The visual scenes we typically view are cluttered with many different objects. Attentional mechanisms are required to select relevant and to filter out irrelevant information. In non-human primates, single-cell recording studies have shown that attention enhances neural responses in visual cortex to attended stimuli as compared to the same stimuli when unattended. The mechanism by which attention enhances neural activity appears to be an increase in the neuron's gain. Several factors determine attentional response enhancement, including stimulus salience, the degree of competition among multiple stimuli for neural representation, the nature of the selection (if featural or spatial), and the expectations of the individual. In the human visual cortex, our previous studies using functional magnetic resonance imaging (fMRI) have shown that attention enhances neural activity to visual stimuli, particularly in the presence of distracter stimuli, and even in the absence of any visual stimulation whatsoever. However, the mechanism by which attention enhances the gain of neural responses and the factors that determine the response modulation are not clear. The long-term objective of this application is to determine these basic attentional mechanisms operating in the human visual cortex with hypotheses derived from monkey physiology using fMRI. fMRI and behavioral studies are proposed to: 1) Determine the effects of attention on stimulus contrast; 2) Determine the relation of mechanisms for featural and spatial attention; and 3) Determine the nature of increases in baseline activity in the absence of visual stimulation. Attentional selection often falls in patients with attentional deficits, which are common in major neurologic disorders and psychiatric diseases. Therefore, it will be important to advance our knowledge of visual attention in healthy subjects in order to advance our understanding of the pathophysiology of attention and ultimately, to develop more efficient treatment for patients with attentional deficits.
Showing the most recent 10 out of 62 publications
