Higher-order thoughts as necessary and sufficient condition of consciousness

As per (Rosenthal, 2005), “[HOTs] do not transfer the property of being conscious from themselves to their targets; indeed, they don’t induce any changes whatever in those targets” [p.185 …] A mental state’s being conscious is not strictly speaking a relational property of that state. A state’s being conscious consists in its being a state one is conscious of oneself as being in. Still, it’s convenient to speak loosely of the property of a state’s being conscious as relational so as to stress that it is in any case not an intrinsic property of mental states” (p.211).

As per (Wilberg, 2010), “a mental state is conscious if and only if it is accompanied by the [suitable] HOT that one is in that state [p.618 …]. It seems that the [suitable] HOT must represent, at least roughly accurately, the individuating features of that state. [p.619 …] Consciousness as a property of token mental states [p.625]”.

As per (Berger, 2014), “Rosenthal’s version of HOT [higher-order thought] theory, according to which a suitable HOT  is both necessary and sufficient for consciousness … consciousness is best understood as a property of individuals, not a property of states. […] Rosenthal’s higher-order thought (“HOT”) theory of consciousness, which holds that one is in a conscious mental state if and only if one is aware of oneself as being in that state via a suitable HOT (see, e.g., (Rosenthal, 2005)). [p1…] Consciousness is, as Rosenthal often emphasizes (e.g., 2009, p. 166), a matter of how one’s mental life appears to one. But appearance and reality can in general diverge. [p2…] Although many theorists do assume that consciousness is a property of states, this assumption is questionable. Indeed, Rosenthal himself has been explicit that consciousness is not a property conveyed by HOTs to first-order states (2005, p. 185). […] The fundamental motivation for HOT theory is the claim that one is in a conscious state only if one is aware of oneself as being in that state. Rosenthal has called this fact about consciousness the “Transitivity Principle” (TP) because it explains what is for one to be in a conscious state in terms of one’s transitive [transitional, intermediate] awareness of being in that state (2005, pp. 3–4). It is clear, however, that the TP offers a necessary, but not sufficient, condition for consciousness. […] According to HOT theory, suitable HOTs are the states in virtue of which there are these subjective impressions. A suitable HOT is an occurrent intentional state that asserts the content that I am in some state. […] Since a targetless HOT does not accurately represent any state, there is no state to exhibit consciousness. In such a case, one only seems to be in a conscious state. Hence Wilberg proposes his No Consciousness Account of targetless HOTs. On Wilberg’s view [(Wilberg, 2010)], HOTs are necessary, but not sufficient, for consciousness. […] But because Wilberg maintains that a state’s being conscious involves its acquiring the property of consciousness, he denies that the subjective appearance of being in a certain state is sufficient for consciousness. On Wilberg’s view, one is in a conscious state if and only if it appears to one that one is in a state via a suitable HOT and, in addition, one is in that state. So while consciousness does concern mental appearance, Wilberg maintains that it is a ‘matter of (rough) correspondence between appearance and reality’ (2010, p. 630). […] If one reports seeing a red apple, such a report signals that it appears to one that one sees a red apple—and all there is to a conscious state, we think, is the subjective appearance that one is in some state. Since appearances in general need not correspond to reality, it is implicit in our folk conception of consciousness that consciousness is a matter of mental appearance, whether or not those appearances are accurate. […] To accurately capture our ordinary conception of consciousness, then, HOT theory should maintain that a suitable HOT is not only necessary but also sufficient for consciousness—and this is the version of HOT theory that Rosenthal defends. On Wilberg’s No Consciousness version of HOT theory, by contrast, the mere subjective appearance of a state is insufficient for consciousness. For Wilberg, if one has a targetless HOT that one sees a red apple, seems to see a red apple, and on that basis reports that one sees a red apple, one’s report does not indicate that one consciously sees the apple. In severing the connection between verbal reports and consciousness, Wilberg’s view violates a feature central to both our commonsense and experimental approaches to consciousness. […] There is thus independent reason to think that consciousness is not a property of existing first-order states conveyed to them by HOTs. […] Again, all that matters for consciousness is a suitable impression that one is in a state. According to HOT theory, HOTs are the states in virtue of which one has the subjective impression that one’s mental life is some way. And HOTs are, of course, states of individuals. It is thus compatible with the folk conception of consciousness that consciousness is not a property of states, but a property of individuals—namely, the property of being aware of oneself as being in a state. […] So though it is acceptable shorthand to say that one is in a conscious state only if one is aware of that state, it is more accurate to say that one is in a conscious state only if one is aware of oneself as being in that state. This way of casting the TP yields a version of HOT theory according to which one is in a conscious state if and only if one is aware of oneself as being in that state via a suitable HOT. This explains why the content of a suitable HOT is that I am in a particular state, not merely that there is such a state. […] On Rosenthal’s view, by contrast, what a HOT makes one aware of is, strictly speaking, oneself. So if one has a suitable HOT, one’s HOT always renders one aware of something that exists—namely, oneself.13 If one has a HOT, then one exists; as Descartes said, a thought requires a thinker […] Sometimes one is aware of oneself as being in a state that exists (when one’s HOT is accurate) and sometimes one is aware of oneself as being in a state that does not exist (when one’s HOT is targetless). […] One way to unpack Rosenthal’s claim would be to hold that when one has a suitable HOT, one exhibits the property of being-in-a-conscious-state, wherein this is understood to be the property of an individual who is suitably aware of being in some state.15 [if HOT theory holds that consciousness is a property of an individual’s representing itself, then the theory is not really a higher-order theory of consciousness at all … HOTs render individuals aware of themselves as being in states. …] What [consciousness] might seem to be a property of a state is actually a property of an individual’s representing itself as being in a state. […] Conscious states are whatever states one is subjectively aware of oneself as being in. So there is, after all, a way in which we can comfortably describe consciousness as attaching to states, even in cases of targetless HOTs. If we do so, as Rosenthal proposes, we apply the property of consciousness to notional states. But this too is accommodated by HOT theory. We often apply properties to notional objects. And it is unclear what evidence could be brought to bear upon the decision between the view that consciousness is a property of notional states and the view that it is a property of actual individuals representing themselves as being in those notional states. […] What it is for one to be in a conscious state is for one to have the suitable appearance of a state, whether or not one is in that state. This is not to say that consciousness does not exist or that it is only a matter of appearances. HOTs and the appearances that they reflect are real, though it is the individuals, and not the states, that are conscious.”

It seems that first we need to investigate the necessary and sufficient conditions for HOT and then investigate if these conditions are the same as that for consciousness.[1]

[1] Email discussion with Jacob Berger (Nov 2015):

Vimal (11/3/15): What are the necessary and sufficient conditions for HOT?

Berger (11/3/15): As Rosenthal conceives of it, HOT theory holds that a mental state is conscious if and only if one has a suitable HOT about it. Now, there is some debate about what 'suitable' means. Rosenthal claims, at least, that the HOT must arise in a way that does not seem to be the product of inference or observation. But there may be other conditions that need to be met. 

Vimal (11/3/15): You and Rosenthal seem to claim that HOT is the necessary and sufficient conditions for consciousness. However Wilber seems to claim that HOT is the necessary but not sufficient conditions for consciousness. Your wrote, “the TP offers a necessary, but not sufficient, condition for consciousness”. 

“Rosenthal claims, at least, that the HOT must arise in a way that does not seem to be the product of inference or observation. But there may be other conditions that need to be met.”

From this it is not clear to me what the necessary and sufficient conditions for HOT are. What is the origin of thought and HOT?

My understanding about necessary and sufficient conditions is as follows: “The necessary conditions for consciousness are those conditions that must be satisfied in order to have consciousness, i.e., if any of them is missing then the entity is not conscious. The sufficient conditions for consciousness are conditions, if satisfied, guarantee that the entity is conscious.” (Vimal, 2015f).

Could you please elaborate it in detail as much as you can?

Berger (11/4/15): You're right that Rosenthal and I think that a suitable HOT is necessary and sufficient for consciousness, whereas Wilburg argues that such a HOT is only necessary.  I try to make the case for why Wilburg is mistaken in the paper I sent you.

As I say there, the TP (which is the motivation for HOT theory) only specifies a necessary condition for consciousness: namely, a mental state is conscious only if you're aware of it somehow.  But being aware of a mental state isn't sufficient for consciousness because, after all, I can become aware of a mental state of yours, but that need not--and indeed cannot--make it conscious.  

So Rosenthal tries to generate a theory of consciousness--HOT theory--that not only captures the TP, but also specifies what kind of awareness is both necessary and sufficient for consciousness.  And this is what we capture by saying that the HOT, in virtue of which a state is conscious, has to be "suitable."  

As Rosenthal argues (e.g., on p. 27 of that book), a mental state need not be conscious if we are aware of it in a way that seems to be mediated by inference or observation.  If you tell me that I'm happy, and I really am happy, my state of happiness still need not be conscious.  My state of happiness is conscious if and only if I'm aware of it *and* it doesn't seem to me that I'm aware of it because of  inference or observation.  That isn't to say that I am not in fact aware of it because of inference or observation, only that it doesn't seem that way to me.  So at least one part of what it is for a HOT to be suitable is for it not to seem to arise as a product of inference or observation.  In other words, you have to seem to be directly aware of it.

Does that help?

Vimal (11/5/15): [I] An individual (consists of body, brain and mind) has innumerable states including conscious, unconscious, and non-conscious states. Therefore, it is more precise to consider a specific state of an entity.

    [II] Let us interpret the HOT theory of consciousness in the least problematic extended dual-aspect monism framework (eDAM) (Vimal, 2008, 2010a, 2013, 2015a, 2015b), where a state of an entity has two inseparable aspects: 1pp-mental aspect and 3pp-physical aspect. [1pp: first person perspective and 3pp-third person perspective.]

     In the eDAM framework, the optimal definition (that has the least number of problems) of consciousness is: consciousness is the mental aspect of a state of a brain-mind system or a brain-process from the first person perspective; consciousness has two sub-aspects: conscious function and conscious experience (Vimal, 2010b).

As per (Berger, 2014), “Rosenthal’s higher-order thought (‘HOT’) theory of consciousness, which holds that one is in a conscious mental state if and only if one is aware of oneself as being in that state via a suitable HOT (see, e.g., (Rosenthal, 2005)).”

In the eDAM, a self is the subjective experience (SE) of a subject (from first person perspective) (Bruzzo & Vimal, 2007), which is different from the objective experience of the subject (from third person perspective). The former includes inner feeling of ‘I”-ness, whereas the latter is the experience of her/his own body as an object similar to that of any object.

Therefore, in the eDAM, Rosenthal’s HOT theory holds one is in a conscious mental state if and only if one is aware of ‘the subjective experience of the subject’ as being in that state via a suitable HOT.

In the eDAM framework, the suitable HOT entails the relationship between three kinds of signals: the self-related neural-network signal interacts with the result of the interaction/matching between stimulus-dependent feed forward (FF) signals and cognitive feedback (FB) signals. Here, stimulus could be either exogenous (external target stimuli) or endogenous (internal stimuli without external target, internal target could be internally generated target as in dreams, imaginations etc.). 

In the eDAM, the first-order state is the state related to “the interaction/matching between stimulus-dependent feed forward signals and cognitive feedback signals”, which is not conscious yet; it needs some entity to experience the matched specific experience; and that entity is the self. The self is the SE of the subject, which is the 1pp-mental aspect of the self-related state of the mind-brain system; the 3pp-physical aspect of this state is the self-related neural-network and its activities. Thus, a HOT theory is consistent with the eDAM framework.

To sum up, if suitable HOT is the necessary and sufficient condition for consciousness, then the necessary and sufficient conditions for the suitable HOT are the same as that for consciousness.

“The necessary conditions for access (reportable) consciousness are as follows: (1) the formation of neural-networks; (2) wakefulness; (3) reentrant interactions among neural populations; (4) fronto-parietal and thalamic-reticular-nucleus attentional signals that modulate consciousness; (5) integrated information (F) at or above threshold level; (6) working memory; (7) stimulus contrast at or above threshold; and (8) neural-network proto-experiences that are superposed potential subjective experiences (SEs) embedded in a neural-network as pre-cursors of SEs. The necessary conditions for phenomenal (non-reportable) consciousness are (1)-(3) and (5)-(8), i.e., the same as access consciousness except attention.” (Vimal, 2015a).

[III] Once we accept a suitable HOT then the eDAM addresses the difference between Rosenthal’s/your and (Wilberg, 2010)’s views and controversy related to external targetless HOTs. It seems that, for Wilberg, target means external stimulus only. Target needs to be precisely defined as it could be internal or external stimulus. This is because we can have experiences or HOT in our imagination (e.g., I can imagine red apple that will create a state) or in our dream for endogenous stimuli/targets even if there is no external target. A state related to this is can be thought of state for the endogenous/internal target without external target; this is what I understood by targetless HOT, where a target is an external stimulus. The state of the tip-of-tongue related to author’s name need to involve self to make the HOT as a suitable HOT. 

[IV] There are over 40 meanings assigned to the term ‘consciousness’, which were categorized in to two groups: functions and experiences as elaborated in (Vimal, 2009). Rosenthal and you seem to include experiences related to both internal and external targets for consciousness; whereas, Wilberg seems to use only external target in his definition of consciousness. Thus, different meanings are assigned to the term “consciousness” leading to further confusion.

Thus, Rosenthal and you seem to defend the view by implicitly using internal target also in consciousness, whereas Wilberg seems to use only external target and hence conclusions appear contradictory, although in reality there is no contradiction. Misunderstanding arose because of the lack of the precise definitions of target and consciousness and their use.

[V] To sum up, as long as self is involved in interactions, i.e., all three kinds of signals (self, FF and FB) interact, then the self should be able to experience the specific experience; otherwise, a first-order state will be created via the interaction/matching of FF and object related FB signals, but there is no entity (such as self) to experience it.  This explanation addresses the controversy between first-order and high-order frameworks.

What are your comments on the above view?

[VI] Query: [1] What are the origins of thoughts? Our thoughts could be due to intrinsic activities and/or because of extrinsic stimuli. Is this correct?

[2] Thoughts, HOTs, attention, memory are included in cognition. Cognition is a functional sub-aspect of consciousness, whereas our subjective experiences are the experiential sub-aspect of consciousness. If a suitable HOT (functional sub-aspect of consciousness) is necessary and sufficient condition for the experiential sub-aspect of consciousness, then this is consistent with the hypothesis of 1-1-1 correspondence between function, related experience, and related neural correlates. Do you agree?

Berger/Vimal (11/10/15)

[II]

Berger: I'm not sure if this is entailed by eDAM, but my version of HOT theory does not hold that the 1pp mental appearances of consciousness are distinct from the 3pp physical features of their brains.  I think mental appearances are just brain activity that a creature is aware of in the right kind of way.  So we have two perspectives, the 1pp and 3pp, on the same thing.  It may not seem like the same thing, but it is.

Vimal:  The eDAM also says that the information is precisely the same for both aspects/perspectives; in addition, they are inseparable, except HOT bases on problematic materialism (consciousness is generated by non-conscious matter such as brain) and eDAM bases on the least problematic dual-aspect monism.

          Berger (11/19/15): I obviously don't think that materialism is problematic--I think it's reasonable to think that consciousness can be generated by non-conscious matter.  This is especially clear if you grant a HOT model on which thoughts can occur nonconsciously (and so are unproblematically material) and that consciousness is just a product of certain kinds of thoughts.

          Vimal: There are over 40 meanings attributed to the term ‘consciousness’ as elaborated in (Vimal, 2009), which were identified and categorized according to whether they were principally about function (thoughts can be considered as functional aspect of consciousness) or about experience. If you attribute the product of certain kinds of thoughts as consciousness then perhaps there is not much problem. However, the experiential aspect of consciousness (what it is like to be …) is different and materialism has serious well-known explanatory gap problem as elaborated in (Vimal, 2010d, 2013).

[IV]

Berger: I agree that the term 'consciousness' is highly ambiguous.  And your idea that Wilberg and I are just talking past each other because we're using different notions of it is an interesting one.  But I'm not sure that's what's going on.  On my view, a conscious state is a state that it (suitably) seems to you that you're in--and I think Wilberg agrees.  And we both agree that, whenever it seems to you that you're in a state, there is something that exists--namely, the HOT in virtue of which it seems to you that way.  What we disagree about is whether or not the state that it seems to you you're in must exist for there to be consciousness.  I say it doesn't: You can be in a conscious state that doesn't exist--which is to say that it can seem to you that you're in a state that you're not in fact in.  Wilberg disagrees.  He thinks that if it seems to you that you're in a state, but you're not in that state, you're not in a conscious state.  I think he's wrong because consciousness isn't a property of the target state (the state you're aware of being in, but might not be in), but a property of you--to be in a conscious state is to be aware of yourself as being in a state.  So in that way we disagree about what the predicate 'conscious' applies to.  I say it applies to creatures; Wilberg says it applies to the target of a HOT.  Perhaps that's what you had in mind? As I say in the paper, though, we often do apply 'conscious' to the target states, and that's fine as far as it goes.  It's just that it is, in a way, misleading and a kind of loose talk.

Vimal: So difference is: you claim that consciousness is property of a creature to resolve the problem of targetless HOT; whereas, Wilberg claims that consciousness (C) is the property of a state (S) of the creature. Since a creature can have innumerable states, one of them may be conscious of seeing redness of ripe tomato, i.e., (S1, C1), second may be related to the taste of the sugar, i.e., (S2, C2), third may be listening to music, i.e., (S3, C3), etc. If you assign consciousness to the creature then it should be the same experience in all states of the same creature, which is obviously not true in this example. Therefore, a specific experience/consciousness should be assigned to a specific state.

Berger (11/19/15): That's a nice argument.  But I don't think it shows my view is problematic. I think that when we say that "a creature C is in a conscious state S" we mean that the creature has a certain property--namely, the property of being aware of itself as being in S.  But that means that the creature can be in many different kinds of conscious states at once--that is, the creature has many kinds of properties.  The creature might be aware of itself as being in state S, state S1, state S2, etc.  And being in those conscious states are all distinct properties of the creature.

Vimal: Yes, S is a unified state as the superposition of S1, S2, etc. Does Wilberg claim that micro-consciousness Ci is the property of ith state Si of the creature? If this is true, then macro-consciousness C is the property of unified macro-state S of the creature at that moment, which is the same as saying C is the property of the creature. Thus, there is no contradiction. Now, let ripe-tomato is missing, then (S1, C1) will be missing also. However, the creature can now imagine ripe-tomato from her memory; this will create S’1, with corresponding C’1. Again, I do not see any contradiction.

[VI]

[1] Berger: I think so.  Sometimes I am caused by extrinsic stimuli to have a thought.  An apple causes me to see the apple and that causes me to think that there's an apple.  But sometimes I can just "bring up" a thought myself endogenously.  I just sit here and think that Paris is in France.  How I do that, however, is something that is still unclear to me.

[2] I'm not sure I agree.  For one thing, I think much thinking (and cognition generally) occurs without being conscious.  According to HOT theory, a mental state is conscious iff one is aware of it via a HOT--but those HOTs need not be, and typically aren't, themselves conscious.  So cognition is not a functional sub-aspect of consciousness, it's the other way around.  And on my view, the expression 'subjective experience' just means conscious mental state.  So while I agree that you can't have a mental function without some neural activity (because mental functions just are neural activities), I don't think you have to have experience--some mental activity occurs without consciousness.  Indeed, that is what HOT theory is trying to account for: the difference between conscious mental states and mental states that occur outside of consciousness.  Does that help?

Vimal: Well, I guess, I was not clear. If you want to pick up a coffee, then function is picking up, experience is where the cup is, and the structure is all related anatomy and physical activities. In this example, structure : function (cognition) : experience (consciousness) :: 1:1:1 and this triad is inseparable for a specific time, specific space, and specific condition; otherwise, if there is any mismatch then you will not able to pick up the cup.

Berger (11/19/15): I agree that our cognition and experience is often wrapped up.  But one of the key motivations for HOT theory is that many kinds of mental activity, including thought, can occur without experience at all.  There are many commonsense and experimental examples of unconscious thinking and arguably even unconscious acting that do not require the creature to be aware of itself as thinking or acting.  If so, then experience isn't inseparable from action.  It often accompanies it, but it need not.  Or at least that's what I think.

Vimal: In that case experience is missing, but function (unconscious thinking) and related structure must remain intact. Missing does not mean separable; it means experience is latent because once you become conscious then experience, function, and structure return back to be inseparable. This example does not violate the doctrine of inseparability of aspects and sub-aspects.

 References:

Berger, J. (2014). Consciousness is not a property of states: A reply to Wilberg. Philosophical Psychology ; , 27(6), 829-842.

Bruzzo, A. A., & Vimal, R. L. P. (2007). Self: An adaptive pressure arising from self-organization, chaotic dynamics, and neural Darwinism. J Integr Neurosci , 6(4), 541-566.

Rosenthal, D. M. (2005). Consciousness and mind. Oxford: Clarendon Press.

Vimal, R. L. P. (2008). Proto-experiences and Subjective Experiences: Classical and Quantum Concepts. J Integr Neurosci [Available at ; Latest update: ], 7(1), 49-73.

Vimal, R. L. P. (2009). Meanings attributed to the term 'consciousness': an overview. J Consciousness Stud , 16(5), 9-27.

Vimal, R. L. P. (2010a). Matching and selection of a specific subjective experience: conjugate matching and subjective experience. J Integr Neurosci [], 9(2), 193-251.

Vimal, R. L. P. (2010b). On the Quest of Defining Consciousness. Mind Matter (Available: < http://sites.google.com/site/rlpvimal/Home/2010-Vimal-DefineC-LVCR-3-2.pdf >), 8(1), 93-121.

Vimal, R. L. P. (2013). Emergence in Dual-Aspect Monism. In A. Pereira Jr. & D. Lehmann (Eds.), The Unity of Mind, Brain and World: Current Perspectives on a Science of Consciousness (pp. 149-181). Cambridge, UK: Cambridge University Press. [Longer version is available for comments: ].

Vimal, R. L. P. (2015a). Necessary and sufficient conditions for consciousness: Extended Dual-Aspect Monism framework. Vision Research Institute: Living Vision and Consciousness Research [Available: ] [DOI: http://dx.doi.org/10.13140/RG.2.1.1587.9124], 7(1), 1-28.

Vimal, R. L. P. (2015b). Segregation and integration of information: extended Dual-Aspect Monism framework. Vision Research Institute: Living Vision and Consciousness Research [Available: ] [DOI: http://dx.doi.org/10.13140/RG.2.1.1974.3445], 7(2), 1-39.

Wilberg, J. (2010). Consciousness and false HOTs. Philosophical Psychology, 23(5), 617-638.