Human Cortex

Mapping Receptive fields in human visual cortex

Most of our understanding of the functional organization of human visual cortex comes from lesion and fMRI studies and by extrapolation from results obtained by neuroanatomical and neurophysiological recordings in nonhuman primates. Although some single unit and field potential recordings have been made in human visual cortex, none has provided quantitative characterization of receptive field properties of individual sites.

We used subdural electrodes to get high-quality recordings that can be used to quantitatively assess these features of cortical organization. We use this technique to show topographic organization in early cortex and provide an estimate of receptive field size and response latency in various regions of visual cortex. The spatial and temporal resolution of this technique are good enough to examine issues such as the modulation by attention and the trial-to-trial the correlation with behavior for activity in different visual areas.

Modulation of visual response by Attention

Although single unit recordings in macaque monkeys have found that attention has only modest effects on the responses of neurons in early visual cortex, fMRI studies with human subjects have described much larger effects. To better understand the origins of this difference, we have recorded visual responses from intracranial electrodes in human subjects. All data were obtained from electrodes placed near the occipital pole (median 3.5 cm, range 0.0-4.0 cm), over sites that had discrete receptive fields of moderate eccentricity (mean eccentricity 5.0°, range 2.0-5.7°). Subjects performed a peripheral orientation change detection task in which two Gabors were presented during fixation: one in the receptive field of the electrode under study and one outside of the receptive field in the diagonally opposite quadrant. Subjects had to respond to an orientation change at the cued location. Eye position was monitored using EOG. Task parameters were adjusted to keep the task challenging (median correct 74%, range 52-76%). Visual responses were obtained by averaging the responses to about 200 stimulus presentations. Attention did not appreciably alter response waveform or increase response amplitude (median 1.00, range 0.98-1.04). These findings are consistent with previous work from macaques that suggests that spatial attention has only a modest effect on the responses of neurons in early visual cortex.