Abstract
Neurons in the supplementary eye field (SEF) are differentially active for go and nogo conditions in an oculomotor decision task (Kim et al., 2005). However, these neurons appear to signal the rule and not the decision when the animal makes errors (Yang et al., 2010). This raises the question of whether the SEF is involved in the decision process, or whether activity here is merely correlative. In the experiments, animals performed a difficult version of our go/nogo ocular baseball task while microstimulation was applied in the attempt to influence the decision. The rule of the task is to pursue a target visually if it intersects a specific zone on the monitor, and maintain fixation if it does not. Target trajectories were chosen from a random set spaced narrowly at the corner of the zone, including one ambiguous trajectory. Microstimulation was applied on average 100–200 msec before zone intersection (duration 300 msec). Current amplitude was usually in the 50–75 μA range, and stimulated and control trials were randomly interleaved. We found that stimulation affected the decision bias of the monkey at 12 of 27 sites. At nine sites, the animal was biased toward go decisions, and at three sites, toward nogo decisions. Surprisingly, at five other sites we found that overall performance was enhanced; the monkey performed better than control at judging both go and nogo trials. We then used Monte Carlo simulations and a bootstrap technique to test the validity of our results. The percentage of significant bias or enhancement sites produced by the bootstrap (58.2% and 51.1% respectively) was significantly smaller than in the original data (66.1% and 67.1%). Therefore, we conclude that SEF activity affects the decision, and that microstimulation here increases the weighting of the normal SEF contribution to the decision process.
Supported by NIH EY117720.