General overview

My work seeks to deepen our understanding of conscious awareness and metacognition using visual perception as a model system. (I will sometimes use the term “awareness” to refer interchangeably to metacognition and consciousness.) A central concern of this work is to distinguish “first-order” or “type 1” perceptual processing from “higher-order” or “type 2” processing, where type 1 processing involves forming objective judgments about the external world and type 2 processing involves forming subjective judgments about one’s internal perceptual processing. For instance, a type 1 judgment might be “there is an apple on the table,” whereas a type 2 judgment might be “I’m not completely sure it’s an apple” (a metacognitive judgment about confidence) or “I didn’t quite see it clearly” (a judgment about the qualities of subjective visual experience).

Distinguishing type 1 and type 2 processes is non-trivial since they typically correlate well– as objective performance in a perceptual task improves, awareness becomes more vivid and confidence increases and becomes more diagnostic of accuracy. Yet, to understand type 2 processes in and of themselves, it is crucial to isolate them from the confounding factor of these typically correlated type 1 processes.

For instance, a common approach to studying awareness is to compare two experimental conditions where an observer reports different levels of awareness of a stimulus while performing a discrimination or detection task. By examining how other cognitive and neural processes differ in the two conditions, we can try to infer something about the cognitive role and neural mechanisms of awareness. However, if care is not taken to dissociate type 1 and type 2 processes, then the two conditions will likely differ not only in awareness (type 2 processing), but also basic stimulus processing and task performance (type 1 processing). This opens the possibility that many of the cognitive functions and neural mechanisms that differ in the two conditions may be attributable to type 1, rather than type 2, processes. This ambiguity obscures our attempts to make specific inferences about type 2 processing, i.e. awareness in and of itself.

My work (in collaboration with many wonderful and talented colleagues) has approached this challenge using a combination of carefully designed visual psychophysics paradigms, an analysis framework couched in signal detection theory, and a wide array of neuroscientific tools for measuring and intervening in neural processes. Visual psychophysics allows for tight control of stimulus properties and type 1 performance, and my work has explored various methods for extending this experimental control to the production of empirical dissociations between type 1 and type 2 performance.

I have also developed a novel extension of signal detection theory to type 2 judgments. This analytical framework provides a theoretically grounded way for separately characterizing type 1 and type 2 performance even in cases where an empirical dissociation cannot be achieved.