic concepts common in the behavioral community share some of the dangers of mentalism if not employed properly, including such concepts as self-reinforcement, response-produced stimulation, and self-rules.

Humans were presented with a task that required moving a light through a matrix. Button presses could produce light movements according to a multiple fixed-ratio 18/differential-reinforcement-of-low-rate 6-s schedule, with components alternating every 2 min. Moving the light through the maze earned points worth chances on money prizes. In Experiment 1 four conditions were assessed through between-subject comparisons: minimal instructions, instructions to press rapidly, instructions to press slowly, and instructions that sometimes rapid responding would work while at other times a slow rate would work best. Subjects responded in three successive sessions of 32 min each. The results suggested that instructions affected the nature of the contact made with the programmed consequences and thus subsequent performance. In some cases, responding seemed to result from added contingencies introduced by stating rules. In Experiment 2 the relative contribution of these two effects was assessed by presenting and then withdrawing two lights that had been paired with two specific instructions: "Go Fast" or "Go Slow." There were three conditions. In one condition, only the Go Fast light was on; in a second, only the Go Slow light was on; and in a third, the lights alternated each minute. In each condition, half the subjects had all instruction lights turned off after the first session. The results once again showed an effect of instructions on contact with the programmed consequences. However, responding sometimes continued in a manner consistent with added contingencies for rule-following even when the programmed consequences had been contacted and would have controlled a different type of responding in the absence of instructions. The relevance of added contingencies for rule-following in determining the effects of explicitly programmed consequences is emphasized.

Schedule sensitivity has usually been examined either through a multiple schedule or through changes in schedules after steady-state responding has been established. This study compared the effects of these two procedures when various instructions were given. Fifty-five college students responded in two 32-min sessions under a multiple fixed-ratio 18/differential-reinforcement-of-low-rate 6-s schedule, followed by one session of extinction. Some subjects received no instructions regarding the appropriate rates of responding, whereas others received instructions to respond slowly, rapidly, or both. Relative to the schedule in operation, the instructions were minimal, partially inaccurate, or accurate. When there was little schedule sensitivity in the multiple schedule, there was little in extinction. When apparently schedule-sensitive responding occurred in the multiple schedule, however, sensitivity in extinction occurred only if differential responding in the multiple schedule could not be due to rules supplied by the experimenter. This evidence shows that rule-governed behavior that occurs in the form of schedule-sensitive behavior may not in fact become schedule-sensitive even though it makes contact with the scheduled reinforcers.

Acquisition and maintenance of autoshaped keypecking with pigeons were studied as a function of immediacy of reinforcement. The basic design manipulated the immediacy of reinforcement following the trial onset while keeping the rate of reinforcement within the trial constant and manipulated the rate of reinforcement within the trial while keeping the immediacy of reinforcement constant. Seven groups of birds were studied at trial durations of either 24 or 60 seconds and immediacy intervals of either 3, 12, 24, or 60 seconds. All groups were exposed to a 180-s inter-food cycle. Acquisition rates varied as a function of the ratio of cycle to trial duration. Higher ratio groups (24 s) resulted in faster acquisition than did lower ratio (60 s). Immediacy of reinforcement exerted a less powerful effect upon acquisition than did the ratio of cycle to trial duration. During maintenance training, probability and rate of responding were positively related to immediacy of reinforcement and unaffected by trial duration. A modification of a scalar expectancy model was shown to account for these results.

The experimental analysis of behavior is also generally characterized by an unhurried attitude toward the as-yet unexplained. Criticism often takes the line that the analysis is oversimplified, that it ignores important facts, that a few obvious exceptions demonstrate that its formulations cannot possibly be adequate, and so on. An understandable reaction might be to stretch the available facts and principles in order to cover more ground, but the general plan of the research suggests another strategy. Unlike hypotheses, theories, and models, together with the statistical manipulations of data which support them, a smooth curve showing a change in probability of a response as a function of a controlled variable is a fact in the bag, and there is no need to worry about it as one goes in search of others. The shortcomings and exceptions will be accounted for in time. (Skinner, 1969, p. 84)

Eating was measured in water-deprived rats when water was presented at regular intervals ranging from 30 to 240 sec. The temporal patterning of eating resembled that of schedule-induced behavior in that the probability of eating was high early in the interval and declined in the end of the interval. Additionally, (a) the number of pellets consumed was controlled by relative time in the interwater interval, (b) the pellets consumed per water presentation was inversely related to water rate, and (c) food-ingestion rate was directly related to water rate. These relationships parallel those found with behavior regarded as schedule induced.

In a series of three experiments the effects of variation in grain duration on automaintenance were evaluated. In the first experiment, key illumination was followed by grain only when pigeons did not peck the key. Each subject was exposed to 2-, 4-, and 8-second feeder durations in blocks of 10 sessions. Subjects pecked on a high percentage of trials at all feeder durations. The mean peck latency was shorter in the 8-second condition than in the two other conditions in five of six subjects. The conditional probability of pecking given successive keylight-grain pairings did not increase as the number of pairings increased. The second experiment was identical to the first, except that key pecking had no scheduled consequence. Under these conditions, all three subjects showed substantial responding. The recorded measures showed no systematic relationship to feeder duration in this study. In the third experiment, two different stimuli were followed by feeder presentations of either identical (2- or 8-second) or different (2- and 8-second) durations within each session. Subjects tended to respond sooner and with a higher overall rate in the presence of the stimulus associated with the longer feeder duration only when different feeder durations were presented within the same session. This result was confirmed by direct observation of the pigeons. The results of these experiments suggest that the effects of varying grain duration may be small, compared to the effects of varying other variables. The results also suggest that the location as well as the frequency of pecking may be an important measure in the analysis of factors controlling the pigeon's key peck.

Running-wheel behavior was examined as a function of the floor area of the experimental chamber in three food-deprived rats when a food pellet was delivered each minute and when it was omitted. Running-wheel behavior when food was omitted was unsystematically related to the floor area. When food was scheduled, three measures of running-wheel behavior were found to be decreasing functions of floor area: percentage of the session time spent in the wheel, percentage of the interfood intervals with a wheel entry, and the mean stay time per interfood interval with a wheel entry. Wheel revolutions per session varied unsystematically, and the local rate of running was an increasing function of floor area when food was scheduled. These results explicate some inconsistent findings in the literature, and provide support for the notion that wheel-running is not a schedule-induced behavior.

Pigeons were trained on concurrent schedules in which key pecking was required by both schedules (concurrent variable‐interval variable‐interval schedules) and on concurrent schedules in which key pecking was required by only one of the schedules (concurrent variable‐interval variable‐time schedules). The distribution of reinforcements was systematically varied with both types of concurrent schedules. The distribution of time between the schedules depended on the reinforcement distribution and was independent of the symmetry of the response requirement. The relation between time and reinforcement distributions appears to be invariant over a wide range of manipulations of responding maintained by concurrent schedules.

In a two-key chamber, one key (the food key) was either red or green with different variable-interval schedules operating concurrently in each color and a second key (the changeover key) served to change the food-key color. Three pigeons were trained with either a 2-sec changeover delay or a 0-sec changeover delay and three birds with a fixed-ratio 2 on the changeover key instead of a changeover delay. The proportion of time spent in red approximated the proportion of reinforcers delivered in red for all birds. When the procedure was changed so that reinforcers were signalled in the green schedule, rates of reinforcement were unaltered, but the pigeons spent virtually the whole session in red. Changeovers to green were allowed only when a reinforcer was assigned by the schedule associated with green. For all pigeons with the fixed-ratio requirement on the changeover key or with a 0-sec changeover delay, the overall rate of red-key responses was higher during the signalling condition than during unsignalled, or baseline, condition. The present data question the generality of previous reports that the rate of one response is independent of the amount of time allocated to the alternative response.