Factorial effects in the categorization of externally distributed stimulus samples

On each trial a sample point randomly drawn from one of two normal probability distributions was exhibited to S, who had to guess whether the sample was a “1” or a “2.” He was then given feedback, which was determined by which of the two distributions the sample point derived from. Two continua were employed: dot position on a file card, and groyness of square in. patches. Three levels of d’ were employed. The function giving the probability of a response “1” for different sample values was sharper for the higher d’ conditions, and for the dot position continuum. Cessation of feedback resulted in improved performance for the low d’ condition. Incentive payoff, confidence ratings, and experimenter had virtually no effects on task performance.