Videos

S04E06 Andy Clark: The Experience Machine



Mind Chat

Andy Clark is leading and incredibly influential philosopher and cognitive scientist. Among other things, Andy has done pioneering work on predictive processing and the extended mind thesis. Here Keith and Philip discuss with Andy his most recent book ‘The Experience Machine.’ https://www.theguardian.com/books/2023/may/04/the-experience-machine-how-our-minds-predict-and-shape-reality-review

Source

Similar Posts

26 thoughts on “S04E06 Andy Clark: The Experience Machine”
  1. Predictive processing fails to explain why there are mistakes in estimations. What happens when predictative processing does not solve a problem? A new set of predictions are made? How do we use the errors in predictions? Are prediction errors useful experiences as well?

  2. Quantitative descriptions of the contents of experience simply do not provide any of the qualitative descriptions of experience. And regardless of what the nature of those qualitative descriptions might be, there is no explanation of how we could get from the quantitative to the qualitative.

  3. The Experience Machine is the wrong title. This is not a theory of Conscious Experience, but it is a theory of Brain function and therefore it is only about the Neural Correlates of Consciousness. Please, will someone show me how this theory Explains any Conscious Experience? What is the Experience of the Redness of Red, the Sound of the Standard A Tone, the Taste of Salt, the Smell of Bleach, the Touch of a Rough Surface, etc.?

  4. Re a theory not being an instantiation of itself, source code isn't an instance of a running program; for that, you need a physical device, about which there is a ton of other information! So that argument seems to fail.
    Surprised that Andy didn't mention the work of John Vervaeke re relevance realization as a core function of mind…

  5. I'm always confused by what Phillip Goff finds salient. Does he not know the difference between psychological concepts and culturally transmitted language? We can't build predictive concepts without learned data and then connect it to how other people are using words (with some Bayesian blurring because it would be impossible to only work with specific instances). Like the multimodal compressions that Lisa Feldman Barrett or Lawrence Barsalou talks about.

  6. Keith thanks for unequivocally saying Consciousness is not an illusion at 1:05:00, because "Consciousness is illusion" is what ends up as being your's and Dan Dennet's position by the detractors – sometimes intentionally and some times mischievously. I think what illusion-ism means is that it is an illusion to which we attribute consciousness – in other words something extra sprinkle as was discussed. The key point and in my opinion a simple point is that a normally functioning brain is needed to register our own consciousness. It is true that the very precise mechanism is being worked on and not yet fully knows but we are getting there thru the works of Andy and Anil and Michael Graziano. I sometimes give this example:

    1. in the distant past people thought that sickness is causes by spells, and wraths of deities
    2. later people realized that contact with an already sick person gives us the sickness.
    3. later we found out the sickness comes from microorganisms
    4. we now know the molecular and chemical basis of how viruses and bacteria disrupt the cellular chemistry and cause sickness
    5. After this point it is no longer biology and the analysis need to turn into chemistry…and eventually physics

    But level 4 satisfies our theory of infectious deceases.

    IMO for consciousness we would like to get to 4 and on that scale we are at 3.8+. The AIs, because we can invasively and by instrumentation look inside them are going to push us over the hump in near future. Of course it will be the understanding of an abstract notion of consciousness slightly dissimilar to our consciousness in precise details but with the same essence.

    Excellent episode.

  7. Uh oh…Philip brings up knowledge argument – which Frank Jackson himself has disavowed. The knowing has two meanings – know = understand is not same as knows = experience. I cannot believe Philip still keeps bringing it up.

    General relativity does not need to know about every orbit of every planet, but it is enough that it predicts correctly how planets will orbit a star. If it does not enumerate orbits every planets – that have existed, exist and will exist that the theory is incomplete.

  8. Nested intelligence – Robots made of robots made of robots – by Dan D. IMO we have not fully realized the significance of Dan's thinking. Global workspace – fame in the brain and so many.

  9. I wonder what the effect of childhood trauma on predictive processing is. In that early developmental stage something disrupts the development of the predictive processing device in our minds and changes how we predict the present for life… something like that. Thoughts?

  10. Hi Philip and Keith. Having recently discovered your channel I'm watching it from the start. I'm half way.
    I'm trying to think about what you're talking about. Is it meaningful to suggest the following:
    It seems as though you both don't distinguish experience and the world, in a certain sense (that is curious and interesting, as someone who perhaps has spent time in a kind of Cartesian-induced quasi-solipsistic space, banging my head against the conceptual problem of other minds, wishing I could perceive other minds.) When I and another look upon and perceive a scene, from different vantage points, naively what is the relation between the two EXPERIENTIAL scenes? Is it possible that for you both the right phenomenology has built deeply in that something is shared, that there is a difference of access point, but to a shared 'reality', and that your views diverge on the metaphysical nature of the reality but coincide as to the availability to both parties of this single reality? For you Keith from here the project is to locate experience in the world, for you Philip, to locate the world in experience. Would you say this very crude sketch is at all meaningful? What struck me as curious when I had the thought was the question of the phenomenological flavour of the world as experienced against the background assumption (or lack of doubt) as to the coincidence of the two minds and worlds.

  11. A bit of a shame you didn't ask him about his background and growing up how he came to his beliefs. Normally I like to go straight into the meat of it but this time would be interesting

  12. Andy Clark's framing of the mind as a prediction machine rather than a passive receiver really shifts how you think about perception itself. The way Philip and Keith push back and engage with his ideas here keeps the conversation grounded rather than letting it float off into pure abstraction.

Comments are closed.

WP2Social Auto Publish Powered By : XYZScripts.com