Brain-stimulation intensity, rate of self-stimulation, and reinforcement strength: an analysis through chaining

Reinforcement strengths of different intensities of brain stimulation were assessed by means of a two-member behavioral chain. A variable interval schedule of 30 sec was the first-member, and five lever presses, each rewarded with stimulation, was the second. It was found that response rate on the VI schedule continued to increase beyond the intensity value which produced peak rate on the second-member, self-stimulation lever. It was concluded (1) that brain-stimulation reinforcement strength cannot be assessed adequately by means of self-stimulation rate, and (2) the chaining technique employed in the present experiment appears promising as an analytical tool in brain-stimulation research. Finally, some aspects of the data suggested a fatigue or stimulation-adaptation phenomenon.     

Reinforcement and instructions with mental patients

An attempt was made to modify a socially desirable response of mental patients. It was found that instructions to the patients had no enduring effect unless accompanied by reinforcement. Also, it was found that reinforcement was not effective unless the reinforcement procedure was accompanied by instructions that specified the basis for the reinforcement. Maximum change in behavior was obtained when the reinforcement procedure took advantage of the existing verbal repertoire of the patients. A significant methodological finding was that substantial modification of the behavior of psychotics could be achieved by briefly delaying, rather than withholding, food reinforcement.     

Continuous opportunity for reinforcing brain stimulation

The performance of a rat provided with continuous opportunity to obtain reinforcing brain stimulation over a 20-day period is described in detail. The animal averaged 29.2 responses per min during the entire time. Testing session was terminated by experimenters, but a day-by-day analysis of behavior provided no evidence that the animal would have stopped responding.     

Exteroceptive control of response under delayed reinforcement

Three white rats, after 50 continuous reinforcements, were exposed successively, under dim illumination, to reinforcement delays of 1, 3, 5, 7.5, 10, 15, and 20 sec, with prolonged training at the 20-sec level. Behavior was maintained at each level, and an increase in interval was accompanied by an increase in post-reinforcement pause. Subsequently, under both 20- and 30-sec delays, the animals were tested during half of each daily session to determine the effect of introducing darkness during each delay interval. The result of this stimulus “support” was to regularize and increase response rate for each animal at both interval values.     

Stimulus control of avoidance behavior

The introduction of a warning signal preceding shocks greatly increased the effectiveness of avoidance responding. Periods of “warm-up” at the beginning of the session were eliminated, and the number of shocks received by the subjects was greatly reduced. With response-shock interval constant, response rate increased as the interval between the response and the onset of the warning signal was shortened. The response tended to occur shortly after the onset of the warning signal regardless of the duration of these “safe” periods. A greatly elevated response rate was maintained even when the duration of the safe period was reduced to 0.3 sec. Thus, the pre-shock signal obtained nearly exclusive control of the responding and overrode the usual “temporal discrimination” of the response-shock interval.     

Stimulus generalization and the response-reinforcement contingency

Generalization gradients along a line-tilt continuum were obtained from groups of pigeons that had been trained to peck a key on different schedules of reinforcement. In Exp. I, gradients following training on a differential-reinforcement-of-low-rate (DRL) schedule proved to be much flatter than gradients following the usual 1-min variable interval (VI) training. In Exp. II, the value of the VI schedule itself was parametrically studied; Ss trained on long VI schedules (e.g., 4-min) produced much flatter gradients than Ss trained on short VI schedules (30-sec; 1-min). The results were interpreted mainly in terms of the relative control exerted by internal, proprioceptive cues on the different reinforcement schedules. Several implications of the results for other problems in the field of stimulus generalization are discussed.