The relationship between mechanisms for curvature and Vernier detections.

Tyler (Journal of Physiology 228 (1973) 637-647) reported that the sensitivity function for curvature detection has band-pass characteristics with respect to shape frequency. The shape of this function suggests that the curvature detection mechanism might consist of a set of sub-mechanisms each tuned for shape frequencies similar to mechanisms for contrast detection. In Experiment 1, we applied an adaptation for shape frequency to investigate whether shape frequency is the critical feature in curvature detection. We successfully replicated the U shape threshold function reported by Tyler (1973), but found that selective adaptation does not occur for shape frequency. These results suggest that shape frequency is not critical for curvature detection. In Experiment 2, we examined the possibility that curvature detection for the highest sensitivity range is mediated by a displacement process similar to that for Vernier acuity. The results indicate that curvature detection for this range is mediated by a mechanism that is sensitive to spatial displacement, and suggests the existence of a common mechanism for spatial displacement that mediates both curvature and Vernier detections. By comparing our results with those of several previous studies, we concluded that the apparent band-pass characteristic for curvature detection is produced by combining the characteristics of three separate limiting factors: local orientation at lower frequencies, spatial acuity at higher frequencies, and displacement in the highest sensitivity area.