Abstract
Wallach (1985) hypothesized that people will acquire a new cue to a perceptual judgment when a novel stimulus is correlated with a known cue. We tested this hypothesis in four experiments in which we introduced novel and systematic correlations between different pairs of sensory signals. The first three experiments each paired the visual motion direction of a random-dot-kinematogram with one of three signals. Experiment 1 used an auditory signal (broadband auditory stimulus) whereas Experiments 2 & 3 used visual signals (binocular disparity or brightness). Using a near-threshold conflict between the actual visual motion direction and the direction indicated by the new signal, subjects were evaluated for whether they used the new signal when making direction judgments. Though the procedure was identical in all three experiments, and the signals were each adjusted to be equally salient, subjects learned to use the auditory signal as a cue to motion direction but failed to learn to use either of the novel visual signals. Experiment 4 checked whether subjects' inability to learn the new visual cues was particular to the motion direction discrimination task by asking subjects to make a different perceptual judgment (determining light source direction). As in Experiments 2 & 3, however, subjects were unable to learn a new visual cue. Overall, the results suggest that the learning processes underlying cue acquisition are not equipotent but rather are biased—novel contingencies involving cross-modal signal pairings are easier to learn than novel contingencies involving only visual signals.
This work was supported by NIH research grant RO1-EY13149.