Abstract
Posterior cingulate cortex (CGp) is strongly connected with both visual and oculomotor circuitry, and dysfunction of this area has been linked to spatial disorientation. These observations suggest that CGp might integrate visual and oculomotor data to compute spatial representations used in orienting and navigation. Consistent with this hypothesis, CGp activation is modulated by visual stimulation, eye movements, and eye position. The goal of the present study was to determine quantitatively whether CGp neuronal responses are anchored to the eyes, head, or world. To discriminate eye-centered from head-centered spatial referencing, single cells were studied while monkeys (M. mulatta) performed delayed-saccade trials initiated from three different starting positions to targets appearing along an axis passing through the previously mapped neuronal response field. Neuronal activity measured during 12 sequential epochs on each trial was first segregated by fixation position and then plotted as a function of either movement vector or final eye position. For most CGp neurons, tuning curves were better aligned when plotted as a function of final eye position than movement vector, suggesting that CGp encodes information in a head- or world-centered coordinate framework. In order to differentiate between these possibilities, subjects were rotated with respect to the visual display and asked to make saccades from a central fixation point to the array of targets used previously. Tuning curves were plotted as a function of either position of the eye in the head or position in the world and compared. These curves were better aligned when plotted as a function of position in the world than as a function of position in the head. These data suggest a role for CGp in signaling the position of salient objects in the world.
Supported by the McDonnell-Pew Program in Cognitive Neuroscience, the Whitehall Foundation, EY013496 (MLP), and F31-NS-046193 (HLD)