As we experience our environments, our brains are constantly computing inferences about the most likely state of the world: Are those lights in the distance headlights or streetlights?  If headlights, how fast are they going?  Is it safe to cross the street?  These perceptual decisions are often accompanied by subjective evaluations of certainty: Are you sure those are headlights? Are you sure they’re approaching slowly enough?  But there are other subjective aspects of perception, too, such as how bright the lights seem to be, regardless of their identity.  Are the neural representations and computations underlying these two kinds of subjective experiences linked in any way, or is a light’s brightness completely separable from your uncertainty about whether it is a headlight or a streetlight?

Despite increasing efforts to answer these questions, how the brain computes metacognitive judgments of certainty versus other subjective evaluations is not well understood.  In this talk, I will present behavioral, neural, and computational evidence that the neural computations underlying subjective confidence are inextricably intertwined with those underlying other subjective experiences, including perceptual awareness.  This fundamental link explains many puzzling and seemingly sub-optimal metacognitive behaviors seen across species by providing a biologically plausible substrate that can parsimoniously describe perception, metacognition, and subjective awareness