W. Kirk Richardson

The NASA/LRC Computerized Test System

A new testing package, including apparatus and tasks, is described for the study of behavior of a variety of species in a variety of experiments. The package is described with respect to the kinds of comparative psychological investigations for which it is well suited. The preliminary data generated within this new testing paradigm demonstrate that the NASA/LRC Computerized Test System provides a flexible yet powerful environment for the investigation of behavioral and psychological processes.

A chimpanzee’s (Pan troglodytes) long-term retention of lexigrams

In this experiment, a chimpanzee’s (Pan troglodytes) long-term retention was examined. The chimpanzee, Lana, was trained to use lexigrams (geometric symbols representing linguistic units) for foods, colors, and objects when she was 2 years of age. At the age of 27, her recognition of three sets of lexigrams was examined. One of these sets (long term lexigrams) included five object lexigrams, one color lexigram, and one food lexigram, none of which had been seen by Lana for more than 20 years. The second set (different-referent lexigrams) contained lexigrams that are still present on current lexigram keyboards, but had been assigned new referents. The third set (same-referent lexigrams) contained lexigrams that had been kept on the keyboards Lana had used and that had retained the same referents. A food, a colored square, or an object was presented, and Lana had to select, by using a joystick, a lexigram on a computer screen. Lana chose the correct lexigram at a level significantly greater than chance for five of the seven lexigrams that she had not seen for more than 20 years.

Extinction responding by septal and normal rats following acquisition under four schedules of reinforcement

Abstract Extinction responding of animals with septal lesions was compared to that of normal animals following acquisition training under either DRL, FI, VI or FR reinforcement schedules (Part 1). During both acquisition and extinction, septals bar pressed more than normals. However, regardless of response levels, a given schedule changed behavior in the same direction for both septals and normals. Thus, if a given schedule led to a decrease in bar pressing by normals, it also led to a decrease by septals, although the absolute level of responding at which this occurred might be different for the two groups. Similarly, if a schedule led to an increase in responding by normals, it also led to an increase by septals. In Part 2 of the experiment operant levels of septals were not found to be different from those of normals, and noncontingent food delivery generated no more nor less bar pressing in septals than in normals. It was proposed that the septal area may be of importance in the initiation of behavior which competes with the required or reinforced response.

The effect of cue-fading on the DRL performance of septal and normal rats

The effect of a cue-light fading procedure on the subsequent noncued performance under a differential-reinforcement-of-low-rate schedule (DRL) of food reinforcement was determined for rats with septal lesions and normal rats. Following initial DRL-20 sec training, all animals were trained with a cued DRL schedule in which a light signaled the end of the required delay and the availability of reinforcement. Subsequently, for one-half of the animals, the cue light was abruptly discontinued; for the remainder of the subjects, the cue was gradually faded out over succeeding sessions. Septal animals, which had the cue gradually faded, were more efficient by the end of training than septal animals which had the light abruptly discontinued. Moreover, the former did not differ from normal controls. For septal animals in the Fade group, the relative frequency distributions of interresponse times (IRTs) were bimodal, a characteristic of normal animals; septal animals which had the cue abruptly terminated, displayed the unimodal distribution of IRTs which is a characteristic of septal animals on the DRL schedule. In summary, the cue-fading procedure completely eliminated the major differences in behavior between normal and septal rats typically observed on a DRL schedule of food reinforcement.

Stimulus stringing by pigeons: Effect of feedback for correct selections

Previous studies have shown that pigeons could learn a serial list of colors and to select each color in the proper order when presented with an array of colors. Errors of jumping forward in the required sequence (forward errors) were most probable, with jumps backward in the required sequence (backward errors) being relatively improbable. One proposed explanation for the higher probability of forward errors suggests that the pigeon pecks at the correct stimulus without activating the response switch and then pecks the next stimulus in the sequence closing the response switch. This “inadequate-peck” hypothesis suggests that forward errors are due to weak stimulus control and that increased feedback for correct pecks should therefore reduce forward error probability relative to a low-feedback condition. The present study compared responding under conditions of high and no experimenter-provided feedback for correct pecks. As the feedback did not affect the probability of forward errors, the inadequate-peck hypothesis was not supported. A short-term memory explanation is consistent with the data.

Stimulus stringing by pigeons: Conditional strings

Pigeons were trained to produce one serial list in the presence of a green background cue and another serial list in the presence of a red background cue when the items for both serial lists were presented on each trial. This demonstrated a combination of serial learning and conditional discrimination learning not previously shown in pigeons. Specifically, when presented with four geometric forms, A B C D, in random locations of a five-key display, the pigeons learned to peck A B C when the background was green and A B D when the background was red. Accuracy on the conditional string ranged from 73% to 85%. Transfer tests using different locations of the stimuli on the keys showed positive transfer, thus ruling out learning of specific locations as the basis of the accurate performance. Above-chance performance was maintained when the conditional colors were presented only on the key that did not contain one of the serial stimuli. The results are interpreted in terms of a chaining model that postulates that the sequential selections were controlled by cues produced by both onset of the trial and prior selections within the trial.

A comparison of the keypeck and treadle-press operants in the pigeon: Fixed-interval schedule of reinforcement

Four pigeons were tested under 30-, 100-, 600-, and 1,000-sec values of the fixed-interval schedule of food presentation using both the treadle-press and the keypeck operants. Response rate was higher for the keypeck response at all schedule values, but the response pattern was similar for both operants. Quantitative measures of response rate and curvature showed considerable variability, with indication of a differential change in quarter-life as a function of schedule value, with the function being steeper for the treadle response than for the key response. Overall, responding on the key and treadle showed more similarities than differences.

The sensitivity of the pigeon’s keypeck to the differential reinforcement of long interresponse times

Keypecking of pigeons was studied under differential-reinforcement-of-low-rate (DRL) and variable-interval (VI) schedules in which the interreinforcement times on the two schedules were equated by a yoking procedure. Each schedule was available for half of every session and a change of schedule was signaled by a change of key color. The value of the DRL schedule was varied from .5 to 300 sec. Response rates were always higher in the VI schedule, but within sessions there was a sharp change in response rate coincident with the change in schedule only under lower schedule values. A group without prior training was tested with a 180-sec schedule value, and it, too, developed a higher response rate during the VI schedule, showing that the effect was not dependent on prior experience under low schedule values. In all conditions except the .5- and 1-sec values of the schedule, the mean proportion of responses emitted during the VI schedule was approximately .85 of the responses emitted during both schedules. The conclusion was that the requirement of a minimum interresponse time for reinforcement may work its effect by determining which responses may occur just prior to the reinforced response and thus receive delayed reinforcement.

The effect of localizing the stimulus at the response key on generalization along the hue dimension in the pigeon

After training with a variable-interval schedule of positive reinforcement, pigeons were tested for stimulus generalization along the hue dimension. For one group, the stimulus was located on the response key. For a second group, the stimulus was located on a surface adjacent to the response key. The stimulus-on-key group produced the typical steep gradients normally found with hue stimuli; the stimulus-off-key group produced flat gradients. After discrimination training between the presence and absence of the hue stimulus, both groups produced decremental gradients. In a second experiment, naive pigeons were trained to peck a transparent key with the stimulus surface located approximately 3.8 cm behind the key. When tested for generalization, the hue gradients were decremental. The results suggest that location of the stimulus in the line of sight with pecking is a necessary condition for stimulus control by hue after nondifferential training.

Autopecking With Gross Movement Physically Restrained

Pigeons were studied under the autoshaping, automaintenance, and omission procedures while either physically restrained or free in the operant chamber. Physical restraint had no effect on autoshaping or automaintenance, possibly because of a ceiling effect. The percentage of trials with at least one response under the omission procedure was lower for the restraint condition than the free condition for most subjects. These data show that, while gross approach movements are not necessary for autoshaping to occur, movement does play some role in autopecking due to light-food pairings.