Comments on: Against functionalism: why I think the Foundational Research Institute should rethink its approach https://archive.opentheory.net/2017/07/why-i-think-the-foundational-research-institute-should-rethink-its-approach/ Speculations on the Frontiers of Science and Culture Sat, 27 May 2023 12:50:11 +0000 hourly 1 https://wordpress.org/?v=7.0 By: MJ https://archive.opentheory.net/2017/07/why-i-think-the-foundational-research-institute-should-rethink-its-approach/#comment-99752 Fri, 07 Dec 2018 10:27:01 +0000 https://archive.opentheory.net/?p=1193#comment-99752 In reply to Sahil.

Thanks for the thoughtful comment. For a general intro to our work on this I’d recommend these two videos: https://qualiaresearchinstitute.org/2018/04/13/videos-from-tsc2018/ – for a more in-depth look, Principia Qualia (https://archive.opentheory.net/PrincipiaQualia.pdf) and Quantifying Bliss (https://qualiacomputing.com/2017/06/18/quantifying-bliss-talk-summary/).

In short, we believe we can generate novel predictions with the combination of Qualia Formalism and the intuition that symmetry in the formalism corresponds to something interesting. More specifically, we think it corresponds to emotional valence, or how pleasant or unpleasant an experience feels). Principia Qualia goes into the philosophical assumptions here, whereas Quantifying Bliss talks about combining this together with Selen Atasoy’s work on connectome-specific harmonic waves to generate specific novel predictions about quantifying pain and pleasure using fMRI.

We’re mostly agnostic on how to generate the formalism; all we’re really assuming is that it exists (that there is in fact an isomorphism between the contents of a phenomenological experience and some mathematical object).

]]>
By: Sahil https://archive.opentheory.net/2017/07/why-i-think-the-foundational-research-institute-should-rethink-its-approach/#comment-99750 Fri, 07 Dec 2018 09:01:11 +0000 https://archive.opentheory.net/?p=1193#comment-99750 In my mind, the problem with finding a mathematical “isomorphism”* is that usually to take an analogy (mathematical or otherwise) seriously, it has to be corroborated. But there is no way _in principle_ to check truly novel predictions (AFAICT). At best we may trust subjective self-reports that we can understand, which would likely be only a narrow, anthropo-/bio- centric subset of all possible experience, assuming there is such a thing.

Aaronson makes similar remarks in his posts on IIT, which I’m sure you’ve seen.

Is there a specific link I could follow for your thoughts on this?

*please no! ‘Isomorphisms’ are between mathematical objects. I suggest the word ‘model’. Using the word isomorphism takes away from formalism – isomorphism means it preserves all the properties you care about at the moment, and this is at best passing the buck as to what properties are enough to really call it the “same” thing, and at worst precisely functionalism.

]]>