Item

Abstract

Modulation of neural signals by saccadic eye movements (SEMs) has been reported in various cortical regions involved in visual and auditory perception. This work typically focused on how a SEM-related signal could prepare sensory areas for novel incoming visual stimuli following the end of SEMs [1]. In contrast to a low-dimensional timing signal, SEM modulation could convey information about the size and direction of SEMs, which could be useful for predicting what types of visual information to expect based on peripheral vision. Indeed, the local field potential (LFP) from visual areas V4 and TE in monkeys showed SEM modulation that was different for contra- and ipsi-versive horizontal SEMs [2]. In addition, single unit activity in primate auditory cortex is known to be sensitive to eye position in the orbit, as measured during spontaneous and auditory-evoked responses [4]. Our previous results showed that the upper bank superior temporal sulcus (STS) and core and belt auditory cortex (ACx) are modulated by SEMs [3], though it was not clear whether this was only a low-dimensional timing signal, or whether the response was also sensitive to the direction and/or magnitude of the saccade. Because oculomotor brain regions that project to STS and ACx are known to encode saccade amplitude while also controlling contraversive SEMs, we measured the influence of these two saccade metrics on ACx and STS activity in two awake, behaving monkeys during visually- and non-visually- guided SEMs. We analyzed the frequency-specific (spectral) LFP with a combination of Hilbert transforms and the Chronux signal processing toolbox. During SEMs, spectral power in the gamma and very high frequency ranges (60-200Hz) was correlated with the degree of contraversive movement and with overall saccade amplitude. The relationship between high-frequency power and saccade measures was seen in STS and ACx, during both visually-driven and non-visually-driven SEMs. In addition, high-frequency increases in power were often accompanied by decreases of power in the delta band (1-4 Hz). Our results suggest SEM modulation of activity in ACx and STS contains both spatial and temporal information, regardless of whether the SEM is visually-driven. Such a signal could be useful for scale-invariant feature integration, for the trans-saccadic integration of objects in complex scenes, or it may merely be the consequence of the signals conveyed from ipsilateral oculomotor areas.