Motor performance and response consistency as a function of age

The specificity versus generality of motor performance and motor response consistency ws investigated as a function of age. 120 Ss, 30 each at age 7, 11, 15 and 19 yr., were given 120 practice trials (60 trials per session with 24 hr. interpolated between sessions) on both simple and choice RT tasks. For motor performance the reliability of individual differences were high in both tasks at all ages, while the amount of generality was moderately high in the two younger groups but diminished with age. The reliability coefficients for motor response consistency were low for both tasks but, with two exceptions, statistically significant. There was no evidences for generality in motor response consistency at any age.     

Schedule-induced drinking as a function of percentage reinforcement

Drinking was recorded in rats while lever pressing was maintained on a series of percentage reinforcement schedules in which the per cent of 1-min fixed intervals terminating with food was 100, 90, 30, 70, 10, 50, and 100%. Intervals in which a pellet was omitted were terminated by brief light flash and click stimuli that were also correlated with food presentations. Drinking failed to develop in five of six subjects following intervals in which the brief stimuli were presented regardless of percentage reinforcement. Postpellet drinking, which followed intervals terminated with pellet delivery, however, increased in both duration and amount ingested per interval as percentage reinforcement was systematically decreased. The increase in postpellet drinking above that produced by 100% reinforcement was interpreted as an analogue of the positive-contrast effect observed with food-reinforced operants.     

Response latency as a function of reinforcement schedule

Four Ss were trained to press and hold down a telegraph key in the presence of a light. Subsequent release of the key during a tone was followed by water reinforcement. The schedule of reinforcement for key release was varied, and its effects on the latency (RT) of key release to the tone were studied. Both median RT and variability of RT were found to be inversely related to frequency of reinforcement as determined by the schedule.     

Behavioral aftereffects of reinforcement and its omission as a function of reinforcement magnitude

Rats responded on a multiple fixed-interval fixed-interval schedule of reinforcement. Each complete cycle of the multiple schedule was separated from the next by a relatively long period of timeout from all schedule contingencies. A response at the end of the second component of each cycle was always reinforced with an invariant reinforcement magnitude, while reinforcement magnitude and reinforcement omission were systematically varied in the first component. Response rate in the first component was a monotonic function of reinforcement magnitude in that component. These changes in response rate in the first component did not affect response rate in the second component. When reinforcement was omitted on 50% of occasions in the first component, following reinforcement there was a reduction in response rate in the second component that was monotonically related to reinforcement magnitude. Following reinforcement omission there was an increase in response rate in the second component that was unrelated to reinforcement magnitude. When reinforcement was omitted on 100% of occasions in the first component, behavioral contrast was observed.     

Conditioned reinforcement as a function of duration of stimulus

Pigeons were provided with three keys. Pecking the center key produced grain on a schedule that alternated at unpredictable times between a variable-interval component and extinction. On concurrent variable-interval schedules, pecking either side key produced a stimulus associated with the variable-interval component on the center key provided that said schedule was currently in effect. The independent variable was the length of time this stimulus remained on the keys. Pecking one side key produced the stimulus for 27 seconds, whereas the duration produced by pecking the other key varied for successive blocks of sessions. For the first four birds, the values tested were 3, 9, 27, and 81 seconds. For the second group, numbering three birds, the values tested were 1, 3, 9, and 27 seconds. The dependent variable was the proportion of total side key pecks that occurred on the variable key. For all birds, the function was positive in slope and negative in acceleration. This finding supports a formulation that ascribes the maintenance of observing responses in a normal setting to the fact that the subject exposes itself to the positive discriminative stimulus for a longer mean duration than it does to the negative stimulus.     

Response latency as a function of amount of reinforcement

Food-deprived rats were trained to press and hold down a telegraph key in the presence of a light. Subsequent release of the key during a tone was followed by 0.15 ml of a 20-percent sucrose solution as reinforcement. The Ss were subsequently shifted to a 0-percent and to a 5-percent solution from the 20-percent base line. The median RT and the variability of RT increased markedly as a result of the shift to the lower sucrose concentrations. For all Ss, the change in median and variability was greater for the shift to the 0-percent solution than for the shift to the 5-percent solution. It is probable that median RT and variability of RT are inversely related to amount of reinforcement.     

Probability learning as a function of momentary reinforcement probability

Pigeons were trained on a probability learning task where the overall reinforcement probability was 0.50 for each response alternative but where the momentary reinforcement probability differed and depended upon the outcome of the preceding trial. In all cases, the maximum reinforcement occurred with a “win-stay, lose-shift” response pattern. When both position and color were relevant cues, the optimal response pattern was learned when the reinforcement probability for repeating the just-reinforced response was 0.80 but not when the probability was 0.65. When only color was relevant, learning occurred much more slowly, and only for subjects trained on large fixed ratio requirements.