Confidence Leak in Perceptual Decision Making

More than the sum of its parts: investigating episodic memory as a multidimensional cognitive process

Why do we remember some events but forget others? Previous studies attempting to decode successful vs. unsuccessful brain states to investigate this question have met with limited success, potentially due, in part, to assessing episodic memory as a unidimensional process, despite evidence that multiple domains contribute to episodic encoding. Using a novel machine learning algorithm known as "transfer learning", we leveraged visual perception, sustained attention, and selective attention brain states to better predict episodic memory performance from trial-to-trial encoding electroencephalography (EEG) activity. We found that this multidimensional treatment of memory decoding improved prediction performance compared to traditional, unidimensional, methods, with each cognitive domain explaining unique variance in decoding of successful encoding-related neural activity. Importantly, this approach could be applied to cognitive domains outside of memory. Overall, this study provides critical insight into the underlying reasons why some events are remembered while others are not.

Manipulating Prior Beliefs Causally Induces Under- and Overconfidence

Humans differ vastly in the confidence they assign to decisions. Although such under- and overconfidence relate to fundamental life outcomes, a computational account specifying the underlying mechanisms is currently lacking. We propose that prior beliefs in the ability to perform a task explain confidence differences across participants and tasks, despite similar performance. In two perceptual decision-making experiments, we show that manipulating prior beliefs about performance during training causally influences confidence in healthy adults (N = 50 each; Experiment 1: 8 men, one nonbinary; Experiment 2: 5 men) during a test phase, despite unaffected objective performance. This is true when prior beliefs are induced via manipulated comparative feedback and via manipulated training-phase difficulty. Our results were accounted for within an accumulation-to-bound model, explicitly modeling prior beliefs on the basis of earlier task exposure. Decision confidence is quantified as the probability of being correct conditional on prior beliefs, causing under- or overconfidence. We provide a fundamental mechanistic insight into the computations underlying under- and overconfidence.

Human-like dissociations between confidence and accuracy in convolutional neural networks

Prior research has shown that manipulating stimulus energy by changing both stimulus contrast and variability results in confidence-accuracy dissociations in humans. Specifically, even when performance is matched, higher stimulus energy leads to higher confidence. The most common explanation for this effect is the positive evidence heuristic where confidence neglects evidence that disconfirms the choice. However, an alternative explanation is the signal-and-variance-increase hypothesis, according to which these dissociations arise from low-level changes in the separation and variance of perceptual representations. Because artificial neural networks lack built-in confidence heuristics, they can serve as a test for the necessity of confidence heuristics in explaining confidence-accuracy dissociations. Therefore, we tested whether confidence-accuracy dissociations induced by stimulus energy manipulations emerge naturally in convolutional neural networks (CNNs). We found that, across three different energy manipulations, CNNs produced confidence-accuracy dissociations similar to those found in humans. This effect was present for a range of CNN architectures from shallow 4-layer networks to very deep ones, such as VGG-19 and ResNet -50 pretrained on ImageNet. Further, we traced back the reason for the confidence-accuracy dissociations in all CNNs to the same signal-and-variance increase that has been proposed for humans: higher stimulus energy increased the separation and variance of the CNNs' internal representations leading to higher confidence even for matched accuracy. These findings cast doubt on the necessity of the positive evidence heuristic to explain human confidence and establish CNNs as promising models for adjudicating between low-level, stimulus-driven and high-level, cognitive explanations of human behavior.

The future of metacognition research: Balancing construct breadth with measurement rigor

Foundational work in the psychology of metacognition identified a distinction between metacognitive knowledge (stable beliefs about one's capacities) and metacognitive experiences (local evaluations of performance). More recently, the field has focused on developing tasks and metrics that seek to identify metacognitive capacities from momentary estimates of confidence in performance, and providing precise computational accounts of metacognitive failure. However, this notable progress in formalising models of metacognitive judgments may come at a cost of ignoring broader elements of the psychology of metacognition - such as how stable meta-knowledge is formed, how social cognition and metacognition interact, and how we evaluate affective states that do not have an obvious ground truth. We propose that construct breadth in metacognition research can be restored while maintaining rigour in measurement, and highlight promising avenues for expanding the scope of metacognition research. Such a research programme is well placed to recapture qualitative features of metacognitive knowledge and experience while maintaining the psychophysical rigor that characterises modern research on confidence and performance monitoring.

Expressions for Bayesian confidence of drift diffusion observers in fluctuating stimuli tasks

We introduce a new approach to modelling decision confidence, with the aim of enabling computationally cheap predictions while taking into account, and thereby exploiting, trial-by-trial variability in stochastically fluctuating stimuli. Using the framework of the drift diffusion model of decision making, along with time-dependent thresholds and the idea of a Bayesian confidence readout, we derive expressions for the probability distribution over confidence reports. In line with current models of confidence, the derivations allow for the accumulation of "pipeline" evidence that has been received but not processed by the time of response, the effect of drift rate variability, and metacognitive noise. The expressions are valid for stimuli that change over the course of a trial with normally-distributed fluctuations in the evidence they provide. A number of approximations are made to arrive at the final expressions, and we test all approximations via simulation. The derived expressions contain only a small number of standard functions, and require evaluating only once per trial, making trial-by-trial modelling of confidence data in stochastically fluctuating stimuli tasks more feasible. We conclude by using the expressions to gain insight into the confidence of optimal observers, and empirically observed patterns.

Serial dependence in timing at the perceptual level being modulated by working memory

Recent experiences bias the perception of following stimuli, as has been verified in various kinds of experiments in visual perception. This phenomenon, known as serial dependence, may reflect mechanisms to maintain perceptual stability. In the current study, we examined several key properties of serial dependence in temporal perception. Firstly, we examined the source of the serial dependence effect in temporal perception. We found that perception without motor reproduction is sufficient to induce the sequential effect; motor reproduction caused a stronger effect and is achieved by biasing the perception of the future target duration rather than directly influencing the subsequent movement. Secondly, we ask how working memory influences serial dependence in a temporal reproduction task. By varying the delay time between standard duration and the reproduction, we showed that the strength of serial dependence is enhanced as the delay increased. Those features of serial dependence are consistent with what has been observed in visual perceptual tasks, for example, orientation perception or location perception. The similarities between the visual and the timing tasks may suggest a similar neural coding mechanism of magnitude between the visual stimuli and the duration.

Boundary updating as a source of history effect on decision uncertainty

When sorting a sequence of stimuli into binary classes, current choices are often negatively correlated with recent stimulus history. This phenomenon-dubbed the repulsive bias-can be explained by boundary updating, a process of shifting the class boundary to previous stimuli. This explanation implies that recent stimulus history can also influence "decision uncertainty," the probability of making incorrect decisions, because it depends on the location of the boundary. However, there have been no previous efforts to elucidate the impact of previous stimulus history on decision uncertainty. Here, from the boundary-updating process that accounts for the repulsive bias, we derived a prediction that decision uncertainty increases as current choices become more congruent with previous stimuli. We confirmed this prediction in behavioral, physiological, and neural correlates of decision uncertainty. Our work demonstrates that boundary updating offers a principled account of how previous stimulus history concurrently relates to choice bias and decision uncertainty.

The impact of feedback on metacognition: Enhancing in easy tasks, impeding in difficult ones

Metacognition refers to the ability to monitor and introspect upon cognitive performance. Abundant research suggests that individual metacognition is easily affected by feedback in daily life, but how feedback affects metacognition in perceptual decision-making remains unclear. Here we investigated how trial-by-trial feedback shapes perceptual metacognition in two experiments with either high (n = 82) or low difficulty (n = 90). Participants were randomly divided into a feedback group in which participants received trial-by-trial performance feedback or a no-feedback group. Results showed that, in the high-difficulty task, participants in the feedback group revealed inferior metacognitive performance than the no-feedback group, manifested as decreased metacognitive efficiency while controlling for performance sensitivity. In the low-difficulty task, however, participants in the feedback group had higher metacognitive efficiency than the no-feedback group. The distinct patterns of findings in the two experiments indicate that whether feedback promotes or impedes metacognition is adjusted by task difficulty.

Neural and computational underpinnings of biased confidence in human reinforcement learning

While navigating a fundamentally uncertain world, humans and animals constantly evaluate the probability of their decisions, actions or statements being correct. When explicitly elicited, these confidence estimates typically correlates positively with neural activity in a ventromedial-prefrontal (VMPFC) network and negatively in a dorsolateral and dorsomedial prefrontal network. Here, combining fMRI with a reinforcement-learning paradigm, we leverage the fact that humans are more confident in their choices when seeking gains than avoiding losses to reveal a functional dissociation: whereas the dorsal prefrontal network correlates negatively with a condition-specific confidence signal, the VMPFC network positively encodes task-wide confidence signal incorporating the valence-induced bias. Challenging dominant neuro-computational models, we found that decision-related VMPFC activity better correlates with confidence than with option-values inferred from reinforcement-learning models. Altogether, these results identify the VMPFC as a key node in the neuro-computational architecture that builds global feeling-of-confidence signals from latent decision variables and contextual biases during reinforcement-learning.

Prior information differentially affects discrimination decisions and subjective confidence reports

According to Bayesian models, both decisions and confidence are based on the same precision-weighted integration of prior expectations ("priors") and incoming information ("likelihoods"). This assumes that priors are integrated optimally and equally in decisions and confidence, which has not been tested. In three experiments, we quantify how priors inform decisions and confidence. With a dual-decision task we create pairs of conditions that are matched in posterior information, but differ on whether the prior or likelihood is more informative. We find that priors are underweighted in discrimination decisions, but are less underweighted in confidence about those decisions, and this is not due to differences in processing time. The same patterns remain with exogenous probabilistic cues as priors. With a Bayesian model we quantify the weighting parameters for the prior at both levels, and find converging evidence that priors are more optimally used in explicit confidence, even when underused in decisions.

Task feedback suggests a post-perceptual component to serial dependence

Decisions across a range of perceptual tasks are biased toward past stimuli. Such serial dependence is thought to be an adaptive low-level mechanism that promotes perceptual stability across time. However, recent studies suggest post-perceptual mechanisms may also contribute to serially biased responses, calling into question a single locus of serial dependence and the nature of integration of past and present sensory inputs. We measured serial dependence in the context of a three-dimensional (3D) motion perception task where uncertainty in the sensory information varied substantially from trial to trial. We found that serial dependence varied with stimulus properties that impact sensory uncertainty on the current trial. Reduced stimulus contrast was associated with an increased bias toward the stimulus direction of the previous trial. Critically, performance feedback, which reduced sensory uncertainty, abolished serial dependence. These results provide clear evidence for a post-perceptual locus of serial dependence in 3D motion perception and support the role of serial dependence as a response strategy in the face of substantial sensory uncertainty.

Using serial dependence to predict confidence across observers and cognitive domains

Our perceptual system appears hardwired to exploit regularities of input features across space and time in seemingly stable environments. This can lead to serial dependence effects whereby recent perceptual representations bias current perception. Serial dependence has also been demonstrated for more abstract representations, such as perceptual confidence. Here, we ask whether temporal patterns in the generation of confidence judgments across trials generalize across observers and different cognitive domains. Data from the Confidence Database across perceptual, memory, and cognitive paradigms was reanalyzed. Machine learning classifiers were used to predict the confidence on the current trial based on the history of confidence judgments on the previous trials. Cross-observer and cross-domain decoding results showed that a model trained to predict confidence in the perceptual domain generalized across observers to predict confidence across the different cognitive domains. The recent history of confidence was the most critical factor. The history of accuracy or Type 1 reaction time alone, or in combination with confidence, did not improve the prediction of the current confidence. We also observed that confidence predictions generalized across correct and incorrect trials, indicating that serial dependence effects in confidence generation are uncoupled to metacognition (i.e., how we evaluate the precision of our own behavior). We discuss the ramifications of these findings for the ongoing debate on domain-generality versus domain-specificity of metacognition.

Feeling lucky? prospective and retrospective cues for sensorimotor confidence

On a daily basis, humans interact with the outside world using judgments of sensorimotor confidence, constantly evaluating our actions for success. We ask, what sensory and motor-execution cues are used in making these judgements and when are they available? Two sources of temporally distinct information are prospective cues, available prior to the action (e.g., knowledge of motor noise and past performance), and retrospective cues specific to the action itself (e.g., proprioceptive measurements). We investigated the use of these two cues in two tasks, a secondary motor-awareness task and a main task in which participants reached toward a visual target with an unseen hand and then made a continuous judgment of confidence about the success of the reach. Confidence was reported by setting the size of a circle centered on the reach-target location, where a larger circle reflects lower confidence. Points were awarded if the confidence circle enclosed the true endpoint, with fewer points returned for larger circles. This incentivized accurate reaches and attentive reporting to maximize the score. We compared three Bayesian-inference models of sensorimotor confidence based on either prospective cues, retrospective cues, or both sources of information to maximize expected gain (i.e., an ideal-performance model). Our findings showed two distinct strategies: participants either performed as ideal observers, using both prospective and retrospective cues to make the confidence judgment, or relied solely on prospective information, ignoring retrospective cues. Thus, participants can make use of retrospective cues, evidenced by the behavior observed in our motor-awareness task, but these cues are not always included in the computation of sensorimotor confidence.

Differences in the detail: Metacognition is better for seen than sensed changes to visual scenes

Metacognition is the process by which we know what we know. Knowing has both declarative and sensed components. Differences exist in the information that moves to our conscious awareness and how it is synthesised with existing knowledge. The current study measured metacognition by extending a visual change detection paradigm that promoted explicit or implicit detection by either a local or global manipulation of a scene. A within-subjects design was used to investigate how 91 participants detected change and made metacognitive judgements. Cognitive modelling, based on confidence judgements, estimated the relative contributions of discrete and continuous cognitive processes to change detection, and to metacognition. Metacognition was sensitive to both the discrete and continuous processes underlying change detection, but was more sensitive to the discrete process. These results demonstrate that metacognition attunes confidence differentially to explicit and implicit processes, and support direct-access theories for discrete processing and meta-representation theories for continuous processing.

Cross-modal metacognition: Visual and tactile confidence share a common scale

Humans can judge the quality of their perceptual decisions-an ability known as perceptual confidence. Previous work suggested that confidence can be evaluated on an abstract scale that can be sensory modality-independent or even domain-general. However, evidence is still scarce on whether confidence judgments can be directly made across visual and tactile decisions. Here, we investigated in a sample of 56 adults whether visual and tactile confidence share a common scale by measuring visual contrast and vibrotactile discrimination thresholds in a confidence-forced choice paradigm. Confidence judgments were made about the correctness of the perceptual decision between two trials involving either the same or different modalities. To estimate confidence efficiency, we compared discrimination thresholds obtained from all trials to those from trials judged to be relatively more confident. We found evidence for metaperception because higher confidence was associated with better perceptual performance in both modalities. Importantly, participants were able to judge their confidence across modalities without any costs in metaperceptual sensitivity and only minor changes in response times compared to unimodal confidence judgments. In addition, we were able to predict cross-modal confidence well from unimodal judgments. In conclusion, our findings show that perceptual confidence is computed on an abstract scale and that it can assess the quality of our decisions across sensory modalities.

A shared brain system forming confidence judgment across cognitive domains

Confidence is typically defined as a subjective judgment about whether a decision is right. Decisions are based on sources of information that come from various cognitive domains and are processed in different brain systems. An unsettled question is whether the brain computes confidence in a similar manner whatever the domain or in a manner that would be idiosyncratic to each domain. To address this issue, human participants performed two tasks probing confidence in decisions made about the same material (history and geography statements), but based on different cognitive processes: semantic memory for deciding whether the statement was true or false, and duration perception for deciding whether the statement display was long or short. At the behavioral level, we found that the same factors (difficulty, accuracy, response time, and confidence in the preceding decision) predicted confidence judgments in both tasks. At the neural level, we observed using functional magnetic resonance imaging that confidence judgments in both tasks were associated to activity in the same brain regions: positively in the ventromedial prefrontal cortex and negatively in a prefronto-parietal network. Together, these findings suggest the existence of a shared brain system that generates confidence judgments in a similar manner across cognitive domains.

Time-consciousness in computational phenomenology: a temporal analysis of active inference

Time plays a significant role in science and everyday life. Despite being experienced as a continuous flow, computational models of consciousness are typically restricted to a sequential temporal structure. This difference poses a serious challenge for computational phenomenology-a novel field combining phenomenology and computational modelling. By analysing the temporal structure of the active inference framework, we show that an integrated continuity of time can be achieved by merging Husserlian temporality with a sequential order of time. We also show that a Markov blanket of the present moment integrates past and future moments of both subjective temporality and objective time in an asynchronous manner. By applying the integrated continuity, it is clear that active inference makes use of both subjective temporality and objective time in an integrated fashion. We conclude that active inference, on a temporal note, qualifies as a computational model for phenomenological investigations.

Confidence reflects a noisy decision reliability estimate

Decisions vary in difficulty. Humans know this and typically report more confidence in easy than in difficult decisions. However, confidence reports do not perfectly track decision accuracy, but also reflect response biases and difficulty misjudgements. To isolate the quality of confidence reports, we developed a model of the decision-making process underlying choice-confidence data. In this model, confidence reflects a subject's estimate of the reliability of their decision. The quality of this estimate is limited by the subject's uncertainty about the uncertainty of the variable that informs their decision ('meta-uncertainty'). This model provides an accurate account of choice-confidence data across a broad range of perceptual and cognitive tasks, investigated in six previous studies. We find meta-uncertainty varies across subjects, is stable over time, generalizes across some domains and can be manipulated experimentally. The model offers a parsimonious explanation for the computational processes that underlie and constrain the sense of confidence.

Consensus Goals in the Field of Visual Metacognition

Despite the tangible progress in psychological and cognitive sciences over the last several years, these disciplines still trail other more mature sciences in identifying the most important questions that need to be solved. Reaching such consensus could lead to greater synergy across different laboratories, faster progress, and increased focus on solving important problems rather than pursuing isolated, niche efforts. Here, 26 researchers from the field of visual metacognition reached consensus on four long-term and two medium-term common goals. We describe the process that we followed, the goals themselves, and our plans for accomplishing these goals. If this effort proves successful within the next few years, such consensus building around common goals could be adopted more widely in psychological science.

Learning with certainty in childhood

Learners use certainty to guide learning. They maintain existing beliefs when certain, but seek further information when they feel uninformed. Here, we review developmental evidence that this metacognitive strategy does not require reportable processing. Uncertainty prompts nonverbal human infants and nonhuman animals to engage in strategies like seeking help, searching for additional information, or opting out. Certainty directs children's attention and active learning strategies and provides a common metric for comparing and integrating conflicting beliefs across people. We conclude that certainty is a continuous, domain-general signal of belief quality even early in life.

Time-sensitive prefrontal involvement in associating confidence with task performance illustrates metacognitive introspection in monkeys

Metacognition refers to the ability to be aware of one's own cognition. Ample evidence indicates that metacognition in the human primate is highly dissociable from cognition, specialized across domains, and subserved by distinct neural substrates. However, these aspects remain relatively understudied in macaque monkeys. In the present study, we investigated the functionality of macaque metacognition by combining a confidence proxy, hierarchical Bayesian meta-d' computational modelling, and a single-pulse transcranial magnetic stimulation technique. We found that Brodmann area 46d (BA46d) played a critical role in supporting metacognition independent of task performance; we also found that the critical role of this region in meta-calculation was time-sensitive. Additionally, we report that macaque metacognition is highly domain-specific with respect to memory and perception decisions. These findings carry implications for our understanding of metacognitive introspection within the primate lineage.

Human perceptual and metacognitive decision-making rely on distinct brain networks

Perceptual decisions depend on the ability to exploit available sensory information in order to select the most adaptive option from a set of alternatives. Such decisions depend on the perceptual sensitivity of the organism, which is generally accompanied by a corresponding level of certainty about the choice made. Here, by use of corticocortical paired associative transcranial magnetic stimulation protocol (ccPAS) aimed at inducing plastic changes, we shaped perceptual sensitivity and metacognitive ability in a motion discrimination task depending on the targeted network, demonstrating their functional dissociation. Neurostimulation aimed at boosting V5/MT+-to-V1/V2 back-projections enhanced motion sensitivity without impacting metacognition, whereas boosting IPS/LIP-to-V1/V2 back-projections increased metacognitive efficiency without impacting motion sensitivity. This double-dissociation provides causal evidence of distinct networks for perceptual sensitivity and metacognitive ability in humans.

Suprathreshold perceptual decisions constrain models of confidence

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.

The feeling of confidence in what we perceive can influence our future behaviour and learning. Understanding how the brain computes confidence is an important goal of researchers. As such, researchers have identified a host of potential models. Yet, rarely are a wide range of models tested against each other to find those that best predict choice behaviour. Our study had human participants compare their confidence for pairs of easy perceptual decisions, reporting if they had higher confidence in the first or second decision. We tested twelve models, covering all three types of models proposed in previous studies, finding strong support for two models. The winning Heuristic model combines all three factors affecting choice uncertainty with an idiosyncratic weighting to compute confidence. The other winning model uses a transformation where the strength of the sensory signal is scaled according to sensory uncertainty. We also assessed the agreement of confidence reports in identical decision scenarios. Humans had higher agreement than almost all model predictions. We propose using confidence agreement intentionally as a second performance benchmark of model fit.

Distinct Generation of Subjective Vividness and Confidence during Naturalistic Memory Retrieval in Angular Gyrus

Our subjective experience of remembering guides and monitors the reconstruction of past and simulation of the future, which enables us to identify mistakes and adjust our behavior accordingly. However, it remains incompletely understood what underlies the process of subjective mnemonic experience. Here, we combined behavior, repetitive TMS, and functional neuroimaging to probe whether vividness and confidence are generated differently during retrieval. We found that preretrieval repetitive TMS targeting the left angular gyrus (AnG) selectively attenuated the vividness efficiency compared with control stimulation while keeping metacognitive efficiency and objective memory accuracy unaffected. Using trialwise data, we showed that AnG stimulation altered the mediating role of vividness in confidence in the accuracy of memory judgment. Moreover, resting-state functional connectivity of hippocampus and AnG was specifically associated with vividness efficiency, but not metacognitive efficiency across individuals. Together, these results identify the causal involvement of AnG in gauging the vividness, but not the confidence, of memory, thereby suggesting a differentiation account of conscious assessment of memory by functionally and anatomically dissociating the monitoring of vividness from confidence.

The Impact of Feedback on Perceptual Decision-Making and Metacognition: Reduction in Bias but No Change in Sensitivity

It is widely believed that feedback improves behavior, but the mechanisms behind this improvement remain unclear. Different theories postulate that feedback has either a direct effect on performance through automatic reinforcement mechanisms or only an indirect effect mediated by a deliberate change in strategy. To adjudicate between these competing accounts, we performed two large experiments on human adults (total N = 518); approximately half the participants received trial-by-trial feedback on a perceptual task, whereas the other half did not receive any feedback. We found that feedback had no effect on either perceptual or metacognitive sensitivity even after 7 days of training. On the other hand, feedback significantly affected participants' response strategies by reducing response bias and improving confidence calibration. These results suggest that the beneficial effects of feedback stem from allowing people to adjust their strategies for performing the task and not from direct reinforcement mechanisms, at least in the domain of perception.

I remember it like it was yesterday: Age-related differences in the subjective experience of remembering

It has been frequently described that older adults subjectively report the vividness of their memories as being as high, or even higher, than young adults, despite poorer objective memory performance. Here, we review studies that examined age-related differences in the subjective experience of memory vividness. By examining vividness calibration and resolution, studies using different types of approaches converge to suggest that older adults overestimate the intensity of their vividness ratings relative to young adults, and that they rely on retrieved memory details to a lesser extent to judge vividness. We discuss potential mechanisms underlying these observations. Inflation of memory vividness with regard to the richness of memory content may stem from age-differences in vividness criterion or scale interpretation and psycho-social factors. The reduced reliance on episodic memory details in older adults may stem from age-related differences in how they monitor these details to make their vividness ratings. Considered together, these findings emphasize the importance of examining age-differences in memory vividness using different analytical methods and they provide valuable evidence that the subjective experience of remembering is more than the reactivation of memory content. In this vein, we recommend that future studies explore the links between memory vividness and other subjective memory scales (e.g., ratings of details or memory confidence) in healthy aging and/or other populations, as it could be used as a window to better characterize the cognitive processes that underpin the subjective assessment of the quality of recollected events.

Prospective Metamemory Monitoring of Episodic Visual Memory in Community-Dwelling Older Adults with Subjective Cognitive Decline and Mild Cognitive Impairment

OBJECTIVE

Metamemory tasks have been utilized to investigate anosognosia in older adults with dementia, though previous research has not systematically compared memory self-awareness in prodromal dementia groups. This represents an important oversight given that remedial and interventional efforts may be most beneficial before individuals' transition to clinical dementia. We examine differences in memory self-awareness and memory self-monitoring between cognitively healthy elderly controls and prodromal dementia groups.

METHODS

Participants with subjective cognitive decline despite intact objective neuropsychological functioning (SCD; n = 82), amnestic mild cognitive impairment (aMCI; n = 18), nonamnestic mild cognitive impairment (naMCI; n = 38), and normal cognitive functioning (HC; n = 120) were recruited from the Einstein Aging Study for a cross-sectional study. Participants completed an experimental visual memory-based global metamemory prediction task and subjective assessments of memory/cognition and self-awareness.

RESULTS

While, relative to HC, memory self-awareness and memory self-monitoring were preserved for delayed memory performance in SCD and aMCI, these processes were impaired in naMCI. Furthermore, results suggest that poor metamemory accuracy captured by our experimental task can be generalized to everyday memory problems.

CONCLUSIONS

Within the framework of the Cognitive Awareness Model, our findings provide preliminary evidence that poor memory self-awareness/self-monitoring in naMCI may reflect an executive or primary anosognosia, with implications for tailored rehabilitative interventions.

A robust confidence-accuracy dissociation via criterion attraction

Many studies have shown that confidence and accuracy can be dissociated in a variety of tasks. However, most of these dissociations involve small effect sizes, occur only in a subset of participants, and include a reaction time (RT) confound. Here, I develop a new method for inducing confidence-accuracy dissociations that overcomes these limitations. The method uses an external noise manipulation and relies on the phenomenon of criterion attraction where criteria for different tasks become attracted to each other. Subjects judged the identity of stimuli generated with either low or high external noise. The results showed that the two conditions were matched on accuracy and RT but produced a large difference in confidence (effect appeared for 25 of 26 participants, effect size: Cohen's d = 1.9). Computational modeling confirmed that these results are consistent with a mechanism of criterion attraction. These findings establish a new method for creating conditions with large differences in confidence without differences in accuracy or RT. Unlike many previous studies, however, the current method does not lead to differences in subjective experience and instead produces robust confidence-accuracy dissociations by exploiting limitations in post-perceptual, cognitive processes.

Contested science: Individuals with higher metacognitive insight into interpretation of evidence are less likely to polarize

Societal polarization over contested science has increased in recent years. To explain this development, political, sociological, and psychological research has identified societal macro-phenomena as well as cognitive micro-level factors that explain how citizens reason about the science. Here we take a radically different perspective, and highlight the effects of metacognition: How citizens reason about their own reasoning. Leveraging methods from Signal Detection Theory, we investigated the importance of metacognitive insight for polarization for the heavily contested topic of climate change, and the less heavily contested topic of nanotechnology. We found that, for climate change (but not for nanotechnology), higher insight into the accuracy of own interpretations of the available scientific evidence related to a lower likelihood of polarization over the science. This finding held irrespective of the direction of the scientific evidence (endorsing or rejecting anthropogenicity of climate change). Furthermore, the polarizing effect of scientific evidence could be traced back to higher metacognitive insight fostering belief-updating in the direction of the evidence at the expense of own, prior beliefs. By demonstrating how metacognition links to polarization, the present research adds to our understanding of the drivers of societal polarization over science.

Atypical meta-memory evaluation strategy in schizophrenia patients

Background

Previous research has reported that patients with schizophrenia would regard false memories with higher confidence, and this meta-memory deficit was suggested as a neurocognitive marker of schizophrenia. However, how schizophrenia patients determine their memory decision confidence has received scant consideration. This study, therefore, aimed to characterize the extent to which meta-memory evaluation strategy differs between schizophrenia patients and healthy individuals, and how such difference contributes to the patients' meta-memory performance.

Methods

27 schizophrenia patients and 28 matched healthy controls performed a temporal-order judgement (TOJ) task, in which they judged which movie frame occurred earlier in an encoded video, and then made retrospective confidence rating. Mixed effect regression models were performed to assess the between-group metacognitive evaluation strategy difference and its relationship to clinical symptoms.

Results

Compared to the control group, the patients' confidence ratings were correlated more with the recent confidence history and less with the TOJ-related evidence. The degree of dependence on recent history of confidence was negatively correlated with the severity of positive symptoms. Furthermore, by controlling for the first-order TOJ performance, we observed that the patients discriminated correct memory decisions from the incorrect ones as accurately as the controls.

Conclusion

The present investigation revealed that schizophrenia patients tend to use more heuristics in making meta-memory evaluations, and such atypical strategy is related to their clinical symptoms. This study provides new insights into how schizophrenia patients perform meta-memory processes. Future research could consider examining such metacognitive deficits in light of other cognitive domains in psychosis.

Examining the robustness of the relationship between metacognitive efficiency and metacognitive bias

We recently found a positive relationship between estimates of metacognitive efficiency and metacognitive bias. However, this relationship was only examined on a within-subject level and required binarizing the confidence scale, a technique that introduces methodological difficulties. Here we examined the robustness of the positive relationship between estimates of metacognitive efficiency and metacognitive bias by conducting two different types of analyses. First, we developed a new within-subject analysis technique where the original n-point confidence scale is transformed into two different (n-1)-point scales in a way that mimics a naturalistic change in confidence. Second, we examined the across-subject correlation between metacognitive efficiency and metacognitive bias. Importantly, for both types of analyses, we not only established the direction of the effect but also computed effect sizes. We applied both techniques to the data from three tasks from the Confidence Database (N > 400 in each). We found that both approaches revealed a small to medium positive relationship between metacognitive efficiency and metacognitive bias. These results demonstrate that the positive relationship between metacognitive efficiency and metacognitive bias is robust across several analysis techniques and datasets, and have important implications for future research.

Variance misperception under skewed empirical noise statistics explains overconfidence in the visual periphery

Perceptual confidence typically corresponds to accuracy. However, observers can be overconfident relative to accuracy, termed "subjective inflation." Inflation is stronger in the visual periphery relative to central vision, especially under conditions of peripheral inattention. Previous literature suggests inflation stems from errors in estimating noise (i.e., "variance misperception"). However, despite previous Bayesian hypotheses about metacognitive noise estimation, no work has systematically explored how noise estimation may critically depend on empirical noise statistics, which may differ across the visual field, with central noise distributed symmetrically but peripheral noise positively skewed. Here, we examined central and peripheral vision predictions from five Bayesian-inspired noise-estimation algorithms under varying usage of noise priors, including effects of attention. Models that failed to optimally estimate noise exhibited peripheral inflation, but only models that explicitly used peripheral noise priors-but used them incorrectly-showed increasing peripheral inflation under increasing peripheral inattention. Further, only one model successfully captured previous empirical results, which showed a selective increase in confidence in incorrect responses under performance reductions due to inattention accompanied by no change in confidence in correct responses; this was the model that implemented Bayesian estimation of peripheral noise, but using an (incorrect) symmetric rather than the correct positively skewed peripheral noise prior. Our findings explain peripheral inflation, especially under inattention, and suggest future experiments that might reveal the noise expectations used by the visual metacognitive system.

Response Bias Reflects Individual Differences in Sensory Encoding

Humans exhibit substantial biases in their decision making even in simple two-choice tasks, but the origin of these biases remains unclear. I hypothesized that one source of bias could be individual differences in sensory encoding. Specifically, if one stimulus category gives rise to an internal-evidence distribution with higher variability, then responses should optimally be biased against that stimulus category. Therefore, response bias may reflect a previously unappreciated subject-to-subject difference in the variance of the internal-evidence distributions. I tested this possibility by analyzing data from three different two-choice tasks (ns = 443, 443, and 498). For all three tasks, response bias moved in the direction of the optimal criterion determined by each subject's idiosyncratic internal-evidence variability. These results demonstrate that seemingly random variations in response bias can be driven by individual differences in sensory encoding and are thus partly explained by normative strategies.

Multitasking costs on metacognition in a triple-task paradigm

Multitasking situations, such as using one's phone while driving, are increasingly common in everyday life. Experimental psychology has long documented the costs of multitasking on task performance; however, little is known of the effects it has on the metacognitive processes that monitor such performance. The present study is a step toward filling this void by combining psychophysical procedures with complex multitasking. We devised a multimodal paradigm in which participants performed a sensorimotor tracking task, a visual discrimination task, and an auditory 2-back working memory task, either separately or concurrently, while also evaluating their task performance every ~15 s. Our main finding is that multitasking decreased participants' awareness of their performance (metacognitive sensitivity) for all three tasks. Importantly, this result was independent of the multitasking cost on task performance, and could not be attributed to confidence leak, psychological refractory period, or recency effects on self-evaluations. We discuss the implications of this finding for both metacognition and multitasking research.

Confidence and central tendency in perceptual judgment

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

Global visual confidence

Visual confidence is the observers’ estimate of their precision in one single perceptual decision. Ultimately, however, observers often need to judge their confidence over a task in general rather than merely on one single decision. Here, we measured the global confidence acquired across multiple perceptual decisions. Participants performed a dual task on two series of oriented stimuli. The perceptual task was an orientation-discrimination judgment. The metacognitive task was a global confidence judgment: observers chose the series for which they felt they had performed better in the perceptual task. We found that choice accuracy in global confidence judgments improved as the number of items in the series increased, regardless of whether the global confidence judgment was made before (prospective) or after (retrospective) the perceptual decisions. This result is evidence that global confidence judgment was based on an integration of confidence information across multiple perceptual decisions rather than on a single one. Furthermore, we found a tendency for global confidence choices to be influenced by response times, and more so for recent perceptual decisions than earlier ones in the series of stimuli. Using model comparison, we found that global confidence is well described as a combination of noisy estimates of sensory evidence and position-weighted response-time evidence. In summary, humans can integrate information across multiple decisions to estimate global confidence, but this integration is not optimal, in particular because of biases in the use of response-time information.

Overlapping and unique neural circuits are activated during perceptual decision making and confidence

The period of making a perceptual decision is often followed by a period of rating confidence where one evaluates the likely accuracy of the initial decision. However, it remains unclear whether the same or different neural circuits are engaged during periods of perceptual decision making and confidence report. To address this question, we conducted two functional MRI experiments in which we dissociated the periods related to perceptual decision making and confidence report by either separating their respective regressors or asking for confidence ratings only in the second half of the experiment. We found that perceptual decision making and confidence reports gave rise to activations in large and mostly overlapping brain circuits including frontal, parietal, posterior, and cingulate regions with the results being remarkably consistent across the two experiments. Further, the confidence report period activated a number of unique regions, whereas only early sensory areas were activated for the decision period across the two experiments. We discuss the possible reasons for this overlap and explore their implications about theories of perceptual decision making and visual metacognition.

Sources of Metacognitive Inefficiency

Confidence judgments are typically less informative about one's accuracy than they could be; a phenomenon we call metacognitive inefficiency. We review the existence of different sources of metacognitive inefficiency and classify them into four categories based on whether the corruption is due to: (i) systematic or nonsystematic influences, and (ii) the input to or the computation of the metacognitive system. Critically, the existence of different sources of metacognitive inefficiency provides an alternative explanation for behavioral findings typically interpreted as evidence for domain-specific (and against domain-general) metacognitive systems. We argue that, contrary to the dominant assumption in the field, metacognitive failures are not monolithic and suggest that understanding the sources of metacognitive inefficiency should be a primary goal of the science of metacognition.

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

The influence of non-visual information on visual awareness judgments has recently gained substantial interest. Using single-pulse transcranial magnetic stimulation (TMS), we investigate the potential contribution of evidence from the motor system to judgment of visual awareness. We hypothesized that TMS-induced activity in the primary motor cortex (M1) would increase reported visual awareness as compared to the control condition. Additionally, we investigated whether TMS-induced motor-evoked potential (MEP) could measure accumulated evidence for stimulus perception. Following stimulus presentation and TMS, participants first rated their visual awareness verbally using the Perceptual Awareness Scale (PAS), after which they responded manually to a Gabor orientation identification task. Delivering TMS to M1 resulted in higher average awareness ratings as compared to the control condition, in both correct and incorrect identification task response trials, when the hand with which participants responded was contralateral to the stimulated hemisphere (TMS-response-congruent trials). This effect was accompanied by longer PAS response times (RTs), irrespective of the congruence between TMS and identification response. Moreover, longer identification RTs were observed in TMS-response-congruent trials in the M1 condition as compared to the control condition. Additionally, the amplitudes of MEPs were related to the awareness ratings when response congruence was taken into account. We argue that MEP can serve as an indirect measure of evidence accumulated for stimulus perception and that longer PAS RTs and higher amplitudes of MEPs in the M1 condition reflect integration of additional evidence with visual awareness judgment. In conclusion, we advocate that motor activity influences perceptual awareness judgments.

Confounding in Studies on Metacognition: A Preliminary Causal Analysis Framework

By definition, metacognitive processes may monitor or regulate various stages of first-order processing. By combining causal analysis with hypotheses expressed by other authors we derive the theoretical and methodological consequences of this special relation between metacognition and the underlying processes. In particular, we prove that because multiple processing stages may be monitored or regulated and because metacognition may form latent feedback loops, (1) without strong additional causal assumptions, typical measures of metacognitive monitoring or regulation are confounded; (2) without strong additional causal assumptions, typical methods of controlling for first-order task performance (i.e., calibration, staircase, including first-order task performance in a regression analysis, or analyzing correct and incorrect trials separately) not only do not deconfound measures of metacognition but may even introduce bias; (3) that the first two problems cannot be solved by using simple models of decision-making derived from Signal Detection Theory. We conclude the paper by advocating robust methods of discovering properties of latent mechanisms.

Confidence as a Priority Signal

When dealing with multiple tasks, we must establish the order in which to tackle them. In multiple experiments, including a preregistered replication (Ns = 16-105), we found that confidence, or the subjective accuracy of decisions, acts as a priority signal, both when ordering responses about tasks already completed or ordering tasks yet to be completed. Specifically, when participants categorized perceptual stimuli along two dimensions, they tended to first give the decision associated with higher confidence. When participants selected which of two tasks they wanted to perform first, they were slightly biased toward the task associated with higher confidence. This finding extends to nonperceptual decisions (mental calculation) and cannot be reduced to effects of task difficulty, response accuracy, response availability, or implicit demands. Our results thus support the role of confidence as a priority signal, thereby suggesting a new way in which it may regulate human behavior.

The suboptimality of perceptual decision making with multiple alternatives

It is becoming widely appreciated that human perceptual decision making is suboptimal but the nature and origins of this suboptimality remain poorly understood. Most past research has employed tasks with two stimulus categories, but such designs cannot fully capture the limitations inherent in naturalistic perceptual decisions where choices are rarely between only two alternatives. We conduct four experiments with tasks involving multiple alternatives and use computational modeling to determine the decision-level representation on which the perceptual decisions are based. The results from all four experiments point to the existence of robust suboptimality such that most of the information in the sensory representation is lost during the transformation to a decision-level representation. These results reveal severe limits in the quality of decision-level representations for multiple alternatives and have strong implications about perceptual decision making in naturalistic settings.

The nature of metacognitive inefficiency in perceptual decision making

Humans have the metacognitive ability to judge the accuracy of their own decisions via confidence ratings. A substantial body of research has demonstrated that human metacognition is fallible but it remains unclear how metacognitive inefficiency should be incorporated into a mechanistic model of confidence generation. Here we show that, contrary to what is typically assumed, metacognitive inefficiency depends on the level of confidence. We found that, across 5 different data sets and 4 different measures of metacognition, metacognitive ability decreased with higher confidence ratings. To understand the nature of this effect, we collected a large dataset of 20 subjects completing 2,800 trials each and providing confidence ratings on a continuous scale. The results demonstrated a robustly nonlinear zROC curve with downward curvature, despite a decades-old assumption of linearity. This pattern of results was reproduced by a new mechanistic model of confidence generation, which assumes the existence of lognormally distributed metacognitive noise. The model outperformed competing models either lacking metacognitive noise altogether or featuring Gaussian metacognitive noise. Further, the model could generate a measure of metacognitive ability which was independent of confidence levels. These findings establish an empirically validated model of confidence generation, have significant implications about measures of metacognitive ability, and begin to reveal the underlying nature of metacognitive inefficiency. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Similar history biases for distinct prospective decisions of self-performance

Metacognition can be deployed retrospectively -to reflect on the correctness of our behavior- or prospectively -to make predictions of success in one's future behavior or make decisions about strategies to solve future problems. We investigated the factors that determine prospective decision making. Human participants performed a visual discrimination task followed by ratings of visibility and response confidence. Prior to each trial, participants made prospective judgments. In Experiment 1, they rated their belief of future success. In Experiment 2, they rated their decision to adopt a focused attention state. Prospective beliefs of success were associated with no performance changes while prospective decisions to engage attention were followed by better self-evaluation of the correctness of behavioral responses. Using standard machine learning classifiers we found that the current prospective decision could be predicted from information concerning task-correctness, stimulus visibility and response confidence from previous trials. In both Experiments, awareness and confidence were more diagnostic of the prospective decision than task correctness. Notably, classifiers trained with prospective beliefs of success in Experiment 1 predicted decisions to engage in Experiment 2 and vice-versa. These results indicate that the formation of these seemingly different prospective decisions share a common, dynamic representational structure.

Can the VTA Come Out to Play? Only When the mPFC's Predictions Go Astray!

Confidence in perceptual decisions scales neural responses to violations in reward expectation. In this issue of Neuron, Lak et al. (2020) show that the medial prefrontal cortex in mice computes a confidence-dependent expectation signal that influences how dopamine neurons convey reward prediction errors to guide learning.

The Confidence Database

Understanding how people rate their confidence is critical for the characterization of a wide range of perceptual, memory, motor and cognitive processes. To enable the continued exploration of these processes, we created a large database of confidence studies spanning a broad set of paradigms, participant populations and fields of study. The data from each study are structured in a common, easy-to-use format that can be easily imported and analysed using multiple software packages. Each dataset is accompanied by an explanation regarding the nature of the collected data. At the time of publication, the Confidence Database (which is available at https://osf.io/s46pr/) contained 145 datasets with data from more than 8,700 participants and almost 4 million trials. The database will remain open for new submissions indefinitely and is expected to continue to grow. Here we show the usefulness of this large collection of datasets in four different analyses that provide precise estimations of several foundational confidence-related effects.

Abnormalities of confidence in psychiatry: an overview and future perspectives

Our behavior is constantly accompanied by a sense of confidence and its' precision is critical for adequate adaptation and survival. Importantly, abnormal confidence judgments that do not reflect reality may play a crucial role in pathological decision-making typically seen in psychiatric disorders. In this review, we propose abnormalities of confidence as a new model of interpreting psychiatric symptoms. We hypothesize a dysfunction of confidence at the root of psychiatric symptoms either expressed subclinically in the general population or clinically in the patient population. Our review reveals a robust association between confidence abnormalities and psychiatric symptomatology. Confidence abnormalities are present in subclinical/prodromal phases of psychiatric disorders, show a positive relationship with symptom severity, and appear to normalize after recovery. In the reviewed literature, the strongest evidence was found for a decline in confidence in (sub)clinical OCD, and for a decrease in confidence discrimination in (sub)clinical schizophrenia. We found suggestive evidence for increased/decreased confidence in addiction and depression/anxiety, respectively. Confidence abnormalities may help to understand underlying psychopathological substrates across disorders, and should thus be considered transdiagnostically. This review provides clear evidence for confidence abnormalities in different psychiatric disorders, identifies current knowledge gaps and supplies suggestions for future avenues. As such, it may guide future translational research into the underlying processes governing these abnormalities, as well as future interventions to restore them.