On the (Dis)unity of Consciousness

Presenter: Jesse Prinz, The Graduate Center, CUNY


  1. Great talk, Jesse, and, as so often, I agree with the vast majority of what you say.

    With one exception, which is very important in this context: unconscious attention. You want to rule out unconscious attention and as far as I can see, your anti-unity argument relies on attention being sufficient for consciousness.

    But there are, of course, those Jiang et al 2006 studies, which really do seem to suggest that we do attend to the nude, without being conscious of it. You respond in various places (including Perceiving the World) that we need to make a distinction between attention to a spatial region and what you call ‘attentionally modulated object representations’. Your strategy is that the former, but not the latter, can be unconscious. And in the Jiang et al case, only the former is present.

    I’m not sure I understand the distinction between these two kinds of attention. You often seem to suggest that the allocation of spatial attention is somehow an early stage of ‘attentionally modulated object representations’, but when you’re responding to arguments against your claim that attention is necessary for consciousness (especially the Lamme and the Koch&Tsuchiya arguments) you seem to presuppose something like ‘attentionally modulated object representations’ without spatial attention.

    One straightforward way of thinking about spatial attention is that it attends to one property of an object: its spatial location. But if we think of spatial attention this way, then it is a subspecies of ‘attentionally modulated object representations’ – so that will not work for your purposes.

    Could you say a bit more on what you take to be spatial attention and what you take to be this distinction between spatial attention and ‘attentionally modulated object representations’? It would also be interesting to see which of these two concepts of attention the neural synchronization data tracks…

  2. Hi Jesse, as you know I am a big fan of the synchrony approach and advocate synchrony as a way to understand what brain states are in general. But I wonder how much you take this to be wedded to your view about attention and consciousness. So, for instance, if one is a higher-order theorist then one can for the most part still agree with everything that has been said here. I know that you say you only want to argue for a conditional claim that if attention is consciousness is right then unity isn’t what we might think that it is but I think it is important to point out that we get much the same results substituting higher-order states=consciousness with a synchrony story.

    Now I know you do not agree with higher-order theories, except in so far as AIR theory is a kind of higher-order theory (by that I mean it accepts the transitivity principle, or the thesis that a conscious state is one that I am aware of myself as being in and identities the proper kind of awareness as attention), but one confound in your argument for attention=consciousness comes from not addressing the issue of the relation between attention and higher-order state. So, one might argue, that attention is ‘attracted’ to a state because it is conscious (by being targeted by a HOT) or even vice versa that attending to a first-order state tends to cause a higher-order thought about it. Do you have any thoughts about the relation of attention to the kinds of higher-order thoughts in question? Given that we very often if not always find them together how could we ever distinguish between the two in practice?

  3. What would be an example of evidence that would count against the thesis that consciousness is attention?

    I’m a bit worried that you seem to equate theories that connect consciousness and the self with theories that consciousness is realised in some very specific location. Just as on your view consciousness involves different areas of the brain forming a specific kind of (not necessarily phenomenal) unity, surely any plausible view of the self is going to hold that the self is something which is constituted of many different cognitive functions unified together in some way, rather than something which exists only at the pineal gland.

    Does the unity thesis hold that all experiences within a single organism/brain/cognitive system are unified, or that all experiences within a single subject are unified?

  4. First one to Bence. Many thanks for this. The concern is well taken. My considered view on the Jiang et al. study may be a bit different from what I say in your volume (I’d have to look). What I think now is that there is a crucial difference between attending and orienting. Orienting is a more ancient family of responses that regulate what information gets into the sensory systems. It includes overt shifts of gave and head position, but also covert plans to execute such shifts (what neuroscientists sometimes call intention as opposed to attention). I think Jiang et al.’s result can be explained by assuming that the suppressed stimulus causes and intention to shift gave, which improves response when the target is present. I think further testing would be needed to confirm that this is intention/orienting rather than attention, but (a) such tests are possible since the two are anatomically dissociable, and (b) the fact the attention is implicated in binocular rivalry suggests that attention is not available during interoccular suppression.

  5. For Richard. PROFUSE thanks for making this incredible conference happen; we are all in your debt, and I am honored to have been a part.

    Great question. I haven’t thought a lot about how attention might relate to HOTs. I think AIR bears a close relationship with Lycan’s HOP theory, and things like the graded nature of consciousness make me incline towards HOPs over HOTs, where I to go higher-order. But, on your question, two things.

    First, I would be happy if the unity story could work with other accounts, including HOT theories, so I’m not wedding to this be right if AND ONLY IF consciousness is attention.

    Second, I think the HOTs and attention are correlated but dissociable. Bottom up attention (e.g., pop-out) tends to produce HOTs, and HOTs may serve to direct top-down attention. But I think there are cases of attention without HOTs, such as the ambient attention spread across a complex scene. And I think there are HOTs without attention. Some visually guided action may require the knowledge that I am seeing certain features without those features being attended. In these cases, I think attention goes with consciousness, and the HOTs do not. Testing is hard because I don’t know how to test for the presence of HOTs in the brain, and higher order judgments are often used to test for consciousness. I’d love suggestions.

  6. For Philip. Many thanks for the questions that cut to the heart of the theory.

    First, on evidence. Many neuroscientists think they have evidence that consciousness and attention are dissociable. For example, they show that cueing an object improved performance even when the cue or the target are not reported to be experienced. The kind of reply I gave to Bence (cueing shifts orientation, not attention) could be tested by finding neural correlates for both orienting and attending, and then testing whether either correlated with experience. A lot of the tests haven’t been done, so one can only raise these alternative interpretations of the data and design corresponding experiments.

    Second, on the self. Interesting questions, which I did not address clearly enough in the paper. If there is a self I agree that it is distributed, but so much the worse for thinking the self can solve the unity problem. The self just introduces a unity problem. Indeed, if the self were a single anatomically isolated thing, it would still be possible that one self could have disunified experiences, so I don’t think the appeal to a self will help, whether or not the self is distributed.

    Third, on disunity, yes. I think there can be disunified experiences and often are. I try to suggest that we find unity when we look for it, but there are reasons to think disunity is common. One reason has to do with the fact that there are multiple control structures for attention, so different stimulus streams could be attended separately.

Comments are closed.