Motion Coherence Thresholds for Adult Participants Viewing Different Patterns of Optic Flow

In adults, radial optic flow evokes stronger brain activity than laminar or rotational flow. Optic flow also evokes different activation patterns depending on motion speed (Fesi et al., 2014). This study examined whether the detection of optic flow in adult observers varies by pattern and speed in ways consistent with prior physiological evidence. Adult observers (n=10; 19.2-22.2 years, M=21.4) viewed two side-by-side, time varying (1.2 Hz coherent/incoherent cycle) annular shaped (18 deg outer/5 deg inner diameter) optic flow displays at a viewing distance of 60 cm. One display depicted random (0% coherent) motion while the other depicted radial or translational motion at one of four fixed coherence levels (5, 10, 15 or 20%). Observers fixated centrally and judged which side contained coherent motion, indicating the choice by a keypress. Within a single run, speed was either 2 or 8 deg/s. Between 2 and 4 runs were collected per participant in a single visit. We analyzed proportion correct and response times using generalized mixed effects modeling in R. As predicted, proportion correct judgments increased with motion coherence, and the response time of correct judgments declined. There were no statistically significant differences in proportion correct judgments by speed or pattern type; however, a significant coherence by pattern interaction, p< .0001, indicated that observers more readily detect low coherence laminar flows than radial ones. Reaction time data suggest that observers are faster to detect laminar patterns, p< .0001, and slow (2 deg/s) speeds, p< .001. Taken together the results suggest that sensitivity to detect optic flow in noise varies by pattern type and speed, in ways partially consistent with prior physiological results.