Measuring Metacognition of Direct and Indirect Parameters of Voluntary Movement

Abstract

were supported by DFG, German Research Foundation (project number 222641018 -SFB/TTR 135). Data, code and pre-registration protocols are available at https://osf.io/kyhu7/ (Experiment 1) and https://osf.io/sy342/ (Experiment 2). The data discussed in this article were first published in "The Confidence Database", https://osf.io/s46pr/ (Rahnev et al., 2020). We have no known conflict of interest to disclose. AbstractWe can make exquisitely precise movements without the apparent need for conscious monitoring.But can we monitor the low-level movement parameters when prompted? And what are the mechanisms that allow us to monitor our movements? To answer these questions, we designed a semi-virtual ball throwing task. On each trial, participants first threw a virtual ball by moving their arm (with or without visual feedback, or replayed from a previous trial) and then made a twoalternative forced choice on the resulting ball trajectory. They then rated their confidence in their decision. We measured metacognitive efficiency using meta-d'/d' and compared it between different informational domains of the first-order task (motor, visuomotor or visual information alone), as well as between two different versions of the task based on different parameters of the movement: proximal (position of the arm) or distal (resulting trajectory of the ball thrown).We found that participants were able to monitor their performance based on distal motor information as well as when proximal information was available. Their metacognitive efficiency was also equally high in conditions with different sources of information available. The analysis of correlations across participants revealed an unexpected result: while metacognitive efficiency correlated between informational domains (which would indicate domain-generality of metacognition), it did not correlate across the different parameters of movement. We discuss possible sources of this discrepancy and argue that specific first-order task demands may play a crucial role in our metacognitive ability and should be considered when making inferences about domain-generality based on correlations.

Judgments of agency are affected by sensory noise without recruiting metacognitive processing

“…We suggest that confidence judgments about agency should be considered as the metacognitive level of an agency processing hierarchy, with agency judgments as explicit first-order judgments. This also brings agency in line with recent motor metacognition research that considers agency-like judgments such as decisions of which trajectory was caused by one's movement to be the first-order motor judgments, followed by metacognitive confidence ratings 43 .…”

I knew that! Response-based Outcome Predictions and Confidence Regulate Feedback Processing and Learning

“…Confidence has typically been studied in two-alternative choice tasks, and only rarely in relation to continuous outcomes 30,34,36,[74][75][76] . By reconceptualizing error detection as outcome prediction, our results shed new light on the well-supported claim that error monitoring and confidence are tightly intertwined 32,38,[77][78][79] and forge valuable links between research on performance monitoring 39,40,67 and on learning under uncertainty 1,3,58,80 .…”

Limited metacognitive access to one’s own facial expressions

“…In contrast to what we described for facial muscles, young, healthy participants have above-chance and precise metacognitive access to movements that are controlled by skeletal muscles 40 . Moreover, unlike the case of metacognition of facial expressions, measures of metacognitive performance in motor control do partially correlate with those from a visual task 41 . Speculatively, at least two factors may explain these discrepancies.…”

“…Participants make a movement (zero-order), then make a (first-order) judgment about it, and finally provide a (secondorder) subjective confidence rating. Examples of a zero-order task include moving a finger at a given pace 40 or throwing a ball to hit a target 41 . A different approach, which we took here, consists in operationalizing the metacognitive judgment not as confidence in accuracy of a binary choice, but instead as a judgment of performance [43][44][45] .…”