Absolute Judgment for One- and Two-Dimensional Stimuli Embedded in Gaussian Noise

Publisher Website
Google Scholar
Abstract
The effect on human performance of adding Gaussian noise or disturbance to the stimuli in absolute judgment tasks involving both 1- and 2-dimensional stimuli was examined. A CRT [Cathode ray tube]-type display was used. For each selected stimulus value (both an X-value and a Y-value were generated in the 2-dimensional case), 10 values (or 10 pairs of values in the 2-dimensional case) were generated from a 0- mean Gaussian variate, added to the selected stimulus value and then served as the coordinate values for the 10 points that were displayed sequentially on the CRT. A PDP-8 digital computer was programmed to generate these points. The independent variables were the stimulus uncertainty and the scatter (variance) of the stimulus noise. Human performance, in terms of the information transmitted and rms [root mean square] error as functions of stimulus uncertainty, was significantly reduced as the noise variance increased. The absolute judgment performance along the X-axis was superior to that along the Y-axis. No significant differences existed in the performance along the X-axis between the 1- and the 2-dimensional case. Variations in the noise variance along 1 axis also affected the performance along the other axis, although not significantly.

This publication has 13 references indexed in Scilit: