Response-Related Signals Increase Confidence But Not Metacognitive Performance

Abstract

Confidence judgements are a central tool in metacognition research. In a typical task, participants first perform perceptual (first-order) decisions and then rate their confidence in these decisions. The relationship between confidence and first-order accuracy is taken as a measure of metacognitive performance. Confidence is often assumed to stem from decision-monitoring processes alone, but processes that co-occur with the first-order decision may also play a role in confidence formation. In fact, some recent studies have revealed that directly manipulating motor regions in the brain, or the time of first-order decisions relative to second-order ones affects confidence judgements. This finding suggests that confidence could be informed by a readout of reaction times in addition to decision-monitoring processes. To test this possibility, we assessed the contribution of response-related signals to confidence and, in particular, to metacognitive performance (i.e., a measure of the adequacy of these confidence judgements). In human volunteers, we measured the effect of making an overt (vs. covert) decision, as well as the effect of pairing an action to the stimulus about which the first-order decision is made. Against our expectations, we found no differences in overall confidence or metacognitive performance when first-order responses were covert as opposed to overt. Further, actions paired to visual stimuli presented led to higher confidence ratings, but did not affect metacognitive performance. These results suggest that confidence ratings do not always incorporate motor information. 4 Significance statement To measure metacognition, or the ability to monitor one s own thoughts, experimental tasks often require human volunteers to, first, make a perceptual decision (first-order task") and, then, rate their confidence in their own decision (second-order task"). In this paradigm, both first and second-order information could, in principle, influence confidence judgements. But only the latter is truly metacognitive. To determine whether confidence is a valid metacognitive measure, we compared confidence ratings between two conditions: with overt responses, where participants provided both first-and second-order responses; and with covert responses where participants reported their confidence in a decision that they had not executed. Removing first-order decisions did not affect confidence, which validates confidence as an introspective measure.

Dual strategies in human confidence judgments

“…Previous studies have shown that certain physical image properties, such as stimulus contrast or the variability in stimulus elements, can impact confidence judgments (Boldt et al, 2017;de Gardelle & Mamassian, 2015;Herce Castanon et al, 2019;Spence et al, 2016;Zylberberg et al, 2014). Others have suggested that observers use non-normative strategies to estimate confidence (e.g., Filevich, Koß, & Faivre, 2020;Maniscalco, Peters, & Lau, 2016;Peters et al, 2017;Winter & Peters, 2021;Zylberberg et al, 2014), much like we do here. Our work differs from previous studies in that we provide both a Bayesian model and an alternative that explains deviations from optimality.…”

Confidence in visual motion discrimination is preserved in individuals with schizophrenia

“…The weaker link between trajectories and confidence in patients with schizophrenia may be related to slower and noisier motor behaviour, or to the tendency of people with schizophrenia to neglect relevant internal cues to control motor actions. 54 The fact that metacognitive performance was preserved in patients with schizophrenia despite a decreased link between confidence and trajectories suggests that senso rimotor signals may globally up or downregulate confidence estimates, with no influence on the calibration between confi dence and firstorder performance as reported recently. 54…”

“…54 The fact that metacognitive performance was preserved in patients with schizophrenia despite a decreased link between confidence and trajectories suggests that senso rimotor signals may globally up or downregulate confidence estimates, with no influence on the calibration between confi dence and firstorder performance as reported recently. 54…”

Transcranial Magnetic Stimulation-Induced Motor Cortex Activity Influences Visual Awareness Judgments

“…Our results provide confirmatory evidence that the findings reported by Siedlecka et al (2019) were a result of motor activity. Additionally, Filevich et al (2020) conclude that key presses in continuous report conditions served as an additional source of evidence available for both the metacognitive judgment and biased participants toward more liberal confidence responses. A similar conclusion comes from the research on perception of voluntary action where both active and passive movements produced overconfidence (Charles et al, 2020).…”

“…Although it cannot be ruled out that in this task the additional response itself or the visual information from the additional task cue were responsible for the observed effect, two other recent studies show an effect of action on confidence ratings. Faivre et al (2020) have shown that sensorimotor conflicts might decrease metacognitive efficiency (a quantitative measure of participants' level of metacognitive ability, given a certain level of 2AFC task performance), and Filevich et al (2020) have presented that continuous report paired to stimulus presentation leads to higher confidence ratings; however, there was no evidence that absolute confidence judgments or metacognitive efficiency varies with the presence or absence of overt responses.…”