Confidence and central tendency in perceptual judgment

This paper theoretically and empirically investigates the role of noisy cognition in perceptual judgment, focusing on central tendency effect: well-known empirical regularity that judgments are biased towards center stimulus distribution. Based a formal Bayesian framework, we generate predictions about relationships between subjective confidence, tendency, response variability. Specifically, our model clarifies lower confidence as measure posterior uncertainty judgment should predict (i) sensitivity magnitude estimates to objective stimuli; (ii) higher mean distribution; (iii) stronger effect at magnitudes; (iv) To test these predictions, collect large-scale experimental data set additionally re-analyze from several previous experiments. Across sets, is strongly predictive variability, both correlationally when exogenously manipulate sensory noise. Our results consistent with (but not necessarily uniquely explained by) models tendency.