![System 2 metacognition for cognitive control across two agents. System 2 metacognitive representations are derived from information in system 1, but they are in a form available for verbal report. For example, the reliability of a sensory signal can be reported in terms of confidence. When agents are cooperating, these reports can be used to optimise control by, for example, giving more weight to the more confident observer [32]. Via system 2, verbal reports can also have long-term effects on the functioning of system 1 [57].](https://www.researchgate.net/profile/Chris-Frith/publication/260444732/figure/fig1/AS:282181290217474@1444288670503/System-2-metacognition-for-cognitive-control-across-two-agents-System-2-metacognitive_Q320.jpg)
![System 1 metacognition with a single agent. The control of cognitive processes in an individual's system 1 occurs automatically and at a sub-personal level [3]. Contention scheduling, a term introduced by Norman and Shallice [18], refers to a mechanism for resolving competition between processes that overlap in their effector system requirements. Metacognitive representations reflect the properties of the functioning of these cognitive processes. System 1 uses metacognitive representations to improve control. An example of such representations would be the reliability of sensory signals. Cognitive processes estimated to be more reliable can be given greater weight [12].](https://www.researchgate.net/profile/Chris-Frith/publication/260444732/figure/fig2/AS:282181298606109@1444288672485/System-1-metacognition-with-a-single-agent-The-control-of-cognitive-processes-in-an_Q320.jpg)
Content uploaded by Chris D Frith
Author content
All content in this area was uploaded by Chris D Frith on Aug 16, 2014
Content may be subject to copyright.
A preview of the PDF is not available
Supra-personal cognitive control and metacognition
Abstract and Figures
The human mind is extraordinary in its ability not merely to respond to events as they unfold but also to adapt its own operation in pursuit of its agenda. This 'cognitive control' can be achieved through simple interactions among sensorimotor processes, and through interactions in which one sensorimotor process represents a property of another in an implicit, unconscious way. So why does the human mind also represent properties of cognitive processes in an explicit way, enabling us to think and say 'I'm sure' or 'I'm doubtful'? We suggest that 'system 2 metacognition' is for supra-personal cognitive control. It allows metacognitive information to be broadcast, and thereby to coordinate the sensorimotor systems of two or more agents involved in a shared task.
Figures - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
... Personality traits and metacognition can both involve automatic or habitual patterns of thinking and behaving. Mindfulness practices promote a shift from automaticity to conscious awareness and intentional action (Shea et al., 2014). By encouraging individuals to be more present and mindful in the moment, mindfulness can help break automatic thought patterns associated with personality traits and facilitate metacognitive processes that involve intentional reflection, evaluation, and self-regulation. ...
The aim of this research is to determine the relationship between personality traits and metacognition and the regulatory role of mindfulness in this relationship. The research is a relationship-scanning model, and Hayes' process plugin was used for regulatory role analysis. The research sample consisted of 413 individuals between the ages of 18-51. The Metacognition Scale-30, Mindfulness Scale, and Five-Factor Personality Scale Short Form were used for data collection. According to the findings of the study, there is a negative relationship between extraversion, agreeableness, conscientiousness, openness to experience, mindfulness, and pathological metacognition and a positive relationship with neuroticism. Mindfulness was not found to play a significant regulatory role in the relationship between personality traits and metacognition.
... It is hard to make sense of such findings when treating the individual as the unit of analysis. Rather, they suggest that confidence judgments can be recast as processes to generate outputs for conspecifics-to facilitate collective behavior-and some authors have already endorsed this way of understanding metacognition (Shea et al., 2014). This special issue asks whether it is time to think about many other cognitive processes in this way. ...
To introduce our special issue How Minds Work: The Collective in the Individual, we propose “radical CI,” a form of collective intelligence, as a new paradigm for cognitive science. Radical CI posits that the representations and processes necessary to perform the cognitive functions that humans perform are collective entities, not encapsulated by any individual. To explain cognitive performance, it appeals to the distribution of cognitive labor on the assumption that the human project runs on countless interactions between locally acting individuals with specialized skills that each retain a small part of the relevant information. Some of the papers in the special issue appeal to radical CI to account for a variety of cognitive phenomena including memory performance, metacognition, belief updating, reasoning, and problem‐solving. Other papers focus on the cultural and institutional practices that make radical CI possible.
... However, our results show that it is possible for some kinds of uncertainty to influence 'control' behaviours (like information seeking) while circumventing subjective confidence, which may put pressure on the idea that the function of subjective confidence is to enable these control processes. It may be that many diverse aspects of cognition and action could be regulated within a single mind via implicit uncertainty representations, but that explicit feelings like subjective confidence instead play an essential role in sharing uncertainty among multiple interacting agents (Bahrami et al., 2010;Shea et al., 2014). ...
Humans and other creatures seek information to improve their cognition and behaviour. Theories in cognitive neuroscience, developmental psychology and animal cognition tend to assume a strong connection between information seeking behaviour and explicit metacognition-conscious introspection about our mental states and subjective metacognitive feelings like confidence or uncertainty. However, recent developments in computational neuroscience have stressed that metacognition and uncertainty are not equivalent, and many forms of uncertainty may be monitored in the brain without generating subjective metacognitive feelings. Here, across a series of experiments in adult humans, we show that information seeking and subjective confidence are controlled by distinct forms of uncertainty. In particular, information seeking (but not confidence) is controlled by uncertainty in sampled sensory evidence while confidence (but not information seeking) is controlled by uncertainty caused by decision boundaries. This double dissociation suggests that separate computations in the mind and brain shape confidence and information seeking: undermining the idea that information seeking behaviour always depends on conscious introspection into our own states of mind.
... The success of this model, based on implicit biomimetic imperatives, suggests that the structure learning problem is actually a structure teaching problem, in which agents can only learn if they are presented with material in the right way. Presenting material in the right way requires co-evolution of the teacher and learner of exactly the sort seen in evolutionary psychology and (cultural) niche construction [92][93][94][95]. If we read natural selection as nature's way of implementing Bayesian model selection [96][97][98][99], the emerging picture is a series of nested free energy minimising-i.e., evidence maximising-processes. On this view, protocols like the above teaching curriculum are themselves selected for structure learning via model selection. ...
This paper concerns structure learning or discovery of discrete generative models. It focuses on Bayesian model selection and the assimilation of training data or content, with a special emphasis on the order in which data are ingested. A key move-in the ensuing schemes-is to place priors on the selection of models, based upon expected free energy. In this setting, expected free energy reduces to a constrained mutual information , where the constraints inherit from priors over outcomes (i.e., preferred outcomes). The resulting scheme is first used to perform image classification on the MNIST dataset to illustrate the basic idea, and then tested on a more challenging problem of discovering models with dynamics, using a simple sprite-based visual disentanglement paradigm and the Tower of Hanoi (cf., blocks world) problem. In these examples, generative models are constructed autodidactically to recover (i.e., disentangle) the factorial structure of latent states-and their characteristic paths or dynamics.
... The ability to reflect on one's own cognitive processes is known as metacognition (Nelson and Narens 1990). Metacognition is suggested as an evolutionary advantage of human beings (Shea et al. 2014). In the context of perception, empirical evidence has indicated that human metacognition is dependent on the activity of the prefrontal cortex (Fleming et al. 2010, Morales et al. 2018, Shekhar and Rahnev 2018, Zheng et al. 2021) and can guide learning (Guggenmos et al. 2016), cognitive offloading (Gilbert et al. 2020), information seeking (Desender et al. 2018), and social interactions (Bahrami et al. 2010, Pescetelli andYeung 2021). ...
Recent evidence shows that people have the meta-metacognitive ability to evaluate their metacognitive judgments of confidence. However, it is unclear whether meta-metacognitive judgments are made by a different system and rely on a separate set of computations compared to metacognitive judgments. To address this question, we asked participants (N = 36) to perform a perceptual decision-making task and provide (i) an object-level, Type-1 response about the identity of the stimulus; (ii) a metacognitive, Type-2 response (low/high) regarding their confidence in their Type-1 decision; and (iii) a meta-metacognitive, Type-3 response (low/high) regarding the quality of their Type-2 rating. We found strong evidence for the existence of Type-3, meta-metacognitive ability. In a separate condition, participants performed an identical task with only a Type-1 response followed by a Type-2 response given on a 4-point scale. We found that the two conditions produced equivalent results such that the combination of binary Type-2 and binary Type-3 responses acts similar to a 4-point Type-2 response. Critically, while Type-2 evaluations were subject to metacognitive noise, Type-3 judgments were made at no additional cost. These results suggest that it is unlikely that there is a distinction between Type-2 and Type-3 systems (metacognition and meta-metacognition) in perceptual decision-making and, instead, a single system can be flexibly adapted to produce both Type-2 and Type-3 evaluations recursively.
Many agree that the more we feel that we can handle a given situation, the less afraid we are. But why? Is the situation no longer dangerous or is fear a response to more than danger? Here I analyze situations in which one reacts in cold blood to danger and argue that the formal object of fear is not the dangerous, but the unsafe. The unsafe indicates not only how the world is, but also how it can be handled. Safety, and its negative counterpart, are characterized by their duality, both evaluative (is the snake dangerous?) and agentive (is it under control?).
To better understand the nature of joint expertise and its underlying processes, we need not only analyses of the general conditions for skilled group action, but also descriptive accounts of the features and dimensions that vary across distinct performances and contexts, such as sport and the arts. And in addition to positioning our accounts against current models of individual skill, we need concepts and lessons from work on collaborative processes in other cognitive domains. This paper examines ecological or situational components of expert joint
action in practice, then offers a selective survey of some key cognitive and affective resources that shape and transform group performance.
Error monitoring is an essential human ability underlying learning and metacognition. In the time domain, humans possess a remarkable ability to learn and adapt to temporal intervals, yet the neural mechanisms underlying this are not clear. Recently, we demonstrated that humans improve sensorimotor time estimates when given the chance to incorporate previous trial feedback ( Bader and Wiener, 2021), suggesting that humans are metacognitively aware of their own timing errors. To test the neural basis of this metacognitive ability, human participants of both sexes underwent fMRI while they performed a visual temporal reproduction task with randomized supra-second intervals (1.5-6 s). Crucially, each trial was repeated following feedback, allowing a "re-do" to learn from the successes or errors in the initial trial. Behaviorally, we replicated our previous finding of improved re-do trial performance despite temporally uninformative (i.e., early or late) feedback. For neuroimaging, we observed a dissociation between estimating and reproducing time intervals. Estimation engaged the default mode network (DMN), including the superior frontal gyri, precuneus, and posterior cingulate, whereas reproduction activated regions associated traditionally with the "timing network" (TN), including the supplementary motor area (SMA), precentral gyrus, and right supramarginal gyrus. Notably, greater and more extensive DMN involvement was observed in re-do trials, whereas for the TN, it was more constrained. Task-based connectivity between these networks demonstrated higher inter-network correlation primarily when estimating initial trials, while re-do trial communication was higher during reproduction. Overall, these results suggest that the DMN and TN jointly mediate subjective self-awareness to improve timing performance.
Metacognition relies on representations, which are commonly viewed as internal knowledge structures. We explore the idea that external embodied diagrams, rather than monomodal symbolic-based entities, can provide an accurate description of metacognition in choreographic dance. When marking, dancers use their bodies to represent properties, dynamics, or structures of dance phrases. Marking-for-self occurs when a dancer marks the dance in their own unique manner, potentially allowing for real-time reflection through the manipulation of external signs. These manipulations can be considered diagrammatic semiosis, as diagrams signify shared relational parts analogous to the parts of their objects. Based on Peirce’s semiotics, we argue that semiosis is crucial for metacognition. This paper is structured to first review marking and marking-for-self, followed by an overview of metacognition. We then define marking-for-self as a metacognitive phenomenon and introduce Peirce’s concept of semiosis and diagrams. Finally, we describe marking as a diagrammatic sign and explain how marking-for-self is an embodied metacognitive process achieved through the manipulation of diagrammatic structures.
Much effort has been made to understand the role of attention in perception; much less effort has been placed on the role attention plays in the control of action. Our goal in this chapter is to account for the role of attention in action, both when performance is automatic and when it is under deliberate conscious control. We propose a theoretical framework structured around the notion of a set of active schemas, organized according to the particular action sequences of which they are a part, awaiting the appropriate set of conditions so that they can become selected to control action. The analysis is therefore centered around actions, primarily external actions, but the same principles apply to internal actions—actions that involve only the cognitive processing mechanisms. One major emphasis in the study of attentional processes is the distinction between controlled and automatic processing of perceptual inputs (e.g., Shiffrin & Schneider, 1977). Our work here can be seen as complementary to the distinction between controlled and automatic processes: we examine action rather than perception; we emphasize the situations in which deliberate, conscious control of activity is desired rather than those that are automatic.
Early studies of intuitive judgment and decision making conducted with the late Amos Tversky are reviewed in the context of two related concepts: an analysis of accessibility, the ease with which thoughts come to mind; a distinction between effortless intuition and deliberate reasoning. Intuitive thoughts, like percepts, are highly accessible. Determinants and consequences of accessibility help explain the central results of prospect theory, framing effects, the heuristic process of attribute substitution, and the characteristic biases that result from the substitution of nonextensional for extensional attributes. Variations in the accessibility of rules explain the occasional corrections of intuitive judgments. The study of biases is compatible with a view of intuitive thinking and decision making as generally skilled and successful.
Book synopsis: The neural computational approach to cognitive and psychological processes is relatively new. However, Neural Computation and Psychology Workshops (NCPW), first held 16 years ago, lie at the heart of this fast-moving discipline, thanks to its interdisciplinary nature — bringing together researchers from different disciplines such as artificial intelligence, cognitive science, computer science, neurobiology, philosophy and psychology to discuss their work on models of cognitive processes.
Once again, the Eleventh Neural Computation and Psychology Workshop (NCPW11), held in 2008 at the University of Oxford (England), reflects the interdisciplinary nature and wide range of backgrounds of this field. This volume is a collection of peer-reviewed contributions of most of the papers presented at NCPW11 by researchers from four continents and 15 countries.