Theta oscillations in primate prefrontal and anterior cingulate cortices in forewarned reaction time tasks.

Previously, we introduced a monkey model for human frontal midline theta oscillations as a possible neural correlate of attention. It was based on homologous theta oscillations found in the monkey's prefrontal and anterior cingulate cortices (areas 9 and 32) in a self-initiated hand-movement task. However, it has not been confirmed whether theta activity in the monkey model consistently appears in other situations demanding attention. Here, we examined the detailed properties of theta oscillations in four variations of forewarned reaction time tasks with warning (S1) and imperative (S2) stimuli. We characterized the theta oscillations generated exclusively in areas 9 and 32, as follows: 1) in the S1-S2 interval where movement preparation and reward expectation were presumably involved, the theta power was higher than in the pre-S1 period; 2) in the no-go trials of go/no-go tasks instructed by S1, the theta power in the S1-S2 interval was lower than in the pre-S1 period in an asymmetrical reward condition, whereas it was moderately higher in a symmetrical condition; 3) the theta power after reward delivery was higher than in the unrewarded trials; 4) the theta power in the pre-S1 period was higher than in the resting condition; and 5) when the monkey had to guess the S1-S2 duration internally without seeing S2, the theta power in the pre-S1 period was higher than in the original S1-S2 experiment. These findings suggest that attentional loads associated with different causes can induce the same theta activity, thereby supporting the consistency of attention-dependent theta oscillations in our model.