This proposal aims to investigate neural mechanisms underlying attention in visual cortex. Attention refers to the ability to preferentially process a limited number of behaviorally relevant stimuli (targets), and suppress processing of irrelevant stimuli (distracters). Studies in monkeys have revealed visual cortical neurons that """"""""mimic"""""""" this ability. When two stimuli are inside such a neuron's receptive field, its response to the first stimulus will be modulated by the second. However, when the first stimulus (target) is made behaviorally relevant, and the monkey attends to it, then the response modulation caused by the second stimulus (distracter) is eliminated, after which the neuron selectively processes the target. Recent data suggest a bias signal that may lead to the selective processing of targets amongst distracters, but the exact mechanisms of target selection remain unknown. One theory holds that target selection can be the result of a serial search, during which an attention focus moves from one to the next item in the stimulus array. Another view holds that target selection follows a parallel operation, in which attention is distributed in parallel across items in the array. We propose an ensemble recording approach in monkeys to investigate attention.
The first aim i s to test serial and parallel theories, by measuring ensemble activity preceding target selection. It is hypothesized that when target search is difficult, different activity patterns will be associated with the serial scanning of each stimulus in the array. According to a parallel hypothesis, only one stable activity pattern, similar to the one observed when the target is presented alone, would emerge preceding target selection.
The second aim i s to measure neural correlates of quickly changing expectations about identity or location of the target, and to show that serial search is a special case of dynamically changing expectations.
The third aim i s to study the contributions of temporal and rate modulations to the biasing signals that lead to target selection. Preliminary data show that neural interactions during attention can be studied successfully at the population level, that area V4 (where recordings will be carried out) is critical for attention, and that ensemble recording is feasible in monkeys. The research is important, as it will settle a century-old debate about the mechanisms underlying target selection. Moreover, upon completion of this project a model system of attention will be available which will help to develop more efficient therapies for the millions of children in the USA affected by attention deficit disorders. Indeed, better-focused drug therapies can be developed by testing their effects directly on attention-related ensemble activity in monkeys.
| De Weerd, P; Smith, E; Greenberg, P (2006) Effects of selective attention on perceptual filling-in. J Cogn Neurosci 18:335-47 |
| De Weerd, Peter; Desimone, Robert; Ungerleider, Leslie G (2003) Generalized deficits in visual selective attention after V4 and TEO lesions in macaques. Eur J Neurosci 18:1671-91 |
| De Weerd, P; Reinke, K; Ryan, L et al. (2003) Cortical mechanisms for acquisition and performance of bimanual motor sequences. Neuroimage 19:1405-16 |
