Metacognition in nonhuman primates: a review of current knowledge

Metacognition, the ability to monitor and control one's own cognitive processes, has long been considered a hallmark of human cognition. However, two decades of research have provided compelling evidence of metacognitive-like abilities in some nonhuman primates. This review synthesizes current knowledge on the subject, highlighting key experimental paradigms and empirical findings, with an emphasis on the latest studies. Thanks to advances in methods and efforts to counter alternative explanations, there is now a consensus that great apes and some macaque species can monitor and control some of their cognitive processes. Despite numerous investigations, however, whether capuchin monkeys are metacognitive remains unclear. Critical gaps persist in our understanding of metacognition across species. We discuss the importance of expanding research to include a wider range of primate species and the potential role of ecological factors in shaping metacognitive capacities. In addition, we consider some promising avenues for future research, including neurophysiological approaches, studies of metacognitive errors, and field experiments.

A Possible Neural Basis for Attentional Capture of Faces Revealed by Functional Magnetic Resonance Imaging and Causal Pharmacological Inactivation in Macaques

In primates, the presence of a face in a visual scene captures attention and rapidly directs the observer's gaze to the face, even when the face is not relevant to the task at hand. Here, we explored a neural circuit that might potentially play a causal role in this powerful behavior. In our previous research, two monkeys received microinfusions of muscimol, a γ-aminobutyric acid type A (GABAA)-receptor agonist, or saline (as a control condition) in separate sessions into individual or pairs of four inferotemporal face patches (middle and anterior lateral and fundal), as identified by an initial localizer experiment. Then, using fMRI, we measured the impact of each inactivation condition on responses in the other face patches relative to the control condition. In this study, we used the same method and measured the impact of each inactivation condition on responses in the FEF and the lateral intraparietal area, two regions associated with attentional processing, while face and nonface object stimuli were viewed. Our results revealed potential relationships between inferotemporal face patches and these two attention-related regions: The inactivation of the middle lateral and anterior fundal face patches had a pronounced impact on FEF, whereas the inactivation of the middle and anterior lateral face patches had a noticeable influence on LIP. Together, these initial exploratory findings document a circuit that potentially underlies the attentional capture of faces. Confirmation of the role of this circuit remains to be accomplished in the context of a paradigm that explicitly tests the attentional capture of faces.

Better bridges: Integrating the neuroscience and philosophy of consciousness

Contemporary consciousness research has given rise to numerous theories in both the philosophical and neuroscientific domains (such as higher-order theory and global neuronal workspace), raising the question as to how well each is supported. This article develops a relatively novel method for determining this, which is to use evidence, not just from a theory's own domain, but also from its complementary domain (e.g., neuroscientific evidence is used to judge a philosophical theory, and vice versa). This approach works when a neuroscientific and a philosophical theory are conceptually linked, allowing evidence confirming or disconfirming one theory to do the same for the other. After developing this method, the article uses it to draw conclusions concerning some of our leading neuroscientific and philosophical theories of consciousness, including first- and second-order representationalism and theories emphasizing the prefrontal cortex's role in consciousness.

Cortical Layer-Dependent Signaling in Cognition: Three Computational Modes of the Canonical Circuit

The cerebral cortex performs computations via numerous six-layer modules. The operational dynamics of these modules were studied primarily in early sensory cortices using bottom-up computation for response selectivity as a model, which has been recently revolutionized by genetic approaches in mice. However, cognitive processes such as recall and imagery require top-down generative computation. The question of whether the layered module operates similarly in top-down generative processing as in bottom-up sensory processing has become testable by advances in the layer identification of recorded neurons in behaving monkeys. This review examines recent advances in laminar signaling in these two computations, using predictive coding computation as a common reference, and shows that each of these computations recruits distinct laminar circuits, particularly in layer 5, depending on the cognitive demands. These findings highlight many open questions, including how different interareal feedback pathways, originating from and terminating at different layers, convey distinct functional signals.

Metacognition as a window into subjective affective experience

When patients seek professional help for mental disorders, they often do so because of troubling subjective affective experiences. While these subjective states are at the center of the patient's symptomatology, scientific tools for studying them and their cognitive antecedents are limited. Here, we explore the use of concepts and analytic tools from the science of consciousness, a field of research that has faced similar challenges in having to develop robust empirical methods for addressing a phenomenon that has been considered difficult to pin down experimentally. One important strand is the operationalization of some relevant processes in terms of metacognition and confidence ratings, which can be rigorously studied in both humans and animals. By assessing subjective experience with similar approaches, we hope to develop new scientific approaches for studying affective processes and promoting psychological resilience in the face of debilitating emotional experiences.

Neural correlates of metacognition: Disentangling the brain circuits underlying prospective and retrospective second-order judgments through noninvasive brain stimulation

Metacognition encompasses the capability to monitor and control one's cognitive processes, with metamemory and metadecision configuring among the most studied higher order functions. Although imaging experiments evaluated the role of disparate brain regions, neural substrates of metacognitive judgments remain undetermined. The aim of this systematic review is to summarize and discuss the available evidence concerning the neural bases of metacognition which has been collected by assessing the effects of noninvasive brain stimulation (NIBS) on human subjects' metacognitive capacities. Based on such literature analysis, our goal is, at first, to verify whether prospective and retrospective second-order judgments are localized within separate brain circuits and, subsequently, to provide compelling clues useful for identifying new targets for future NIBS studies. The search was conducted following the preferred reporting items for systematic reviews and meta-analyses guidelines among PubMed, PsycINFO, PsycARTICLES, PSYNDEX, MEDLINE, and ERIC databases. Overall, 25 studies met the eligibility criteria, yielding a total of 36 experiments employing transcranial magnetic stimulation and 16 ones making use of transcranial electrical stimulation techniques, including transcranial direct current stimulation and transcranial alternating current stimulation. Importantly, we found that both perspective and retrospective judgments about both memory and perceptual decision-making performances depend on the activation of the anterior and lateral portions of the prefrontal cortex, as well as on the activity of more caudal regions such as the premotor cortex and the precuneus. Combining this evidence with results from previous imaging and lesion studies, we advance ventromedial prefrontal cortex as a promising target for future NIBS studies.

Neural circuits for retrospective and prospective introspection for the past, present and future in macaque monkeys and humans

For animals, including humans, to have self-awareness, the ability to reflect on one's own perceptions and cognitions, which is known as metacognition, and an understanding of consistency of the self from the past to the present and into the future based on metacognition is essential. Through the mediation of self-consciousness, animals are thought to be able to proactively act to change their environment rather than passively responding to changes in their environment. However, it has not been known whether animals have self-awareness, and, if so, how it is implemented neurobiologically. In this review article, I introduce our studies examining the neural basis of metacognitive abilities for past, present, and future actions in macaque monkeys and humans, and explore the evolutionary origins of self-awareness.

Metacognition and Confidence: A Review and Synthesis

Determining the psychological, computational, and neural bases of confidence and uncertainty holds promise for understanding foundational aspects of human metacognition. While a neuroscience of confidence has focused on the mechanisms underpinning subpersonal phenomena such as representations of uncertainty in the visual or motor system, metacognition research has been concerned with personal-level beliefs and knowledge about self-performance. I provide a road map for bridging this divide by focusing on a particular class of confidence computation: propositional confidence in one's own (hypothetical) decisions or actions. Propositional confidence is informed by the observer's models of the world and their cognitive system, which may be more or less accurate-thus explaining why metacognitive judgments are inferential and sometimes diverge from task performance. Disparate findings on the neural basis of uncertainty and performance monitoring are integrated into a common framework, and a new understanding of the locus of action of metacognitive interventions is developed.

An easy-to-implement, non-invasive head restraint method for monkey fMRI

Functional magnetic resonance imaging (fMRI) in behaving monkeys has a strong potential to bridge the gap between human neuroimaging and primate neurophysiology. In monkey fMRI, to restrain head movements, researchers usually surgically implant a plastic head-post on the skull. Although time-proven to be effective, this technique could create burdens for animals, including a risk of infection and discomfort. Furthermore, the presence of extraneous objects on the skull, such as bone screws and dental cement, adversely affects signals near the cortical surface. These side effects are undesirable in terms of both the practical aspect of efficient data collection and the spirit of "refinement" from the 3R's. Here, we demonstrate that a completely non-invasive fMRI scan in awake monkeys is possible by using a plastic head mask made to fit the skull of individual animals. In all of the three monkeys tested, longitudinal, quantitative assessment of head movements showed that the plastic mask has effectively suppressed head movements, and we were able to obtain reliable retinotopic BOLD signals in a standard retinotopic mapping task. The present, easy-to-make plastic mask has a strong potential to simplify fMRI experiments in awake monkeys, while giving data that is as good as or even better quality than that obtained with the conventional head-post method.

Characterization of behavioral changes in T-maze alternation from dopamine D1 agonists with different receptor coupling mechanisms

RATIONALE

Dopamine D1 receptor agonists have been shown to improve working memory, but often have a non-monotonic (inverted-U) dose-response curve. One hypothesis is that this may reflect dose-dependent differential engagement of D1 signaling pathways, a mechanism termed functional selectivity or signaling bias.

OBJECTIVES AND METHODS

To test this hypothesis, we compared two D1 ligands with different signaling biases in a rodent T-maze alternation task. Both tested ligands (2-methyldihydrexidine and CY208243) have high intrinsic activity at cAMP signaling, but the former also has markedly higher intrinsic activity at D1-mediated recruitment of β-arrestin. The spatial working memory was assessed via the alternation behavior in the T-maze where the alternate choice rate quantified the quality of the memory and the duration prior to making a choice represented the decision latency.

RESULTS

Both D1 drugs changed the alternate rate and the choice latency in a dose-dependent manner, albeit with important differences. 2-Methyldihydrexidine was somewhat less potent but caused a more homogeneous improvement than CY208243 in spatial working memory. The maximum changes in the alternate rate and the choice latency tended to occur at different doses for both drugs.

CONCLUSIONS

These data suggest that D1 signaling bias in these two pathways (cAMP vs β-arrestin) has complex effects on cognitive processes as assessed by T-maze alternation. Understanding these mechanisms should allow the identification or discovery of D1 agonists that can provide superior cognitive enhancement.

Natural statistics support a rational account of confidence biases

Previous work has sought to understand decision confidence as a prediction of the probability that a decision will be correct, leading to debate over whether these predictions are optimal, and whether they rely on the same decision variable as decisions themselves. This work has generally relied on idealized, low-dimensional models, necessitating strong assumptions about the representations over which confidence is computed. To address this, we used deep neural networks to develop a model of decision confidence that operates directly over high-dimensional, naturalistic stimuli. The model accounts for a number of puzzling dissociations between decisions and confidence, reveals a rational explanation of these dissociations in terms of optimization for the statistics of sensory inputs, and makes the surprising prediction that, despite these dissociations, decisions and confidence depend on a common decision variable.

Do I know what they know? Linking metacognition, theory of mind, and selective social learning

Young children are often dependent on learning from others and to this effect develop heuristics to help distinguish reliable sources from unreliable sources. Where younger children rely heavily on social cues such as familiarity with a source to make this distinction, older children tend to rely more on an informant's competence. Little is known about the cognitive mechanisms that help children to select the best informant; however, some evidence points toward mechanisms such as metacognition (thinking about thinking) and theory of mind (thinking about other's thoughts) being involved. The goals of the current study were to (a) explore how the monitoring and control components of metacognition may predict selective social learning in preschoolers and (b) attempt to replicate a reported link between selective social learning and theory of mind. In Experiment 1, no relationship was observed across the measures. In Experiment 2, only selective social learning and belief reasoning were found to be related as well as when both experiments' samples were combined. No links between selective social learning and metacognition were observed in the two experiments. These results suggest that theory of mind is a stronger correlate of selective learning than metacognition in young children. The implications regarding the kind of tasks used to measure metacognition are discussed.

Imagining the future self through thought experiments

The ability of the mind to conceptualize what is not present is essential. It allows us to reason counterfactually about what might have happened had events unfolded differently or had another course of action been taken. It allows us to think about what might happen - to perform 'Gedankenexperimente' (thought experiments) - before we act. However, the cognitive and neural mechanisms mediating this ability are poorly understood. We suggest that the frontopolar cortex (FPC) keeps track of and evaluates alternative choices (what we might have done), whereas the anterior lateral prefrontal cortex (alPFC) compares simulations of possible future scenarios (what we might do) and evaluates their reward values. Together, these brain regions support the construction of suppositional scenarios.

Two-Year-Olds' Eye Movements Reflect Confidence in Their Understanding of Words

We studied the fundamental issue of whether children evaluate the reliability of their language interpretation, that is, their confidence in understanding words. In two experiments, 2-year-olds (Experiment 1: N = 50; Experiment 2: N = 60) saw two objects and heard one of them being named; both objects were then hidden behind screens and children were asked to look toward the named object, which was eventually revealed. When children knew the label used, they showed increased postdecision persistence after a correct compared with an incorrect anticipatory look, a marker of confidence in word comprehension (Experiment 1). When interacting with an unreliable speaker, children showed accurate word comprehension but reduced confidence in the accuracy of their own choice, indicating that children's confidence estimates are influenced by social information (Experiment 2). Thus, by the age of 2 years, children can estimate their confidence during language comprehension, long before they can talk about their linguistic skills.

On the assumptions behind metacognitive measurements: Implications for theory and practice

Theories of visual confidence have largely been grounded in the gaussian signal detection framework. This framework is so dominant that idiosyncratic consequences from this distributional assumption have remained unappreciated. This article reports systematic comparisons of the gaussian signal detection framework to its logistic counterpart in the measurement of metacognitive accuracy. Because of the difference in their distribution kurtosis, these frameworks are found to provide different perspectives regarding the efficiency of confidence rating relative to objective decision (the logistic model intrinsically gives greater meta-d'/d' ratio than the gaussian model). These frameworks can also provide opposing conclusions regarding the metacognitive inefficiency along the internal evidence continuum (whether meta-d' is larger or smaller for higher levels of confidence). Previous theories developed on these lines of analysis may need to be revisited as the gaussian and logistic metacognitive models received somewhat equivalent support in our quantitative model comparisons. Despite these discrepancies, however, we found that across-condition or across-participant comparisons of metacognitive measures are relatively robust against the distributional assumptions, which provides much assurance to conventional research practice. We hope this article promotes the awareness for the significance of hidden modeling assumptions, contributing to the cumulative development of the relevant field.

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.

Metacognition as a Consequence of Competing Evolutionary Time Scales

Evolution is full of coevolving systems characterized by complex spatio-temporal interactions that lead to intertwined processes of adaptation. Yet, how adaptation across multiple levels of temporal scales and biological complexity is achieved remains unclear. Here, we formalize how evolutionary multi-scale processing underlying adaptation constitutes a form of metacognition flowing from definitions of metaprocessing in machine learning. We show (1) how the evolution of metacognitive systems can be expected when fitness landscapes vary on multiple time scales, and (2) how multiple time scales emerge during coevolutionary processes of sufficiently complex interactions. After defining a metaprocessor as a regulator with local memory, we prove that metacognition is more energetically efficient than purely object-level cognition when selection operates at multiple timescales in evolution. Furthermore, we show that existing modeling approaches to coadaptation and coevolution-here active inference networks, predator-prey interactions, coupled genetic algorithms, and generative adversarial networks-lead to multiple emergent timescales underlying forms of metacognition. Lastly, we show how coarse-grained structures emerge naturally in any resource-limited system, providing sufficient evidence for metacognitive systems to be a prevalent and vital component of (co-)evolution. Therefore, multi-scale processing is a necessary requirement for many evolutionary scenarios, leading to de facto metacognitive evolutionary outcomes.

Conversion of concept-specific decision confidence into integrative introspection in primates

Introspection based on the integration of uncertain evidence is critical for acting upon abstract thinking and imagining future scenarios. However, it is unknown how confidence read-outs from multiple sources of different concepts are integrated, especially considering the relationships among the concepts. In this study, monkeys performed wagering based on an estimation of their performance in a preceding mnemonic decision. We found that the longer the response times for post-decision wagering, the more relieved the impairments having been caused by frontal disruption. This suggests the existence of a time-consuming compensatory metacognitive process. We found posterior inferior parietal lobe (pIPL) as its candidate, which was not coding the wagering per se (i.e., just high bet or low bet), but became more active when monkeys successfully chose the optimal bet option based on mnemonic decision performance. Thereafter, the pIPL prompts dorsal anterior cingulate cortex to carry the chosen wagering option. Our findings suggest a role for the pIPL in metacognitive concept integration.

Rodents monitor their error in self-generated duration on a single trial basis

A fundamental question in neuroscience is what type of internal representation leads to complex, adaptive behavior. When faced with a deadline, individuals' behavior suggests that they represent the mean and the uncertainty of an internal timer to make near-optimal, time-dependent decisions. Whether this ability relies on simple trial-and-error adjustments or whether it involves richer representations is unknown. Richer representations suggest a possibility of error monitoring, that is, the ability for an individual to assess its internal representation of the world and estimate discrepancy in the absence of external feedback. While rodents show timing behavior, whether they can represent and report temporal errors in their own produced duration on a single-trial basis is unknown. We designed a paradigm requiring rats to produce a target time interval and, subsequently, evaluate its error. Rats received a reward in a given location depending on the magnitude of their timing errors. During the test trials, rats had to choose a port corresponding to the error magnitude of their just-produced duration to receive a reward. High-choice accuracy demonstrates that rats kept track of the values of the timing variables on which they based their decision. Additionally, the rats kept a representation of the mapping between those timing values and the target value, as well as the history of the reinforcements. These findings demonstrate error-monitoring abilities in evaluating self-generated timing in rodents. Together, these findings suggest an explicit representation of produced duration and the possibility to evaluate its relation to the desired target duration.

Metamemory: Rats know the strength of their memory

Assessing the quality of one's own memories is a core cognitive function, but it has been unclear whether rodents possess this ability. Evidence that they do has come from research using a new behavioural paradigm in which rats make temporal bets guided by memory confidence.

Rats use memory confidence to guide decisions

Memory enables access to past experiences to guide future behavior. Humans can determine which memories to trust (high confidence) and which to doubt (low confidence). How memory retrieval, memory confidence, and memory-guided decisions are related, however, is not understood. In particular, how confidence in memories is used in decision making is unknown. We developed a spatial memory task in which rats were incentivized to gamble their time: betting more following a correct choice yielded greater reward. Rat behavior reflected memory confidence, with higher temporal bets following correct choices. We applied machine learning to identify a memory decision variable and built a generative model of memories evolving over time that accurately predicted both choices and confidence reports. Our results reveal in rats an ability thought to exist exclusively in primates and introduce a unified model of memory dynamics, retrieval, choice, and confidence.

Chemogenetic sensory fMRI reveals behaviorally relevant bidirectional changes in primate somatosensory network

Concurrent genetic neuromodulation and functional magnetic resonance imaging (fMRI) in primates has provided a valuable opportunity to assess the modified brain-wide operation in the resting state. However, its application to link the network operation with behavior still remains challenging. Here, we combined chemogenetic silencing of the primary somatosensory cortex (SI) with tactile fMRI and related behaviors in macaques. Focal chemogenetic silencing of functionally identified SI hand region impaired grasping behavior. The same silencing also attenuated hand stimulation-evoked fMRI signal at both the local silencing site and the anatomically and/or functionally connected downstream grasping network, suggesting altered network operation underlying the induced behavioral impairment. Furthermore, the hand region silencing unexpectedly disinhibited foot representation with accompanying behavioral hypersensitization. These results demonstrate that focal chemogenetic silencing with sensory fMRI in macaques unveils bidirectional network changes to generate multifaceted behavioral impairments, thereby opening a pivotal window toward elucidating the causal network operation underpinning higher brain functions in primates.

Metacognitive resources for adaptive learning⋆

Biological organisms display remarkably flexible behaviours. This is an area of active investigation, in particular in the fields of artificial intelligence, computational and cognitive neuroscience. While inductive biases and broader cognitive functions are undoubtedly important, the ability to monitor and evaluate one's performance or oneself -- metacognition -- strikes as a powerful resource for efficient learning. Often measured as decision confidence in neuroscience and psychology experiments, metacognition appears to reflect a broad range of abstraction levels and downstream behavioural effects. Within this context, the formal investigation of how metacognition interacts with learning processes is a recent endeavour. Of special interest are the neural and computational underpinnings of confidence and reinforcement learning modules. This review discusses a general hierarchy of confidence functions and their neuro-computational relevance for adaptive behaviours. It then introduces novel ways to study the formation and use of meta-representations and nonconscious mental representations related to learning and confidence, and concludes with a discussion on outstanding questions and wider perspectives.

Common functional localizers to enhance NHP & cross-species neuroscience imaging research

Functional localizers are invaluable as they can help define regions of interest, provide cross-study comparisons, and most importantly, allow for the aggregation and meta-analyses of data across studies and laboratories. To achieve these goals within the non-human primate (NHP) imaging community, there is a pressing need for the use of standardized and validated localizers that can be readily implemented across different groups. The goal of this paper is to provide an overview of the value of localizer protocols to imaging research and we describe a number of commonly used or novel localizers within NHPs, and keys to implement them across studies. As has been shown with the aggregation of resting-state imaging data in the original PRIME-DE submissions, we believe that the field is ready to apply the same initiative for task-based functional localizers in NHP imaging. By coming together to collect large datasets across research group, implementing the same functional localizers, and sharing the localizers and data via PRIME-DE, it is now possible to fully test their robustness, selectivity and specificity. To do this, we reviewed a number of common localizers and we created a repository of well-established localizer that are easily accessible and implemented through the PRIME-RE platform.

Minimal specifications for non-human primate MRI: Challenges in standardizing and harmonizing data collection

Recent methodological advances in MRI have enabled substantial growth in neuroimaging studies of non-human primates (NHPs), while open data-sharing through the PRIME-DE initiative has increased the availability of NHP MRI data and the need for robust multi-subject multi-center analyses. Streamlined acquisition and analysis protocols would accelerate and improve these efforts. However, consensus on minimal standards for data acquisition protocols and analysis pipelines for NHP imaging remains to be established, particularly for multi-center studies. Here, we draw parallels between NHP and human neuroimaging and provide minimal guidelines for harmonizing and standardizing data acquisition. We advocate robust translation of widely used open-access toolkits that are well established for analyzing human data. We also encourage the use of validated, automated pre-processing tools for analyzing NHP data sets. These guidelines aim to refine methodological and analytical strategies for small and large-scale NHP neuroimaging data. This will improve reproducibility of results, and accelerate the convergence between NHP and human neuroimaging strategies which will ultimately benefit fundamental and translational brain science.

Combining brain perturbation and neuroimaging in non-human primates

Brain perturbation studies allow detailed causal inferences of behavioral and neural processes. Because the combination of brain perturbation methods and neural measurement techniques is inherently challenging, research in humans has predominantly focused on non-invasive, indirect brain perturbations, or neurological lesion studies. Non-human primates have been indispensable as a neurobiological system that is highly similar to humans while simultaneously being more experimentally tractable, allowing visualization of the functional and structural impact of systematic brain perturbation. This review considers the state of the art in non-human primate brain perturbation with a focus on approaches that can be combined with neuroimaging. We consider both non-reversible (lesions) and reversible or temporary perturbations such as electrical, pharmacological, optical, optogenetic, chemogenetic, pathway-selective, and ultrasound based interference methods. Method-specific considerations from the research and development community are offered to facilitate research in this field and support further innovations. We conclude by identifying novel avenues for further research and innovation and by highlighting the clinical translational potential of the methods.

Explaining distortions in metacognition with an attractor network model of decision uncertainty

Metacognition is the ability to reflect on, and evaluate, our cognition and behaviour. Distortions in metacognition are common in mental health disorders, though the neural underpinnings of such dysfunction are unknown. One reason for this is that models of key components of metacognition, such as decision confidence, are generally specified at an algorithmic or process level. While such models can be used to relate brain function to psychopathology, they are difficult to map to a neurobiological mechanism. Here, we develop a biologically-plausible model of decision uncertainty in an attempt to bridge this gap. We first relate the model's uncertainty in perceptual decisions to standard metrics of metacognition, namely mean confidence level (bias) and the accuracy of metacognitive judgments (sensitivity). We show that dissociable shifts in metacognition are associated with isolated disturbances at higher-order levels of a circuit associated with self-monitoring, akin to neuropsychological findings that highlight the detrimental effect of prefrontal brain lesions on metacognitive performance. Notably, we are able to account for empirical confidence judgements by fitting the parameters of our biophysical model to first-order performance data, specifically choice and response times. Lastly, in a reanalysis of existing data we show that self-reported mental health symptoms relate to disturbances in an uncertainty-monitoring component of the network. By bridging a gap between a biologically-plausible model of confidence formation and observed disturbances of metacognition in mental health disorders we provide a first step towards mapping theoretical constructs of metacognition onto dynamical models of decision uncertainty. In doing so, we provide a computational framework for modelling metacognitive performance in settings where access to explicit confidence reports is not possible.

Identification and disruption of a neural mechanism for accumulating prospective metacognitive information prior to decision-making

More than one type of probability must be considered when making decisions. It is as necessary to know one's chance of performing choices correctly as it is to know the chances that desired outcomes will follow choices. We refer to these two choice contingencies as internal and external probability. Neural activity across many frontal and parietal areas reflected internal and external probabilities in a similar manner during decision-making. However, neural recording and manipulation approaches suggest that one area, the anterior lateral prefrontal cortex (alPFC), is highly specialized for making prospective, metacognitive judgments on the basis of internal probability; it is essential for knowing which decisions to tackle, given its assessment of how well they will be performed. Its activity predicted prospective metacognitive judgments, and individual variation in activity predicted individual variation in metacognitive judgments. Its disruption altered metacognitive judgments, leading participants to tackle perceptual decisions they were likely to fail.

Single-unit Recording in Awake Behaving Non-human Primates

Non-human primates (NHPs) have been widely used as a species model in studies to understand higher brain functions in health and disease. These studies employ specifically designed behavioral tasks in which animal behavior is well-controlled, and record neuronal activity at high spatial and temporal resolutions while animals are performing the tasks. Here, we present a detailed procedure to conduct single-unit recording, which fulfils high spatial and temporal resolutions while macaque monkeys (i.e., widely used NHPs) perform behavioral tasks in a well-controlled manner. This procedure was used in our previous study to investigate the dynamics of neuronal activity during economic decision-making by the monkeys. Monkeys' behavior was quantitated by eye position tracking and button press/release detection. By inserting a microelectrode into the brain, with a grid system in reference to magnetic resonance imaging, we precisely recorded the brain regions. Our experimental system permits rigorous investigation of the link between neuronal activity and behavior.

Socially relative reward valuation in the primate brain

Reward valuation in social contexts is by nature relative rather than absolute; it is made in reference to others. This socially relative reward valuation is based on our propensity to conduct comparisons and competitions between self and other. Exploring its neural substrate has been an active area of research in human neuroimaging. More recently, electrophysiological investigation of the macaque brain has enabled us to understand neural mechanisms underlying this valuation process at single-neuron and network levels. Here I show that shared neural networks centered at the medial prefrontal cortex and dopamine-related subcortical regions are involved in this process in humans and nonhuman primates. Thus, socially relative reward valuation is mediated by cortico-subcortically coordinated activity linking social and reward brain networks.

Fine-Grained Topography and Modularity of the Macaque Frontal Pole Cortex Revealed by Anatomical Connectivity Profiles

The frontal pole cortex (FPC) plays key roles in various higher-order functions and is highly developed in non-human primates. An essential missing piece of information is the detailed anatomical connections for finer parcellation of the macaque FPC than provided by the previous tracer results. This is important for understanding the functional architecture of the cerebral cortex. Here, combining cross-validation and principal component analysis, we formed a tractography-based parcellation scheme that applied a machine learning algorithm to divide the macaque FPC (2 males and 6 females) into eight subareas using high-resolution diffusion magnetic resonance imaging with the 9.4T Bruker system, and then revealed their subregional connections. Furthermore, we applied improved hierarchical clustering to the obtained parcels to probe the modular structure of the subregions, and found that the dorsolateral FPC, which contains an extension to the medial FPC, was mainly connected to regions of the default-mode network. The ventral FPC was mainly involved in the social-interaction network and the dorsal FPC in the metacognitive network. These results enhance our understanding of the anatomy and circuitry of the macaque brain, and contribute to FPC-related clinical research.

Off-Peak 594-nm Light Surpasses On-Peak 532-nm Light in Silencing Distant ArchT-Expressing Neurons In Vivo

For large brain volume manipulations using optogenetics, both effective opsin excitation and efficient light delivery with minimal light absorption are required to minimize the illuminating light intensity and concomitant off-target effects. ArchT, a widely used potent inhibitory opsin, is commonly activated by 532-nm light, which lies on its in vitro excitation peak. However, 532-nm light also lies on a peak range of the hemoglobin absorption spectrum. Therefore, we predicted that 594-nm light is superior in suppressing distant ArchT-expressing neurons, which is slightly off the ArchT-excitation-plateau and largely off the peak of the hemoglobin absorption spectrum. We quantitatively tested this prediction by the electrophysiological recording of the rat cortex in vivo. At illumination distances greater than 500 μm, 594-nm light was more effective than 532-nm light. Its superiority increased with distance. These results validate our prediction and highlight the significance of excitation-absorption trade-off in selecting illumination wavelength for optogenetics in vivo.

•

Wavelength-dependency of optogenetic neuronal control was directly measured in vivo

•

Off-peak light silence 1-mm-distant ArchT-neuron twice more than on-peak light in vivo

•

Superiority of off-peak light at distance arose from its less absorption of light

•

Simulation of light propagation supported unexpectedly large effect of hemoglobin

Behavior- and Modality-General Representation of Confidence in Orbitofrontal Cortex

Every decision we make is accompanied by a sense of confidence about its likely outcome. This sense informs subsequent behavior, such as investing more-whether time, effort, or money-when reward is more certain. A neural representation of confidence should originate from a statistical computation and predict confidence-guided behavior. An additional requirement for confidence representations to support metacognition is abstraction: they should emerge irrespective of the source of information and inform multiple confidence-guided behaviors. It is unknown whether neural confidence signals meet these criteria. Here, we show that single orbitofrontal cortex neurons in rats encode statistical decision confidence irrespective of the sensory modality, olfactory or auditory, used to make a choice. The activity of these neurons also predicts two confidence-guided behaviors: trial-by-trial time investment and cross-trial choice strategy updating. Orbitofrontal cortex thus represents decision confidence consistent with a metacognitive process that is useful for mediating confidence-guided economic decisions.

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.

Explicit memory and cognition in monkeys

Taxonomies of human memory, influenced heavily by Endel Tulving, make a fundamental distinction between explicit and implicit memory. Humans are aware of explicit memories, whereas implicit memories control behavior even though we are not aware of them. Efforts to understand the evolution of memory, and to use nonhuman animals to model human memory, will be facilitated by better understanding the extent to which this critical distinction exists in nonhuman animals. Work with metacognition paradigms in the past 20 years has produced a strong case for the existence of explicit memory in nonhuman primates and possibly other nonhuman animals. Clear dissociations of explicit and implicit memory by metacognition have yet to be demonstrated in nonhumans, although dissociations between memory systems by other behavioral techniques, and by brain manipulations, suggest that the explicit-implicit distinction applies to nonhumans. Neurobehavioral studies of metamemory are beginning to identify neural substrates for memory monitoring in the frontal cortex of monkeys. We have strong evidence that at least some memory systems are explicit in rhesus monkeys, but we need to learn more about the distribution of explicit processes across cognitive systems within monkeys, and across species.

Disruptions of frontoparietal control network and default mode network linking the metacognitive deficits with clinical symptoms in schizophrenia

The metacognitive deficit in awareness of one's own mental states is a core feature of schizophrenia (SZ). The previous studies suggested that the metacognitive deficit associates with clinical symptoms. However, the neural mechanisms underlying the relationship remain largely unknown. We here investigated the neural activities associated with the metacognitive deficit and the neural signatures associated with clinical symptoms in 38 patients with SZ using functional magnetic resonance imaging with a perceptual decision-making task accompanied with metacognition, in comparison to 38 age, gender, and education matched healthy control subjects. The metacognitive deficit in patients with SZ was associated with reduced regional activity in both the frontoparietal control network (FPCN) and the default mode network. Critically, the anticorrelational balance between the two disrupted networks was substantially altered during metacognition, and the extent of alteration positively scaled with negative symptoms. Conversely, decoupling between the two networks was impaired when metacognitive monitoring was not required, and the strength of excessive neural activity positively scaled with positive symptoms. Thus, disruptions of the FPCN and the default mode network underlie the metacognitive deficit, and alternations of network balance between the two networks correlate with clinical symptoms in SZ. These findings implicate that rebalancing these networks holds important clinical potential in developing more efficacious therapeutic treatments.

Evaluation of Self-generated Behavior: Untangling Metacognitive Readout and Error Detection

When producing a duration, for instance, by pressing a key for 1 sec, the brain relies on self-generated neuronal dynamics to monitor the “flow of time.” Evidence has suggested that the brain can also monitor itself monitoring time, the so-called self-evaluation. How are temporal errors inferred on the basis of purely internally driven brain dynamics with no external reference for time? Although studies have shown that participants can reliably detect temporal errors when generating a duration, the neural bases underlying the evaluation of this self-generated temporal behavior are unknown. Theories of psychological time have also remained silent about such self-evaluation abilities. We assessed the contributions of an error-detection mechanism, in which error detection results from the ability to estimate the latency of motor actions, and of a readout mechanism, in which errors would result from inferring the state of a duration representation. Error detection predicts a V-shape association between neural activity and self-evaluation at the offset of a produced interval, whereas the readout predicts a linear association. Here, human participants generated a time interval and evaluated the magnitude of their timing (first- and second-order behavioral judgments, respectively). Focusing on the MEG/EEG signatures after the termination of the self-generated duration, we found several cortical sources involved in performance monitoring displaying a linear association between the power of alpha (α = 8–14 Hz) oscillations and self-evaluation. Altogether, our results support the readout hypothesis and indicate that duration representation may be integrated for the evaluation of self-generated behavior.

MRI-based visualization of rTMS-induced cortical plasticity in the primary motor cortex

Repetitive transcranial magnetic stimulation (rTMS) induces changes in cortical excitability for minutes to hours after the end of intervention. However, it has not been precisely determined to what extent cortical plasticity prevails spatially in the cortex. Recent studies have shown that rTMS induces changes in “interhemispheric” functional connectivity, the resting-state functional connectivity between the stimulated region and the symmetrically corresponding region in the contralateral hemisphere. In the present study, quadripulse stimulation (QPS) was applied to the index finger representation in the left primary motor cortex (M1), while the position of the stimulation coil was constantly monitored by an online navigator. After QPS application, resting-state functional magnetic resonance imaging was performed, and the interhemispheric functional connectivity was compared with that before QPS. A cluster of connectivity changes was observed in the stimulated region in the central sulcus. The cluster was spatially extended approximately 10 mm from the center [half width at half maximum (HWHM): approximately 3 mm] and was extended approximately 20 mm long in depth (HWHM: approximately 7 mm). A localizer scan of the index finger motion confirmed that the cluster of interhemispheric connectivity changes overlapped spatially with the activation related to the index finger motion. These results indicate that cortical plasticity in M1 induced by rTMS was relatively restricted in space and suggest that rTMS can reveal functional dissociation associated with adjacent small areas by inducing neural plasticity in restricted cortical regions.

Contributions of anterior cingulate cortex and basolateral amygdala to decision confidence and learning under uncertainty

The subjective sense of certainty, or confidence, in ambiguous sensory cues can alter the interpretation of reward feedback and facilitate learning. We trained rats to report the orientation of ambiguous visual stimuli according to a spatial stimulus-response rule that must be learned. Following choice, rats could wait a self-timed delay for reward or initiate a new trial. Waiting times increase with discrimination accuracy, demonstrating that this measure can be used as a proxy for confidence. Chemogenetic silencing of BLA shortens waiting times overall whereas ACC inhibition renders waiting times insensitive to confidence-modulating attributes of visual stimuli, suggesting contribution of ACC but not BLA to confidence computations. Subsequent reversal learning is enhanced by confidence. Both ACC and BLA inhibition block this enhancement but via differential adjustments in learning strategies and consistent use of learned rules. Altogether, we demonstrate dissociable roles for ACC and BLA in transmitting confidence and learning under uncertainty.

Perirhinal circuits for memory processing

The perirhinal cortex (PRC) serves as the gateway to the hippocampus for episodic memory formation and plays a part in retrieval through its backward connectivity to various neocortical areas. First, I present the evidence suggesting that PRC neurons encode both experientially acquired object features and their associative relations. Recent studies have revealed circuit mechanisms in the PRC for the retrieval of cue-associated information, and have demonstrated that, in monkeys, PRC neuron-encoded information can be behaviourally read out. These studies, among others, support the theory that the PRC converts visual representations of an object into those of its associated features and initiates backward-propagating, interareal signalling for retrieval of nested associations of object features that, combined, extensionally represent the object meaning. I propose that the PRC works as the ventromedial hub of a 'two-hub model' at an apex of the hierarchy of a distributed memory network and integrates signals encoded in other downstream cortical areas that support diverse aspects of knowledge about an object.

The Blind Men and the Elephant: The Quest for Open Data Repositories

Nonhuman primate imaging is maturing into a solid subfield within basic and translational neurosciences. In the present issue of Neuron, Milham et al. (2018) present a data repository to openly share anatomical, functional, and diffusion-based neuroimaging data from monkeys.

Individual susceptibility to TMS affirms the precuneal role in meta-memory upon recollection

A recent virtual-lesion study using inhibitory repetitive transcranial magnetic stimulation (rTMS) confirmed the causal behavioral relevance of the precuneus in the evaluation of one's own memory performance (aka mnemonic metacognition). This study's goal is to elucidate how these TMS-induced neuromodulatory effects might relate to the neural correlates and be modulated by individual anatomical profiles in relation to meta-memory. In a within-subjects design, we assessed the impact of 20-min rTMS over the precuneus, compared to the vertex, across three magnetic resonance imaging (MRI) neuro-profiles on 18 healthy subjects during a memory versus a perceptual task. Task-based functional MRI revealed that BOLD signal magnitude in the precuneus is associated with variation in individual meta-memory efficiency. Moreover, individuals with higher resting-state functional connectivity (rs-fcMRI) between the precuneus and the hippocampus, or smaller gray matter volume in the stimulated precuneal region exhibit considerably higher vulnerability to the TMS effect. These effects were not observed in the perceptual domain. Thus, we provide compelling evidence in outlining a possible circuit encompassing the precuneus and its mnemonic midbrain neighbor the hippocampus at the service of realizing our meta-awareness during memory recollection of episodic details.

Developing a Reflective Mind: From Core Metacognition to Explicit Self-Reflection

Metacognition is the ability to monitor and control cognition. Because young children often provide inaccurate metacognitive judgments when prompted to do so verbally, it has long been assumed that this ability does not develop until late childhood. This claim is now challenged by new studies using nonverbal paradigms and revealing that basic forms of metacognition—such as the ability to estimate decision confidence or to monitor errors—are present even in preverbal infants. This new line of evidence suggests that young children adapt to their environment not only by considering their physical and social surroundings but also by reflecting on their own cognitive states.

Mnemonic Introspection in Macaques Is Dependent on Superior Dorsolateral Prefrontal Cortex But Not Orbitofrontal Cortex

The human PFC has been associated more with meta-perceptual as opposed to meta-memory decisions from correlational neuroimaging investigations. Recently, metacognitive abilities have also been shown to be causally dependent upon anterior and dorsal PFC in nonhuman primate lesion studies. Two studies, using postdecision wagering paradigms and reversible inactivation, challenged this meta-perceptual versus meta-memory notion and showed that dorsal and anterior prefrontal areas are associated with metamemory for experienced objects and awareness of ignorance, respectively. Causal investigations are important but scarce; nothing is known, for example, about the causal contributions of prefrontal subregions to spatial metamemory. Here, we investigated the effects of dorsal versus ventral PFC lesions on two-alternative forced-choice spatial discrimination tasks in male macaque monkeys. Importantly, we were rigorous in approach and applied three independent but complementary indices used to quantify individual animals' metacognitive ability (“Type II sensitivity”) by two variants of meta-d′/d′ and phi coefficient (φ). Our results were consistent across indices: while neither lesions to superior dorsolateral PFC nor orbitofrontal cortex impaired spatial recognition performance, only monkeys with superior dorsolateral PFC lesions were impaired in meta-accuracy. Together with the observation that the same orbitofrontal cortex lesioned monkeys were impaired in updating rule value in a Wisconsin Card Sorting Test analog, we therefore document a functional double-dissociation between these two PFC regions. Our study presents important causal evidence that other dimensions, namely, domain-specific processing (e.g., spatial vs nonspatial metamemory), also need considerations in understanding the functional specialization in the neural underpinnings of introspection.

SIGNIFICANCE STATEMENT This study demonstrates macaque monkeys' metacognitive capability of introspecting its own memory success is causally dependent on intact superior dorsolateral prefrontal cortices but not the orbitofrontal cortices. Combining neurosurgical techniques on monkeys and state-of-the-art measures of metacognition, we affirm a critical role of the PFC in supporting spatial meta-recognition memory and delineate functional specificity within PFC for distinct elements of metacognition.

The macaque anterior cingulate cortex translates counterfactual choice value into actual behavioral change

The neural mechanisms mediating sensory-guided decision making have received considerable attention but animals often pursue behaviors for which there is currently no sensory evidence. Such behaviors are guided by internal representations of choice values that have to be maintained even when these choices are unavailable. We investigated how four macaque monkeys maintained representations of the value of counterfactual choices– choices that could not be taken at the current moment but which could be taken in the future. Using functional magnetic resonance imaging, we found two different patterns of activity co-varying with values of counterfactual choices in a circuit spanning hippocampus, anterior lateral prefrontal cortex, and anterior cingulate cortex (ACC). ACC activity also reflected whether the internal value representations would be translated into actual behavioral change. To establish the causal importance of ACC for this translation process, we used a novel technique, Transcranial Focused Ultrasound Stimulation, to reversibly disrupt ACC activity.

Origin and evolution of human cognition

We investigate which of the higher cognitive abilities or types of intelligence characteristic of humans are found, even in preliminary form, in non-human animals, predominantly primates, or whether qualitatively different ("unique") human abilities exist. This concerns (1) tool use and fabrication, (2) problem solving, (3) gaze following, (4) mirror self-recognition, (5) imitation, (6) metacognition, (7) theory of mind, (8) consciousness, (9) prosociality, and (10) language. We found that none of these abilities can be regarded as unique to humans without precursors in non-human primates. The observed differences in cognitive functions, underlying brain mechanisms and resulting behaviors correlate best with differences in the information processing capacity as an equivalent of general intelligence based on the number of cortical neurons, packing density and axonal conduction velocity plus long-range cortical fascicles. The biggest quantitative change appears to concern the origin of syntactical language, but this was preceded by an increased mental ability to manipulate sequential events within working memory.

Dopamine is a double-edged sword: dopaminergic modulation enhances memory retrieval performance but impairs metacognition

While memory encoding and consolidation processes have been linked with dopaminergic signaling for a long time, the role of dopamine in episodic memory retrieval remained mostly unexplored. Based on previous observations of striatal activity during memory retrieval, we used pharmacological functional magnetic resonance imaging to investigate the effects of dopamine on retrieval performance and metacognitive memory confidence in healthy humans. Dopaminergic modulation by the D2 antagonist haloperidol administered acutely during the retrieval phase improved recognition accuracy of previously learned pictures significantly and was associated with increased activity in the substantia nigra/ventral tegmental area, locus coeruleus, hippocampus, and amygdala during retrieval. In contrast, confidence for new decisions was impaired by unsystematically increased activity of the striatum across confidence levels and restricted range of responsiveness in frontostriatal networks under haloperidol. These findings offer new insights into the mechanisms underlying memory retrieval and metacognition and provide a broader perspective on the presence of memory problems in dopamine-related diseases and the treatment of memory disorders.