NEURAL CORRELATES OF PARTIAL TRANSMISSION OF SENSORIMOTOR INFORMATIONIN THE CEREBRAL-CORTEX

Using single neuron recordings in monkey primary motor (MI) cortex, tw o series of experiments were conducted in order to know whether respon se preparation can begin before perceptual processing finishes, thus p roviding evidence for a temporal overlap of perceptual and motor proce sses. In Experiment 1, a ''left/right, Go/No-Go'' reaction time (RT) t ask was used. One monkey was trained to perform wrist flexion/extensio n movements to align a pointer with visual targets. The visual display was organized to provide a two-dimensional stimulus: side (an easy di scrimination between left and right targets) which determined movement direction, and distance (a difficult discrimination between distal an d proximal targets) which determined whether or not the movement was t o be made. Changes in neuronal activity, when they were time-locked to the stimulus, were almost similar in the Go and No-Go trials, and whe n they were time-locked to movement onset, were markedly reduced in No -Go as compared to Go trials. In Experiment 2, a stimulus-response com patibility (SRC) task was used. Two monkeys were trained to align a po inter with visual targets, on either left or right. In the spatially ' 'compatible'' trials, they had to point at the stimulus position, wher eas in the ''incompatible'' trials, they had to point at the target lo cated in the opposite side. For 12.5% of neurons, changes in activity associated with incompatible trials looked like changes in activity as sociated with movements performed in the opposite direction during com patible trials, thus suggesting the hypothesis of an automatic activat ion of the congruent, but incorrect response. Results of both experime nts provide evidence for a partial transmission of information from vi sual to motor cortical areas: that is, in the No-Go trials of the firs t task, information about movement direction, before the decision to p erform or not this movement was made, and, in the incompatible trials of the SRC task, information about the congruent, but incorrect respon se, before the incongruent, but correct response was programmed.