Toward Noninvasive 3-D Imaging of the Time Course of Cortical Activity: Investigation of the Depth of the Event-Related Optical Signal

Abstract

The event-related optical signal (EROS) has been recently proposed as a method for studying noninvasively the time course of activity in localized cortical areas (G. Gratton and M. Fabiani, 1998, Psychonomic Bull. Rev. 5: 535–563). Previous data have shown that EROS has very good temporal resolution and can provide detailed surface activity maps. In the present study we investigated whether the depth of the active area can also be estimated. Nine subjects were run in a study in which the eccentricity of the visual stimuli was varied, and EROS was recorded from medial occipital areas using multiple source–detector distances. Seven of the same subjects were also run through a functional magnetic resonance imaging (fMRI) study using the same protocol. The fMRI data indicated that the depth from the head surface to the cortical area activated increased systematically with the eccentricity of the visual stimuli. The EROS recording indicated a response with a latency of 60–80 ms from stimulation. This response varied systematically with eccentricity, so that the greater the eccentricity of the stimuli, the longer the source–detector distance (and thus the depth) at which the EROS effect was observed. The depth of the brain area generating the EROS effect was estimated using a simple algorithm derived from phantom studies on homogeneous media. The average depth estimates for each eccentricity condition obtained with EROS corresponded with those obtained with fMRI, with discrepancies of less than 1 mm. These data demonstrate that multiple source–detector distances can be used to estimate the depth of the cortical areas responsible for the EROS effects.