Relationship between recognition accuracy and order of reporting stimulus dimensions.

Abstract

Stimulus objects differing in color, form, and number of objects were presented tachisto-scopically to groups of Ss. Four types of instruction were given: (a) Equal instructions in which Ss were asked to pay equal attention to and to record all three dimensions, (b) Emphasis instructions in which Ss were asked to pay primary attention to one dimension only but to record all three, (c) One-only instructions in which Ss were asked to pay primary attention to and to record only one dimension, and (d) Ordered instructions in which Ss were asked to pay equal attention to and to record all three dimensions but the order in which they were recorded was specified immediately after the stimulus exposure. The main findings and conclusions were as follows: The average accuracy of report on all three dimensions combined differed but slightly between the various types of instruction. Under Emphasis instructions the accuracy of report on the dimension Ss had been instructed to pay attention to was considerably greater than on the other two dimensions. This accuracy on the dominant dimension was not increased under one-only instructions even though Ss did not report the other two dimensions. Under ordered instructions the difference in accuracy between the first recorded dimension and the average of the other two was as large as the difference reported for Emphasis instructions. It was concluded that the selective effect of instructions on the accuracy of report for tachistoscopically presented stimuli can be accounted for in large part by the following assumptions: (a) a constant amount of information is transmitted irrespective of instructions; (b) the distribution of this between dimensions depends upon the order in which they are reported; and (c) the effect of instructions is to determine which order of report occurs.