Ten years ago, Susanna Siegel proposed the method of phenomenal contrast in order to determine the type of properties that are represented in perceptual experiences. In brief, do we see only lines and colors or do we also see pine trees? Her method proceeds in two steps. First, one describes a situation in which there is intuitively a contrast between two experiences, only one of which instantiates the high-level property (e.g., seeing a tree before and after becoming an expert pine tree spotter). The second step consists in drawing an inference to the best explanation of this contrast by ruling out alternative explanations. Most debates on the permissible content of perception have focused on visual awareness but it may as well be applied to bodily awareness: do we experience only bodily posture and pressure or do we also experience our body as our own? One should then consider borderline cases in which the awareness of the high-level property of ownership is altered, namely disownership disorders and the rubber hand illusion.
Somatoparaphrenia: After a lesion in the right parietal lobe some patients deny that one of their limbs belongs to them: “Ex: Whose arm is this? A.R.: It’s not mine”.
They can regain a sense of ownership towards their hand after vestibular stimulation: “Ex: (Raises the patient’s left arm.) Is this arm yours? A.R: Why, yes”.
This recovery is only temporary and two hours later they deny again that this is their own hand: “Ex: (Points to the patient’s felt arm.) Whose arm is this? A.R.: It’s my mother’s.” (Bisiach et al. 1991, p. 1030).
The Rubber Hand Illusion(RHI): Participants sit with their arm hidden behind a screen, while looking at a rubber hand presented in their bodily alignment; the rubber hand can then be touched either in synchrony or in asynchrony with their own hand. Participants report ownership over the rubber hand after synchronous stroking only: “I found myself looking at the dummy hand thinking it was actually my own.” (Botvinick and Cohen, 1998, p. 756)
In both cases, there is a clear contrast between the time at which subjects judge that this is not their hand (before vestibular stimulation and in the asynchronous condition) and the time at which they judge that this is their hand (after vestibular stimulation and in the synchronous/ condition). But do their subjective reports express more than their cognitive attitudes?
One may, for instance, claim that what somatoparaphrenic patients lack (and momentarily regain) is their ability to believe/accept/suppose/imagine that their own hand is their own. In favour of the cognitive account is the fact that somatoparaphrenic patients are delusional: they are convinced that the hand is not their own. Their cognitive disruption, however, does not suffice to exclude disruption at the experiential level too. One may actually argue that delusions are caused by abnormal experiences that the patients try to account for. The cognitive interpretation furthermore faces a general problem. Its formulation remains relatively hand waving because it is not clear what kind of cognitive attitude can account for the contrast highlighted above. It seems indeed that it must answer opposing demands.
A primary feature of the (dis)ownership thought is that it is cognitively impenetrable. Some patients know that their hand belongs to them, and yet it still seems to them as if it did not: “my eyes and my feelings don’t agree, and I must believe my feelings. I know they look like mine, but I can feel they are not, and I can’t believe my eyes”(Nielsen, 1938, p. 555). Likewise, subjects in the RHI rarely self-attribute the rubber hand and merely state that it seems as if it were their hand. Hence, the (dis)ownership thought must be insensitive to the influence of beliefs. This narrows down the range of potential candidates to cognitive attitudes such as supposition and acceptance.
The most promising candidate may then be imagination, which can escape control and which can be affectively loaded. However, on some interpretations, imagination involves mentally recreating selected experiences (Currie and Ravenscroft, 2002). If the subjects in the RHI imagine their hand to be their own, it means that they imagine experiencing it as being their own. The imaginative proposal is thus not purely cognitive; it also includes the experience that is re-enacted in imagination. This gives rise to a first difficulty. The disownership thought in the clinical case seems to have an emotional impact. Patients are indeed greatly distressed even if they know that their hand is their own. However, merely supposing or accepting that this is not one’s hand is not affectively disturbing precisely because one does not take what the thought represents to be true. A solution is to assume that it is a deficit at the subpersonal level that explains the distress expressed by the patients, instead of the disownership thought itself. This would solve the affective problem but it does not help to account for the lack of control that patients and subjects have over their (dis)ownership thought. Indeed, patients cannot stop at will to entertain the idea that this is not their hand, whereas many subjects fail to experience the RHI despite wanting to have it. The difficulty is that cognitive attitudes such as acceptance and supposition (unlike beliefs) are typically under voluntary control. One cannot decide to believe in God but one can voluntarily suppose that God exists or even accept it (as in Pascal’s wager).
To recap, patients describe that they feel that their hand is not their own, whereas participants in the RHI report that they feel that the rubber hand is their own. The burden of proof is on those who refuse to take them to their words. Most of their arguments rely on criticisms of the experiential interpretation of these cases but they need to offer a positive account of the type of cognitive attitude that can do the job as well as a (dis)ownership experience. So far it has been hard to find any.
To continue the line of questioning that I began in Mind the body (1): Is the feeling of ownership in the rubber hand illusion a part of, or separate from, the feeling of the hand being stroked? Let’s stipulate that, for the reasons you give (cognitive impenetrability, affective loading), it is not a cognitive attitude—a belief or an imaginative state. This does not settle the question of separateness: it could be a separate perceptual state, for example, that arises after synchronous stroking. If the feeling of ownership is not separate—if it is immanent in a perceptual state, then non-human animals should feel be capable of it. Question: do dogs suffer a rubber paw illusion? Would they flinch if a synchronously stroked rubber paw was struck with a mallet?
About the issue of separateness. The question as I take it is whether the feeling of ownership is an experience fully independent from bodily sensations. Another way to put it is whether one can experience a feeling of ownership without any bodily sensation (because of complete anesthesia or prolonged stay in deprivation tank). On my view, this is impossible because the feeling of ownership arises from the spatial frame of reference of bodily sensations: it is because sensations are felt to be located within the protective body map that one experiences the felt body part as one’s own.
It should be noted that the question of separateness goes beyond the case of ownership and bodily awareness. One might ask the same for the feeling of presence and for the feeling of familiarity. In the latter case, one always mentions Capgras syndrome: sensory content with no affective content. The Fregoli syndrome (unknown people feel familiar) is also interesting: affective content wrongly associated with sensory content. But in both syndromes, there is always a sensory content. Could one have a floating feeling of familiarity, with no sensory grounds? My friend and colleague Jérôme Dokic claims that the feeling of déjà vu is such a case. But there is a visual scene attached to the déjà vu. Could a deaf blind person experience déjà vu?
There is already a rubber tail illusion in mice (Wada et al., 2016). The tail was not threatened but simply grasped. Mice responded in a different way when the rubber tail was grasped after synchronous and asynchronous stroking. If we assume that this is really a feeling of ownership that the mice experience, then this is one more argument against a cognitive account I’d say. More on separateness soon.
I agree with your argument for the cognitive impenetrability of the feeling of ownership. Such cognitive impenetrability is illustrated by the striking examples of feelings of disownership in patients with somatoparaphrenia you report and in feelings of external control in patients with schizophrenia. It is also exemplified in metacognitive feelings, which keep influencing epistemic decisions even when subjects are instructed that these feelings were formed in a “non-diagnostic” way (Nussinson & Koriat, 2008). In addition, social stereotypical feelings have been shown to persist in contradiction with one’s own explicit beliefs in impulsive situations of decision-making (Greenwald & Banaji, 1995).
Cognitive impenetrability, then, is expressed is in our case by the dissociability between the feeling that one owns a limb and believing it. One can feel and not believe, or believe and not think that one owns a segment attached to the body. Cognitive impenetrability, however, seems prima facie to clash with Anscombe’s view that a bodily feeling fails to offer a separable ground for justifying the judgment that expresses it. How does this view hold, when a judgment of ownership may be formed against a feeling of disownership, and vice versa? Is not it furthermore justified to draw on cognitive and neuroscientific studies as well as on mere observation of one’s own reactions in order to further characterize the content of our experience of ownership ? One’s reactions (when a hammer is about to hit the rubber hand) are guided by the feeling, not by the judging ; and do not our spontaneous reactions carry information about the kind of feeling we have, independently from the self-descriptions that can be formed (or not) on the basis of this experience?
Does this line of reasoning capture the gist of your argument?
Keying error
Please correct the following sentence in my comment on post 2 : “One can feel and not believe, or believe and not *feel* that one owns a segment attached to the body.”
You are right, Joelle, the argument of cognitive impenetrability plays an important role here. But it is not sufficient. As noted by McDowell, feelings and experiences are not the only states that can be immune to the influence of beliefs. We know that supposition, acceptance, and imagination are too. This may also be true of Anscombe’s notion of knowledge without observation. But cognitive impenetrability is not the only criterion to meet. Based on the empirical cases we know of, the (dis)ownership state must satisfy the following three conditions: (i) dissociable from beliefs; (ii) beyond voluntary control; (iii) with affective impact. The argument then is an inference to the best explanation. It appears that feelings are the best candidate.
To me, ownership seems to be not one sense, but a combination of senses. There is a proprioceptive sense, of where the body part is, but that is experimentally separable from the agentive sense, which is the sense that ‘I’ cause the body part to move. The entraining in the rubber hand illusion is about using the tactile sense to overwhelm the more subtle proprioceptive sense.