Abstract
We investigated the influence of covert spatial attention on the perception of visual motion in human subjects. In our paradigm, an exogenous cue was used to attract the spatial attention of subjects to one of two random dot patterns (RDP, diameter 5 deg, moving at 8 deg/s within stationary virtual apertures), centered at 5 degrees eccentricity to the left and right, respectively, of a central fixation point. Importantly, we aimed to manipulate the amount of spatial attention deployed by varying the validity of the cue. The target stimulus was a brief coherent motion signal (75 ms) within one RDP, which was accompanied by a contralateral and followed by a bilateral RDP mask containing a mixture of all possible directions of motion. By using different levels of cue validity, we created four attentional conditions (100%, 75%, 50%, 25%). Direction discrimination thresholds of well-trained subjects were measured using a 4-alternative forced-choice (4AFC) direction discrimination task. They were asked to report the location (left or right) as well as the motion direction (up or down relative to horizontal) of the target stimulus by pressing one of four buttons. Using such a 4AFC discrimination task, true attentional effects can be discriminated from effects of stimulus uncertainty. We additionally used blocks of trials with four different levels of motion coherence (40%, 60%, 80%, 100%) in the target stimulus. Our results show that the validity of the cue affected subjects’ discrimination thresholds in a graded manner, i.e. 100% attention to one location resulted in highest performance whereas 25% attention led to the weakest performance, approximately doubling the discrimination threshold. Motion coherence also affected performance, with discrimination thresholds increasing with decreasing levels of motion strength. These two effects on performance seem to operate independently, with no interaction found between coherence and level of graded attention.
Meeting abstract presented at VSS 2013