Latency differentiation of hits and false alarms in an operant-psychophysical test.

Rats detected the luminance difference of standard and comparison stimuli in a go/no-go procedure. A key press was reinforced by brain stimulation only when the key's luminance was 10.53 ft-L (36.01 cd/m(2)), and key presses to dimmer comparison values produced a 5-sec timeout. These asymmetrical reinforcement contingencies maximized the bias toward hits and false alarms ("yes" reports), and thus the number of latencies available for analysis. False alarm latencies exceeded hit latencies, with the magnitude of differentiation proportional to luminance difference, demonstrating stimulus control on the very occasions that errors (key presses to comparison luminances) were emitted. Overall latencies decreased when the standard-comparison luminance difference was made smaller, suggesting a reduction in observing time when the stimuli became indiscriminable.