Biological Mechanisms Are Necessary for Consciousness, Not Just Computation

Key Quotes

"It's become increasingly clear that the Turing Test -- determining whether human interlocutors can tell whether a conversation is being carried out by a human or a machine -- is not a good way to think about consciousness. Modern LLMs can mimic human conversation with extraordinary verisimilitude, but most people would not judge them to be conscious."

Sean Carroll explains that the Turing Test, which focuses on output and conversational ability, is insufficient for determining consciousness. He notes that current large language models (LLMs) can convincingly imitate human conversation but are generally not considered conscious. This highlights a shift in thinking away from purely behavioral tests for consciousness.

"There's a point of view that really puts the emphasis on kind of an input output mechanism this would go back to the Turing test with Alan Turing right Turing suggested that if you had a computer program that could have a conversation with a human and trick them into thinking that it was conscious then it should count as conscious what really matters in other words is the output of the computation going on and this grew into a view called computational functionalism."

Sean Carroll introduces the concept of computational functionalism, which posits that consciousness is determined by the computational processes and the input-output mechanisms of a system. This view, rooted in Turing's ideas, suggests that the function and computation are what matter, regardless of the underlying physical implementation. Carroll indicates he is moving away from this perspective.

"Ned Block has long argued that consciousness involves something more than simply the 'functional' aspects of inputs and outputs."

Sean Carroll states Ned Block's long-held position that consciousness is not solely defined by functional aspects like input and output. This suggests that Block believes there are additional, perhaps deeper, components to consciousness beyond what computational functionalism accounts for. This sets the stage for Block's alternative views on the requirements for consciousness.

"Ned Block has long argued that consciousness involves something more than simply the 'functional' aspects of inputs and outputs. Ned is actually a super well respected philosopher in the field of consciousness. I quote him in the big picture mentioning his distinction between access consciousness which is what uh more or less what David Chalmers classifies as the easy problem it's sort of your ability to access different pieces of information globally in your cognition versus phenomenal consciousness which is the feeling of experiencing something and that is what is hard to explain."

Sean Carroll highlights Ned Block's significant contributions to the study of consciousness, particularly his distinction between access consciousness and phenomenal consciousness. Carroll explains that access consciousness relates to the global availability of information for cognitive processes (the "easy problem"), while phenomenal consciousness refers to the subjective experience or "what it's like" to feel something (the "hard problem"). This distinction is foundational to understanding Block's arguments.

"So Ned wants to argue that maybe at least he's very open minded he's a good philosopher is like just suggesting possibilities we should take seriously maybe these things that we think of as experiences of conscious states have something to do with the subconscious processes that are going on in our biological manifestation or instantiation and maybe therefore you could build a computer program that was arbitrarily good at tricking you at giving all the output that you might expect a conscious creature to give you and nevertheless it would not qualify as what we think of as conscious."

Sean Carroll discusses Ned Block's hypothesis that conscious experiences might be linked to subconscious processes within biological organisms. Block suggests that a computer program, even one perfectly mimicking conscious output, might not be truly conscious if it lacks these underlying biological or subconscious processes. Carroll finds this possibility compelling, indicating a departure from purely functionalist views.

"So the idea of the sales pitch i guess for computational functionalism would be look the the brain clearly computes some things you can at some level think of how you communicate with the human being as it gets some input it gives some output clearly there is a computation underlying that and the computational functionalist view is that is what it is there's nothing really extra going on."

Sean Carroll outlines the core argument for computational functionalism, which posits that the brain's processes are fundamentally computational. This perspective suggests that consciousness arises directly from these computations, with inputs and outputs being the primary focus. The view implies that there is no additional element beyond the computational function itself that constitutes consciousness.

"Yeah so what i think is if you want the machine to be conscious you may need a certain kind of implementation of those computations."

Ned Block, as relayed by Sean Carroll, proposes that for a machine to achieve consciousness, the specific way computations are implemented might be crucial. This suggests that simply performing the correct computations is not enough; the underlying physical or biological substrate and its mechanisms could be essential for consciousness. This challenges the purely functionalist view that implementation details are irrelevant.

"So the question is whether consciousness is like that and i think we we just don't know so the turing thesis doesn't help us."

Ned Block, as discussed by Sean Carroll, questions whether consciousness is analogous to processes like gravity or rainstorms, where a simulation does not produce the actual phenomenon. He suggests that consciousness might be an intrinsic property tied to specific implementations, rather than something that can be replicated through mere computation. Block concludes that the Turing thesis, focusing on computability, does not resolve this fundamental question.