Saturday 7 July 2012

Hakwan Lau: How to study the functions of subjective awareness?

      Abstract: I give some examples of subliminal priming studies suggesting that subjective awareness may not be as functionally powerful as we might think. However these studies as well as most others in the field suffer from a methodological problem: in rendering stimuli unconscious we substantially lower the relevant signal strength, so subliminal priming effects are invariably weak. It is a problem for a whole field to rely on weak effects because null results are hard to interpret and positive results are subject to selection and publication bias. So I propose a new approach to address this problem. The key is to keep signal strength / perceptual sensitivity constant while manipulating subjective awareness, and to see how that affects cognitive functions. This is hard to achieve but I show preliminary data demonstrating how this could be done.

      Unconscious activation of the cognitive control system in the human prefrontal cortex. Lau HC, Passingham RE. J Neurosci. 2007 May 23;27(21):5805-11.
      Empirical support for higher-order theories of conscious awareness
Hakwan Lau and David Rosenthal
      Subliminal stimuli in the near absence of attention influence top-down cognitive control.
Rahnev DA, Huang E, Lau H. Atten Percept Psychophys. 2012
      Does response interference depend on the subjective visibility of flanker distractors?
Maniscalco B, Bang JW, Iravani L, Camps-Febrer F, Lau H. Atten Percept Psychophys. 2012
      Attention induces conservative subjective biases in visual perception. Rahnev D, Maniscalco B, Graves T, Huang E, de Lange FP, Lau H. Nat Neurosci. 2011 Oct 23;14(12):1513-5
      Direct injection of noise to the visual cortex decreases accuracy but increases decision confidence. Rahnev DA, Maniscalco B, Luber B, Lau H, Lisanby SH. J Neurophysiol. 2012 Mar;107(6):1556-63.

Comments invited


  1. Lau suggests that if we look hard enough, we will inevitably find that any cognitive function can occur unconsciously. Just because we have not yet found one that is exclusive conscious does not mean that we never will.

    1. I'll tend to have the same reasoning as Hakwan Lau, mainly that any cognitive ability might be possible without consciousness. However, although removing consciousness for a single given cognitive ability at a time might be possible without necessarily lose the ability's efficiency, what about removing consciousness in all the cognitive abilities at the same time? I would highly doubt about the efficiency of such an individual to maintain his/her survival and allow his/her reproduction. What I mean is that if we have some evidence such as blindsight that consciousness is not necessarily needed for an object detection that is usually done by conscious visual stimuli, we cannot infere that consciousness is ablsolutely not necessary for species like ours who evolved with consciousness. To my knowledge, there's no example of an individual who is able to behave with a complete loss of consciousness. Actually, such individuals are simply in a comatose or a vegetable state. So yes, some cognitive abilities can occur without consciousness, but we have to keep thinking in terms of an individual as belonging to a specific species. If a human cannot behave without consciousness, therefore, consciousness is needed for a human's life and therefore, consciousness is at some point necessary for life. That's why I think using the robot argument to discredit the real utility (or adaptive function).
      Also, I think that there might have a cognitive function that is exclusively conscious and that might be the one that shut everything down when an individual fall in a comatose state...

    2. Very elegant designs, and I am curious to see findings from his TMS study. What I really appreciated about Lau's talk - within the broader context of the Summer Institute - was that he defined his working definition of consciousness (more or less "subjective awareness of relevant perceptual info"). If we were, instead, to take Pr. Harnad's consciousness = feeling definition, then the notion that a 'cognitive function that occurs unconsciously' would not make any sense (apart maybe from some cognitive functioning that occurs during sleep). His definition of the term spared me some confusion!


      "Subjective awareness of relevant perceptual stimuli"?

      Why the "subjective"? Is there any other kind of awareness?

      And why the "relevant"? Awareness of irrelevant stimuli is not consciousness?

      And what's "perceptual"? If a feeling of anxiety or doubt perceptual? Or just a feeling?

      And why a "stimulus"? If you hallucinate, so it feels like you're hearing a sound, but there's no acoustic stimulus, is that not consciousness?

      What does that leave us: "subjective" is gone, "relevant" is gone, "perceptual" is gone, "stimuli" is gone.

      That leaves "awareness of feeling something/"

      But can you feel something without being aware?

      And can you be aware without feeling something?

      Welcome back to weasel-word-free feeling...

    4. I agree with your view Prof. Harnad, but Dr. Lau put forth a definition, which I think is commendable - even if, ironically, it makes portions of his talk more amenable to consciousness-critique.

  2. Hi Martha,
    Yeah that's basically what I think. But perhaps my point is more that we will find *empirical evidence that apparently support* subliminal priming of almost any function. Whether these effects are real, I don't really know. It's hard to distinguish small effects some null effects some times....

    1. But although subliminal priming is in theory possible for any cognitive function, does it necessarily entail that conscious cognitive functions are superfluous ? I believe work in that perspective would also lead to subliminal priming's limits. In that sense, it would concur to the very important problem of defining consciousness' boundaries.
      Keep it up, Hakwan !

    2. Thanks for the reply Hakwan :)

  3. This comment has been removed by the author.

  4. I wonder whether the first experiment discussed by Dr. Lau -- the one involving subliminal priming with diamonds and squares in a cognitive control task -- shows without doubt that cognitive control can occur unconsciously. Here is an alternative interpretation: the subliminal priming only prepares certain areas of the brain for one the two tasks (semantic or phonological), but it doesn’t actually start the whole process of answering the relevant question (‘Is it concrete?’ in the semantic case ‘Is it monosyllabic?’ in the phonological case). The relevant area (be it the semantic module or the phonological module) then waits for the allocation of mental resources like working memory in order to operate fully. And this allocation only occur as the result of a conscious decision. Of course, this interpretation could only be plausible insofar as we can make sense of the notion of ‘preparing’ a module. Is this in any way plausible considering what we know in neuroscience?

  5. Xavier Dery ‏@XavierDery

    Lau believes that if you wait long enough, all cognitive tasks will be proven to be unconsciously done... Still no function then! #TuringC

    11:34 AM - 7 Jul 12 via Twicca Twitter app