CAREER: Neural Mechanisms of Learning to Attend

Princeton University, Princeton NJ

Investigators

Abstract

The world is brimming with sensory information, flooding our senses with sights and sounds and smells. To avoid being overwhelmed, we must filter this stream of sensory information and focus on what is most important for our current goals. Attention allows us to focus on stimuli that are most relevant to us while ignoring distracting stimuli. But, how do we learn, in a given task or situation, what is relevant and what is a distractor? This project explores the fundamental question of how we learn to attend to task-relevant information. By providing a deeper understanding of attention, our work will lay the foundation for improving diagnoses and treatments of individuals who struggle with attention-deficit disorders. These research goals are accompanied by a complementary educational plan will engage K-12 and undergraduate students in curiosity-driven, interactive research experiences to study how attention affects their everyday lives. Previous work suggests that visual attention relies on an ‘attentional template’ that represents what information is relevant in a given situation. Attentional templates are represented in prefrontal and parietal cortex in the brain and act, through top-down connections, to bias sensory representations in sensory cortex. This allows the brain to selectively increase the neural representation of task-relevant stimuli and reduce the representation of distracting stimuli. In this way, attentional templates allow us to focus our cognitive resources on task-relevant stimuli. Earlier studies have investigated the mechanisms of attention after the template has been acquired in well-learned visual tasks. The current project will build on this work, using novel experimental tasks to study how the brain discovers what is relevant and learns the appropriate attentional template. This project will utilize a non-human primate (the macaque monkey) as a model organism to study the neural basis of attention and cognition yielding implications and insights for human cognition. A first research goal is to understand the role of prefrontal and parietal cortex in learning new attentional templates by simultaneously recording neural activity from both prefrontal and parietal cortex while monkeys repeatedly learn new attentional templates. A second research goal is to understand how learning a new attentional template transforms or ‘warps’ the high-dimensional neural representation of stimuli in prefrontal and visual cortex. Recent work has suggested visual stimuli are represented in a continuous ‘cognitive map’ that represents different visual features along different dimensions in neural space. Experiments will test the hypothesis that attention alters this cognitive map by ‘warping’ neural space such that attended features are represented more strongly. This will be studied by analyzing how new attentional templates change the representations of stimuli in the neural population, and how they alter the dynamic functional connectivity between regions in the visual attention network. The results of these experiments in non-human primates will be integrated with parallel studies of attention in humans.The results of these experiments promise to lead to a deeper understanding of the neural correlates of learning, attention, and cognition at both a neuronal and a brain network level. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.