Abstract
There is converging psychophysical and neuropsychological evidences for distinct processing of object attributes within the ventral and dorsal stream of the visual system depending on the purpose of the behaviour, i.e. semantic categorisation or goal directed action. Motor act can however be triggered according to linguistic information processed within the ventral stream or spatial information processed within the dorsal stream but with different time constraints due to differences in the magno-parvocellular neural latencies. We tested the influence of presenting congruent or incongruent spatial information when processing linguistic stimulus (or the opposite) to specifying a right or left motor response. We found that the time to respond to a linguistic stimulus (the word “DROITE” (right) or “GAUCHE” (left) displayed within a rectangular frame) was 186ms greater than that to react to a spatial stimulus (a target-dot to the right or left side of the rectangular frame). No interference effect on spatial accuracy and reaction time was observed when responding to the spatial cue whatever the linguistic context. On the contrary, a strong interference of non-congruent spatial information on both reaction time and response accuracy was found when responding to the linguistic cue but predominantly for responses with short reaction time (300 to 400ms). Response selection can thus be influenced by non-relevant visual information suggesting that the ventral and dorsal visual streams are activated simultaneously and compete to specifying the relevant visual signal for action but with different time constraints.
Supported by European Science Foundation, Eurocores CNCC CRP grant, and ANR “neurosciences, neurologie et psychiatrie” program from the French Ministry.