A Theory Of Mind Through Simulation

For those of you who have followed my comments here and there and the Free Will series I’ve been doing, you know that I’ve been talking about a theory of mind where we experience the world through a simulation that is running in our brains. It turns out that Alva Noe` appears to be saying exactly the same thing – but using different words. To wit, I give you himself:



Alva gives another talk on color perception here

For me colors are like shapes. Just as a three-dimensional shape has a hidden backside so colors have hidden ways they would look if the conditions of lighting were changed.

He doesn’t come out and speak about the simulation running in our brains, but he describes it.


  1. Digesting rather than reading..defining experience. Look into my eyes and hear what i am saying.

    • Let me hear how you feel? Let me touch how you taste…. perspective is everything, perception is understanding.

      • Indeed. i am now very intrigued, craving more. A blank slate. i do believe i ill immerse myself in this theory, for now.

        • I shall try to keep it entertaining

  2. So if someone has synesthesia….

    Question. Recently, I said the following:

    “The less probable it is that an account or body of ideas originated with man, the more probable it is that it DIDN’T originate with man.”

    You said you disagreed. I’m curious to know why.

    • Wow, pick stuff out of the blue? Okay, I’ll have a shot at this without any context.

      The first problem is basing the presumptions on humans. Outside of humans there is only other animal life. The less probable that a body of ideas originated with man the more probable it originated with animals that are are not human. This is not more probable. Wee have only evidence for life on this planet. The boundaries of your presuppositions almost always include a magic man in the sky. I have no reason to do so, nor evidence to believe it possible.

  3. I think part of what Noë is getting at here is the inadequacy of a representationalist approach to cognition. In fact, Noë is at least tangentially involved in a movement in philosophy and cognitive science called radical embodied cognitive science, which holds that computational and representational theories of cognition don’t cut the mustard, and instead seek answers in dynamical systems theory.

    I’m on the fence about it, myself. Dynamical systems theory definitely has shown more success in some areas (such as building robots that can move like actual organisms), but fails in others (such as playing chess, where a computationalist approach is much more successful). Honestly I expect that both the embodied/nonrepresentational approach and the computationalist/symbolic approach are partially right and partially wrong.

    • I’m not fully up on but dynamical systems theory/chaos theory has a chance. I see both models applied as if they are simply too narrowly focused to tackle the whole problem.

      With child development, we see the human brain building rules for the simulation. With a very poor simulation running, context switching and stitching two simulation events together takes learning. The a-not-b error as an example. You can watch children building rules for their simulation. When learning, it takes a bit to understand that each action is independent and outcomes can change. This requires more rules than figuring out a simple task when the same task is applied in multiple ways in the same environment. It’s a complex behavior that requires continuous cause and effect prediction.

      Cognition is not merely computational or representational but a mix where rules based computational short cuts lessen energy requirements to get it right most of the time. There is no reason to think that a human child should be fully cognitive, nor a human adult for that matter… at least where fully cognitive is some measurable state. There are many people that ‘patch’ the rules in their simulation of the world so that they no longer question things like how cars work, or tv sets. These are just objects with special characteristics that follow specific rule patterns. This abstracts a complex thing to a simple set of rules. By rules I mean that cars don’t float in the air or go sideways and televisions don’t lay down at night or cook food. By adopting these simpler rules of behavior for complex objects, we can more quickly assimilate them into the simulation successfully. Contrast this to getting a new remote control for the cable box. This is not quickly assimilated and sometimes only barely assimilated because the rules remain more complex for interaction in the simulation.

      The topic is quite large for a comment, but I can see all the learning issues as building rules sets and improving computational skills while learning also the parameters of representational/symbolic relationships. It’s clearly not a discussion started and finished over one glass.

  4. I have really enjoyed your thoughts on this.

    • holly,
      Thanks for commenting, and for your kind words.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: