Each daily session had 50 new visual discrimination problems, presented consecutively, with an interval between problems of 10 sec. For
each new problem, a positive and a negative stimulus were chosen at
random from the large population of available stimuli (see Apparatus).
For each trial within a problem, the 2 stimuli were presented side by
side, occupying the left and right positions at random, and the animal
chose one stimulus by touching it. If the chosen stimulus was the negative,
both stimuli immediately disappeared from the screen. If the
chosen stimulus was the positive, the negative stimulus immediately
disappeared, and, at the same time, a vertical white line 1 mm thick
and extending the full height of the screen appeared over the center of
the positive stimulus. The positive stimulus and the white line disappeared
together 0.5 set later. The intertrial interval, measured from the
animal’s touching the stimulus, was 4 sec. Trials continued in this fashion
until either the negative or the positive stimulus was chosen consistently
for 4 trials consecutively. If the positive stimulus was chosen
4 times consecutively, a reward was dispensed at the end of the fourth
white-line display. If the negative stimulus was chosen 4 times consecutively,
the fourth choice immediately produced 10 set darkness. Either
of these events, the primary reinforcer and the primary nonreinforcer,
terminated the current problem. The number of trials per problem was
thus not fixed but was always at least 4, enabling learning curves to be
drawn for trials l-4.