Sustained Attention's Floodlight Dynamically Influences Cognitive and Neural Representations

Despite our best efforts, humans are unable to maintain the focus of our attention indefinitely. The mnemonic consequences of attentional lapses are evident in our daily lives, such as forgetting where we left our phone after mindlessly setting it down. Our ability to sustain attention is consistent at the trait level---individuals attend either reliably well or reliably poorly---but varies at the state level, with all individuals fluctuating between periods of high and low sustained attentional states across seconds to minutes. Yet, we lack a full understanding of the brain-based mechanisms involved in attention’s dynamics. The current dissertation aims to better characterize the nature of sustained attention through investigation of its neural and behavioral correlates. Using patterns of brain connectivity, I identify distributed networks that track sustained attention across individuals (Chapter 1), across contexts (Chapter 2), and across moments (Chapter 3), suggesting that attentional maintenance reconfigures functional relationships throughout the brain. I then explore the consequences of dynamic attentional states on the fate of information reflected in memory (Chapter 4) and in neural representations during encoding (Chapters 5 and 6). Evidence suggests that heightened sustained attentional states, rather than sharpening spotlight-like selectivity, increase processing capacity broadly in a manner more akin to a floodlight. Increased processing is further supported by a strengthening in representational fidelity under better attentional states as measured with spatially- and temporally-resolved neural signals. Results converge on the characterization of sustained attention as a global mechanism that supports cognition through the strengthening of functional connections and information representations in the brain.