Date of Degree


Document Type


Degree Name



Cognitive Neuroscience


Tony Ro


Richard Brown

Subject Categories

Cognitive Neuroscience


cognition, perception, color, metacognition, 2AFC, detection


Confidence measures are sometimes used to index awareness below a participant’s criterion for what counts as being subjectively aware of a stimulus or stimulus property. If confidence ratings index awareness below subjective thresholds, then one might be tempted to use confidence ratings in place of subjective reports of whether a stimulus/stimulus property was “seen” or “unseen”. As such, a dissociation of confidence from performance may be construed as a dissociation of subjective awareness from objective task performance. This methodology has been used as supporting evidence for blindsight in typical observers (Balsdon and Azzopardi 2015; Peters and Lau 2015; Peters et al. 2017; Knotts et al., 2018). To further examine whether confidence can be used to reliably index a dissociation of awareness from objective performance, we conducted a web-based experiment using a 2AFC task to evaluate discrimination sensitivity for shape-majority decisions (a fixed ratio of two groups of shapes) under the independent variable: differentially colored shape ratios where either shape may be colored differently. Using a near-threshold color difference, we created conditions to render the color manipulation subjectively unaware. We found that confidence reports did not co-vary with performance on the primary task for participants who experienced performance increases with near-threshold color differences. Further research is necessary to support these findings.