Is consciousness in control? Does it matter?

Self-portrait as an Epiphenomenon

(Photo credit: alexschlegel)

A new scientific paper claims to describe an experiment that shows that consciousness controls our actions.  From the paper:

These results indicate that conscious intentions govern motor function… until today, it was unclear whether conscious motor intention exists prior to movement, or whether the brain constructs such an intention after movement initiation.

The Neuroskeptic takes a look at this study, and in the process succinctly describes the debate.

To simplify, one school of thought holds that (at least some of the time), our intentions or plans control our actions. Many people would say that this is what common sense teaches us as well.

But there’s an alternative view, in which our consciously-experienced intentions are not causes of our actions but are actually products of them, being generated after the action has already begun. This view is certainly counterintuitive, and many find it disturbing as it seems to undermine ‘free will’.

He then goes on to show that the study has a significant design flaw that calls its conclusions into question.  You can read the details in his post.

But I want to back up a bit and ask a question.  Why does it matter?

To illustrate why I’m asking, consider the proposed sequence of the two debated scenarios.

Scenario 1: Consciousness controls actions:

  1. You consciously decide what to do.
  2. You do it.
  3. You have conscious and unconscious knowledge of 2 and how it turned out.
  4. Loop back to step 1.

Scenario 2: Consciousness does not control actions:

  1. You unconsciously decide what to do.
  2. You do it.
  3. You have conscious knowledge (at least sometimes) of the results of 2.
  4. The information in 3 is available to the unconscious parts of your brain.
  5. Loop back to step 1.

I think most people agree that scenario 2 happens at all the time.  For example, we usually don’t consciously think about walking or driving to work, or striking each key on a keyboard when writing an email.  The question is whether scenario 1 ever happens.

But again my question is, does it matter?  Look again at the sequences.  What changes if scenario 1 or scenario 2 are happening?  Isn’t consciousness still having a causal effect on actions in scenario 2, albeit a delayed one?

Maybe the real distinction is how often and how early step 3 in scenario 2 happens?  I think there’s no question that it varies depending on the situation.  I presented the scenarios above as two discrete possibilities, but I suspect the reality is more of a spectrum, with various actions arising with varying frequencies into consciousness.

Many people who are overly impressed for evidence against scenario 1 think that consciousness is an epiphenomenon, a side effect with no causal power.  In other words, in scenario 2, they omit step 4 completely.  But if you think about it, that makes no evolutionary sense.  Why would an awareness mechanism evolve with no causal influence?

Even if consciousness doesn’t directly pull the action levers, it still provides information to the systems that do.  It has to at least have a causal influence on the communication modules in the brain, otherwise we wouldn’t even be able to discuss it.  And given that communication requires movement (of vocal cords, fingers on the keyboard, etc), there no reason to suspect its causal influence excludes the broader movement modules.

It is widely accepted that deliberations that involve consciousness are slower and more cognitively expensive than those which don’t.  This makes me more inclined to think that scenario 1, if it happens, is rare.  It’s why we train, so that our automatic unconscious decisions will be the right ones.  Because having to consciously process our actions takes more time.

What relevance does this have to the free will debate?  None as far as I can see.  It might if consciousness were in fact an epiphenomenon, but only if we narrowly defined the self as the conscious portion of the mind.  Regardless, thoughts still arise from brain operations, from physical laws.  Libertarian free will still has no logical space to exist, and compatibilist free will remains a bigger picture perspective on the entire brain and society.

This entry was posted in Mind and AI and tagged , , , , , , . Bookmark the permalink.

9 Responses to Is consciousness in control? Does it matter?

  1. I think it matters for people who are invested in some notion of strong programmatic AI. Thinking of Searle’s “Minds, Brains, and Programs”, when the details of intentionality are left vague, the jump to “well it must be biological matter that makes the difference!” or “It’s some metaphysical abstraction to unity from complex individual parts” are really the only options unless you’ve got more to go on than a just “a directness” which could take a variety of forms.

    I don’t think it would matter for more of a machine learning style of AI though, but I’m out of my depths when it comes to this conversation.

    Like

    • Hmmm, not quite seeing the issue with strong AI. But then I long ago concluded that either Searle and like minded people see something I don’t, or they’re confused. Naturally I lean toward them being confused :-), but of course this might simply be me missing that key issue again.

      Like

      • I don’t think those are mutually exclusive (Searle et all seeing something others don’t and them being confused)!

        I also don’t think you’re missing a key issue – I think Searle’s biggest problem is that his version of intentionality amounts to “It just IS, ok??” and so anyone attempting to do strong AI through programming has to have a much more detailed version understanding of intentionality (when, where, how it works, etc) if they’re to overcome the challenge that you can’t program intentionality because it’s shrouded in mystical confusion.

        But I tend to also think Searle is seeing something that isn’t there (he says intentionality is bound to organic matter, but for all his clarity he might as well just call it a “soul” and be done with it) AND that programming intentionality is impossible, so maybe I’m the one who’s confused.

        Like

        • Actually, I think that is an insightful comment. It seems to me that what Searle can’t see (what he, perhaps, subconsciously won’t see) is that intentionality is programming. Learning of course makes a difference, but so do inputs to a program. Machine programming comes from a human programmer (mostly, for now) but organic programming comes from evolution.

          Like

  2. amanimal says:

    I’m confused – given:

    “… in scenario 2, they omit step 4 completely.”

    … did you mean “conscious” rather than “unconscious” in Step 4?

    Regardless, I’m pretty sure I don’t know enough to agree or disagree – maybe someday:

    ‘The New Unconscious’, Hassin, Uleman & Bargh (editors) 2005

    http://www.academia.edu/5378163/The_New_Unconscious

    So far I’ve found much of it fairly straightforward, but some parts are going to require a more fundamental understanding before they’re going mean much to me.

    Like

    • No, I actually mean “unconscious”. The idea being that your unconscious knows what your conscious knows and that information is available for your next action, even if your unconscious is the one doing it. Therefore your conscious knowledge of what happened, and how you consciously think about it, influences future actions. (Now that I write that, it occurs to me that I might should have explicitly stated it in the post.)

      BTW, finished ‘Who’s In Charge?’. I feel another consciousness post coming on.

      Like

      • amanimal says:

        Ahh, that clears it up, thanks. Congrats on ‘WiC?’, you were about due for a reminder :) I’ll look forward to that post!

        Like

  3. guymax says:

    This post is the most sensible thing I’ve read about consciousness for a long time. It seems spot on. This is an issue I hadn’t thought through before, so thanks for doing it for me.

    Like

Comments are closed.