Modulation of local field potentials in macaque frontal cortex during visual and memory guided saccades

Working memory refers to short term storage of information used to guide ongoing actions. An extensive literature in neuroscience attempts to explain the brain physiology associated with working memory in humans and nonhuman primates. These studies have primarily relied on measures of single neuron action potential rates or measures of the electroencephalogram. In the present thesis, local field potentials (LFPs) were recorded from electrodes in periarcuate and anterior cingulate cortices of a rhesus monkey performing a delayed saccade task. The task involves trails in which the target for a saccade remains illuminated during the delay period and trials in which the target for a saccade disappears during the delay period. This thesis characterizes task dependent modulation of the power spectra of the LFPs recorded during task performance. The results of this thesis showed that LFP power decreased significantly during the delay compared to baseline periods of both trial types in the majority of recordings from both the periarcuate and anterior cingulate regions. In the recordings from the periarcuate cortex, delay period power increased relative to baseline period power and was significantly tuned for one saccade target location in 20% of the recordings. No recordings from anterior cingulate showed significant direction tuning. In periarcuate recordings, delay vs. baseline period increases in power in the gamma band were more likely in shallow cortical sites (<1.5 mm from the cortical surface). In anterior cingulate recordings, smaller decreases in delay vs. baseline period power in the alpha and theta bands were more likely in shallow cortical sites. These results were interpreted as evidence for synchronization of shallow cortical LFPs and desynchronization of deep cortical LFPs caused by increased activity of pyramidal neurons and interneurons during saccade planning. Additionally, for both periarcuate and anterior cingulate recordings, power was significantly lower at frequencies below 30 Hz and higher at frequencies above 30 Hz during the delay period of memory vs. visually guided saccade trials. This effect was not correlated with cortical depth of recording site. These results support the premise that working memory is associated with increased synchronization in local cortical circuits in the primate brain.