Variable-interval escape from stimuli accompanied by shocks

Individual performances of three rats were examined under a procedure in which steady rates of bar pressing were maintained by conditioned aversive stimulation. Originally neutral visual and auditory stimuli were accompanied by widely and irregularly spaced pulses of shock; they were terminated on a variable-interval schedule by pressing a bar. The contingencies between behavior and shock were also duplicated in a control procedure in which no visual or auditory stimuli were provided. Pressing observed under the control procedure was attributed to differences in the aversiveness of pressing and nonpressing behavior engendered by differences in the incidence of shock following the two classes of behavior. Increased rates with visual and auditory stimuli were attributed to termination of conditioned aversive stimulation. Control rates declined more rapidly than did experimental rates as the mean interval between successive shocks was lengthened; both rates tended to decline when less than 60 sec was allowed as time out from shocks following the successful response. In the control procedure, discrimination between the continuation and discontinuation of the shock series, as measured by relative rates, depended on the relative length of the interval between shocks and the time-out period. Regular warm-up accelerations in rate were noted following an initial delay in responding at the beginning of each session. The length of time required for the warm-up depended on the length of the mean interval between shocks, indicating that exposure to a certain amount of shock was required to establish a supporting state for the observed performance.     

A note on mathematical models for learning

It is shown that a rational learning curve developed by Estes for paired associate learning is a special case of a more general function. The latter is the product of two functions and assumes that the discovery and fixation aspects of learning are independent. The indications are that the form of one of the functions has not been tested sharply by Estes in the paired associate learning setting.