Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5

Is there a symmetry between metacognition & mindreading (theory of mind)?

#1
C C Offline
https://aeon.co/essays/is-there-a-symmet...indreading

EXCERPTS (Stephen M Fleming): . . . A common view is that self-knowledge is special, and immune to error, because it is gained through introspection – literally, ‘looking within’. While we might be mistaken about things we perceive in the outside world (such as thinking a bird is a plane), it seems odd to say that we are wrong about our own minds. If I think that I’m feeling sad or anxious, then there is a sense in which I am feeling sad or anxious. We have untrammelled access to our own minds, so the argument goes, and this immediacy of introspection means that we are rarely wrong about ourselves.

This is known as the ‘privileged access’ view of self-knowledge, and has been dominant in philosophy in various guises for much of the 20th century. René Descartes relied on self-reflection in this way to reach his conclusion ‘I think, therefore I am,’ noting along the way that: ‘I know clearly that there is nothing that can be perceived by me more easily or more clearly than my own mind.’

An alternative view suggests that we infer what we think or believe from a variety of cues – just as we infer what others think or feel from observing their behaviour. This suggests that self-knowledge is not as immediate as it seems. For instance, I might infer that I am anxious about an upcoming presentation because my heart is racing and my breathing is heavier. But I might be wrong about this – perhaps I am just feeling excited. This kind of psychological reframing is often used by sports coaches to help athletes maintain composure under pressure.

The philosopher most often associated with the inferential view is Gilbert Ryle, who proposed in The Concept of Mind (1949) that we gain self-knowledge by applying the tools we use to understand other minds to ourselves: ‘The sorts of things that I can find out about myself are the same as the sorts of things that I can find out about other people, and the methods of finding them out are much the same.’ Ryle’s idea is neatly summarised by another New Yorker cartoon in which a husband says to his wife: ‘How should I know what I’m thinking? I’m not a mind reader.’

Many philosophers since Ryle have considered the strong inferential view as somewhat crazy, and written it off before it could even get going. [...] But if we take a longer view of history, the idea that we have some sort of special, direct access to our minds is the exception, rather than the rule...

Other aspects of the mind – most famously, perception – also appear to operate on the principles of an (often unconscious) inference. The idea is that the brain isn’t directly in touch with the outside world (it’s locked up in a dark skull, after all) – and instead has to ‘infer’ what is really out there by constructing and updating an internal model of the environment, based on noisy sensory data...

[...] Elegant evidence for this perception-as-inference view comes from a range of striking visual illusions. [...] These striking illusions are simply clever ways of exposing the workings of a system finely tuned for perceptual inference. And a powerful idea is that self-knowledge relies on similar principles – whereas perceiving the outside world relies on building a model of what is out there, we are also continuously building and updating a similar model of ourselves – our skills, abilities and characteristics. And just as we can sometimes be mistaken about what we perceive, sometimes the model of ourselves can also be wrong.

Let’s see how this might work in practice. [...] We now know from laboratory research that metacognition, just like perception, is also subject to powerful illusions and distortions – lending credence to the inferential view. [...] In our research, we have also found surprisingly large and consistent differences between individuals on these measures – one person might have limited insight into how well they are doing from one moment to the next, while another might have good awareness of whether are likely to be right or wrong.

This metacognitive prowess is independent of general cognitive ability, and correlated with differences in the structure and function of the prefrontal and parietal cortex. In turn, people with disease or damage to these brain regions can suffer from what neurologists refer to as anosognosia – literally, the absence of knowing. For instance, in Alzheimer’s disease, patients can suffer a cruel double hit – the disease attacks not only brain regions supporting memory, but also those involved in metacognition, leaving people unable to understand what they have lost.

This all suggests – more in line with Socrates than Descartes – that self-awareness is something that can be cultivated, that it is not a given, and that it can fail in myriad interesting ways. And it also provides newfound impetus to seek to understand the computations that might support self-awareness. This is where Premack and Woodruff’s more expansive notion of theory of mind might be long overdue another look.

Saying that self-awareness depends on similar machinery to theory of mind is all well and good, but it begs the question – what is this machinery? What do we mean by a ‘model’ of a mind, exactly?

[...] There is now tantalising evidence that similar types of brain cell also encode abstract conceptual spaces. For instance, if I am thinking about buying a new car, then I might think about how environmentally friendly the car is, and how much it costs. These two properties map out a two-dimensional ‘space’ on which I can place different cars – for instance, a cheap diesel car will occupy one part of the space, and an expensive electric car another part of the space. The idea is that, when I am comparing these different options, my brain is relying on the same kind of systems that I use to navigate through physical space.

[...] So far, these lines of work – on abstract conceptual models of the world, and on how we think about other minds – have remained relatively disconnected, but they are coming together in fascinating ways...

[...] Recently, the psychologist Anthony G Vaccaro and I surveyed the accumulating literature on theory of mind and metacognition, and created a brain map that aggregated the patterns of activations reported across multiple papers. Clear overlap between brain activations involved in metacognition and mindreading was observed in the medial PFC. This is what we would expect if there was a common system building models not only about other people, but also of ourselves – and perhaps about ourselves in relation to other people. Tantalisingly, this very same region has been shown to carry grid-like signatures of abstract, conceptual spaces.

At the same time, computational models are being built that can mimic features of both theory of mind and metacognition. These models suggest that a key part of the solution is the learning of second-order parameters – those that encode information about how our minds are working, for instance whether our percepts or memories tend to be more or less accurate. Sometimes, this system can become confused...

[...] Another implication of a symmetry between metacognition and mindreading is that both abilities should emerge around the same time in childhood. By the time that children become adept at solving false-belief tasks – around the age of four – they are also more likely to engage in self-doubt, and recognise when they themselves were wrong about something...

[...] Indeed, when Simon Baron-Cohen, Alan Leslie and Uta Frith outlined their influential theory of autism in the 1980s, they proposed that theory of mind was only ‘one of the manifestations of a basic metarepresentational capacity’. The implication is that there should also be noticeable differences in metacognition that are linked to changes in theory of mind. In line with this idea, several recent studies have shown that autistic individuals also show differences in metacognition. And in a recent study of more than 450 people, Elisa van der Plas, a PhD student in my group, has shown that theory of mind ability (measured by people’s ability to track the feelings of characters in simple animations) and metacognition (measured by the degree to which their confidence tracks their task performance) are significantly correlated with each other. People who were better at theory of mind also formed their confidence differently – they were more sensitive to subtle cues, such as their response times, that indicated whether they had made a good or bad decision.

Recognising a symmetry between self-awareness and theory of mind might even help us understand why human self-awareness emerged in the first place. The need to coordinate and collaborate with others in large social groups is likely to have prized the abilities for metacognition and mindreading. [...] A symmetry between self-awareness and other-awareness also offers a subversive take on what it means for other agents such as animals and robots to be self-aware... (MORE - missing details)
Reply




Users browsing this thread: 1 Guest(s)