The act of detecting a stimulus contaminates measures of conscious experience with decision biases.
Nature communications May 8, 2026 Nicolás Sánchez-Fuenzalida, Chris Jungerius, Stephen M Fleming et al.
A key challenge in consciousness research is separating genuine conscious perception from non-perceptual biases when observers report whether they saw a stimulus. In experiments with 505 participants who detected and reproduced dim or absent visual patterns under different conditions—attentional cues, asymmetrical base rates, and payoff schemes—a Hurdle-Gaussian model decomposed responses into a perceptual continuous component and a non-perceptual hurdle component. Statistical priors and reward structures shifted the non-perceptual hurdle, while attentional cues selectively shifted the perceptual component, consistent with changes in conscious experience. Crucially, including detection trials contaminated reproduction reports with non-perceptual criterion effects, urging caution in interpreting detection-based measures in consciousness research.