Suprathreshold perceptual decisions constrain models of confidence

Abstract

Perceptual confidence is an important internal signal about the certainty of our decisions and there is a substantial debate on how it is computed. We highlight three confidence metric types from the literature: observers either use 1) the full probability distribution to compute probability correct (Probability metrics), 2) point estimates from the perceptual decision process to estimate uncertainty (Evidence-Strength metrics), or 3) heuristic confidence from stimulus-based cues to uncertainty (Heuristic metrics). These metrics are rarely tested against one another, so we examined models of all three types on a suprathreshold spatial discrimination task. Observers were shown a cloud of dots sampled from a dot generating distribution and judged if the mean of the distribution was left or right of centre. In addition to varying the horizontal position of the mean, there were two sensory uncertainty manipulations: the number of dots sampled and the spread of the generating distribution. After every two perceptual decisions, observers made a confidence forced-choice judgement whether they were more confident in the first or second decision. Model results showed that the majority of observers were best-fit by either: 1) the Heuristic model, which used dot cloud position, spread, and number of dots as cues; or 2) an Evidence-Strength model, which computed the distance between the sensory measurement and discrimination criterion, scaled according to sensory uncertainty. An accidental repetition of some sessions also allowed for the measurement of confidence agreement for identical pairs of stimuli. This N-pass analysis revealed that human observers were more consistent than their best-fitting model would predict, indicating there are still aspects of confidence that are not captured by our modelling. As such, we propose confidence agreement as a useful technique for computational studies of confidence. Taken together, these findings highlight the idiosyncratic nature of confidence computations for complex decision contexts and the need to consider different potential metrics and transformations in the confidence computation.

Modality Independent or Modality Specific? Common Computations Underlie Confidence Judgements in Visual and Auditory Decisions

“…The poor performance of the unscaled evidence strength class suggests that in mixed difficulty designs, observers modulate their confidence by the amount of sensory uncertainty in their internal representation of the stimulus in both the visual and auditory domain. The role of sensory uncertainty in adjusting response criteria has been demonstrated previously in the visual modality Denison et al, 2018;Locke et al, 2022;Qamar et al, 2013) but has not been demonstrated in the auditory modality. In addition to providing evidence that observers take sensory uncertainty into account across modalities, our study also provides insight into the extent to which these processes are consistent with Bayesian or non-Bayesian accounts of confidence.…”

“…), which is unbounded and allows for greater differences in confidence at different levels of sensory uncertainty (see bottom panel of Fig 1E ; Adler Locke et al, 2022).…”

Towards characterizing the canonical computations generating phenomenal experience

“…Such arbitration studies do exist, but they are relatively few (see e.g. (Adler & Ma, 2018b;Locke et al, 2022;Peters, Fesi, et al, 2017;Peters, Thesen, et al, 2017)). If we want to identify which model is (more) correct, we must design our stimuli and experiments specifically with such arbitration in mind.…”

Feeling lucky? Prospective and retrospective cues for sensorimotor confidence

“…Having different best-fitting models across participants is a common phenomenon in cue combination and multi-sensory research (e.g., 76,77), as well as perceptual confidence (35,46).…”

“…In order to incorporate a confidence judgment into a naturalistic reaching task more complex models need to be used. Bayesian approaches to confidence have gained popularity for handling non-binary decisions among other tasks, describing confidence as the posterior probability of being correct given the presence of noise (13,(29)(30)(31)(32)(33)(34)(35). Sensorimotor confidence is positioned uniquely to benefit from the Bayesian approach given that measurements in motor tasks are effector specific and recorded in a way that captures the noise naturally present in the system.…”