Our new paper "Intrinsic Computational Functionalism (ICF) and Simulated Consciousness" is out (with @Shuqin_Ma).
https://t.co/e3MfNAQC5C
In this paper, we show that if we accept the assumptions of ICF, a simulation of a conscious system should also be conscious.
Another preprint from us "Intrinsic Computational Functionalism: From Observer-Relative Maps to Observer-Independent Structures" with @Shuqin_Ma
Intrinsic Computational Functionalism (ICF) described in this paper is very close to my current philosophical position on consciousness.
Although computational functionalism has many problems, I endorse this intrinsic version where the main idea is that if consciousness is computationally constituted, it must depend on computational structures that are physically realized and intrinsic to the system itself.
I think this is an important viewpoint when we consider the possibility of AI consciousness.
Our other recent work on canonical functionalism is one example of this direction, but only captures just one aspect of the larger framework of ICF.
https://t.co/HFTKtZex0e
The paper proposes two criteria:
C1: System-intrinsic instantiation
The relevant computational property must not depend on an external observer’s labels. It should be invariant under structure-preserving relabellings of the system’s variables.
C2: Causal-dynamical organization under intervention
The property must be grounded in the system’s own causal state-space structure, revealed by how its variables mutually constrain one another under intervention.
The goal is to move beyond both naive computationalism and anti-computationalism and try to understand what kinds of physical-computational organization might be important for consciousness.
I have also additional thoughts on how this might be relevant for IIT's implication that computer-based simulations of consciousness cannot be conscious. But this is a topic for the next paper.
https://t.co/GB0bnfJnlm
@Spectrum_cj The Consciousness of Neuroscience
Alex Gomez-Marin, eNeuro, 2023 のSignificance Statementに、「ask not what neuroscience can do for consciousness but what consciousness can do for neuroscience」と言っています。