Neuronal correlates of motivated inhibition in monkey motor cortices

The ability to rapidly inhibit actions is critical to the effective and flexible interactions with the environment, and it has been widely studied by using the countermanding (stop-signal) task. An open question is how this ability can be affected by other cognitive functions, such as motivation, and relatedly, which are the neural correlates of this influence. To address this question we designed a combined countermanding/motivation task intended to modulate the motivation to stop and to move by using the reward prospect, and we recorded the Multi-Unit Activity (MUA) by means of Utah arrays (2x48 channels) from the dorsal premotor cortex (PMd) and the primary motor cortex (M1) of a macaque monkey while engaged in the task. Indeed previous data have shown that motor cortices neuronal activity signals both movement inhibition (Mattia et al., 2013) and reward (Roesch and Olson, 2004), but it is unclear whether and how these phenomena are combined. In the countermanding/motivation task, the monkey was instructed to reach to a target after a go signal (no-stop trials) but to withhold his response when a stop-signal followed after a variable delay (stop trials). In every trial, one out of three possible reward cues (RC) was presented 1 sec before the go signal to inform about the amount of reward that would have been delivered if a correct response had been produced. RC conditions were: Go+Stop- with a higher reward for no-stop trials compared to the stop trials; Go-Stop+, with reversed reward amounts; and Go Stop with equal amounts delivered. We found that the monkey strategically adapted his behavior to the different reward prospects: reaction times (RTs) diminished and the probability of errors in the stop trials increased from Go-Stop+, to Go Stop and then to Go+Stop- conditions. Recorded MUA in both PMd and M1 was modulated by task conditions. In some channels (22/96), MUA was significantly modulated before go signal by the RC value. Of these channels, 41% showed a progressive increase of MUA after the RC presentation until movement onset; this activity was progressively higher for Go-Stop+, Go Stop and Go+Stop-. Fifty-five percent showed a decrease during the RT epoch and a further increase of MUA just before or after movement onset. Others channels (48/96) showed RC related modulations around movement onset and/or after reaching the final target but not before go signal. Interestingly, in correct stop trials, some of the channels showed an early stop-signal response which magnitude was affected by the RC value. These results suggest that motor cortices can signal and integrate motivational aspects into movement control and inhibition.