Purchase this article with an account.
Nestor Matthews; Fine Motion Discriminations at Isoluminance. Journal of Vision 2002;2(7):389. doi: https://doi.org/10.1167/2.7.389.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: To determine the extent to which fine direction and speed discrimination depend on luminance contrast when these two tasks are performed under identical stimulus conditions.
Method: 20 naïve undergraduates participated in a within-subjects experimental design. Two random-dot cinematograms (RDCs), that always differed from each other in speed and direction of motion, were presented successively on each trial. The dots were differently colored than the surround, and either matched the surround's luminance (isoluminant condition = 0% contrast) or differed from the surround's luminance (luminance-contrast condition = 40% contrast). The task was randomly varied, requiring either speed judgments or direction judgments. Across these two tasks, the stimulus conditions were held constant. Discrimination thresholds, defined as one half the stimulus increment needed to alter the response rate from 0.25 to 0.75, were measured for each subject, task, and luminance condition.
Results: Within-subjects ANOVAs indicated that neither speed thresholds ( F(1,19) = 0.07, p=0.78, n.s. ) nor direction thresholds ( F(1,19)=0.05, p=0.82, n.s. ) were affected when the luminance difference between the surround and the dots was eliminated. A control experiment with the same statistical power and subjects suggested that performance on the speed task was based upon motion cues and not upon time cues that were also available: Speed thresholds were significantly lower than time thresholds in both the isoluminant ( F(1,19)=27.29, p<0.0001 ) and the luminance-contrast (F(1,19)=22.8, p=0.0001 ) conditions.
Conclusions: The findings are consistent with the notion that, for RDCs, fine direction and speed discrimination are limited by neural noise that does not require luminance contrast. The present data could be modeled either by a shared noise source for the two tasks, or by task-specific noise sources that obey similar rules regarding luminance contrast.
This PDF is available to Subscribers Only