By self-awareness, however, philosophers have in mind the experience of being an entity that exists through time, which is not something that can be produced by reflexive processing. The organism needs to be able to represent itself, not just moment-to-moment but as an entity with a history and a future (“to consider itself the same thing at a time and over time”.) It must therefore be able to link affective experience to memory and prospection in the same way as it links it to perception and sensory processing moment to moment. That is to say that it must be able to appraise episodes of memory and foresight for self-relevance.
Because the temporal window of human cognition extends beyond the present we have evolved systems that recapitulate important aspects of reflexive affective processing for those higher level cognitive processes involved in planning, recollection, prospection, and decision-making. These systems simulate temporally distant experiences by rehearsing some of the same perceptual and emotional mechanisms activated by the simulated situation. As a result we can recall previous episodes of experience and imagine future episodes of experience and link those simulations to other high-level cognitive processes in order to plan and decide. We remember being sunburnt and imagine getting skin cancer when deciding whether to go to the beach at noon (Gusnard et al. 2001; Buckner et al. 2008; Fair et al. 2008; Broyd et al. 2009).
These simulations are the raw material of autobiographical narratives whose structure and duration can vary depending on cognitive context. They may be as simple as recall of a single event that triggers a flash of affect, but can also be assembled into elaborate histories and imaginative rehearsals depending on the cognitive context. This narrative capacity provides a crucial aspect of cognitive control possibly unique to humans. The most important aspect of these simulations is sometimes overlooked in studies that emphasise their quasi-perceptual content. That is the fact that the simulation of perceptual and sensory experience evokes affective associations. We simulate a scene in order to evoke the affective responses that represent the significance of events and objects for us. When we imagine or recall an episode of experience its affective significance is also represented in experience via the offline rehearsal of affective processing. The ventromedial prefrontal cortex is a structure which “traffics” or makes available the affective information. In effect, the ventromedial prefrontal cortex recapitulates at a higher level the properties of the amygdala. In so doing it associates affective information with explicitly represented information used in reflective decision making and planning (Ochsner et al. 2002; Bechara & Damasio 2005). It thus allows the subject to make explicit reflective appraisals. When I lie on the beach I have pleasant feelings produced by low-level appraisal systems. When I imagine or recall lying on the beach while trying to decide whether to holiday in Thailand or Senegal my ventromedial prefrontal cortex makes available the affective information prompted by that simulation.
This is why “pure” episodic memory studies (such as recall of content of visual scenes) do not activate the ventromedial prefrontal cortex, whereas “activations in the ventromedial PFC [prefrontal cortex] … are almost invariably found in autobiographical memory studies” (Gilboa 2004, p. 1336; my emphasis). Gilboa (2004) suggests that this is because “autobiographical memory relies on a quick intuitive ‘feeling of rightness’ to monitor the veracity and cohesiveness of retrieved memories in relation to an activated self-schema.” This is consistent with studies showing activity in the ventromedial and related subcortical structures when people make intuitive (that is, rapid and semiautomatic) judgments about themselves. When people make judgments about themselves using semantic knowledge and symbolic reasoning, ventromedial structures are less active.
This idea is supported by studies of patients with lesions to the ventromedial prefrontal cortex. These patients oscillate between various forms of reflexive cognition and more abstract forms of thinking using semantic knowledge and procedural reasoning. What they have lost is the ability, provided by ventromedial structures, to simulate affective and motivational response in the absence of the stimulus, while they retain the ability to process information in an abstract way. Consequently, a ventromedial patient may be able to do a utility calculation about her personal future but be unable to act on that knowledge. It appears that semantic knowledge is motivationally inert. Such results are often used to emphasize the necessary role of affect in deliberation, but they also suggest that what those affective responses do is provide the necessary personal perspective on information. They make the information mine, so to speak. Furthermore, this diminishment is not just at a time, but over time. These patients, although not amnesic in the strict sense of the term, have very limited ability for autobiographical recall or prospection. They have no sense of a persisting self (Damasio 1994; Bechara & Damasio 2005; Gerrans & Kennett 2010).
This suggests that disorders in which people feel a diminished sense of self would be characterized by hypoactivity in the ventromedial prefrontal cortex. In a review of the neuropsychological and imaging literature, Koenigs & Grafman concluded that “one could conceive of the VMPFC patients’ selective reduction in depressive symptoms as a secondary effect of a primary lack of self-awareness and self-reflection” (2009, p. 242; my emphasis). In other words, patients with ventromedial damage do not “feel” personally affected when considering even quite distressing events because they cannot access or activate the required affective responses.
It seems that “mine-ness” of experience is a cognitive achievement mediated by the ventromedial prefrontal cortex. As we noted above the ventromedial prefrontal cortex is suited to play this role because it recapitulates at a higher level many of the processing properties of lower-level hubs of emotional processing that represent self-relevance. Rather than reinvent the cognitive wheel for controlled processing, evolution has provided pathways that traffic affective and reward-predictive information processed automatically at lower levels to controlled processing coordinated by the ventromedial prefrontal cortex.
In effect, these studies suggest that in both online reflexive and offline reflective processing affective processes are needed to represent the significance of the information for the subject, and it is the consequent bodily feelings that produce the feeling of self-awareness My version of this view is in some ways an amalgam of ideas found in Seth (2013) and Proust (2013). All three of us share the view that the mind is hierarchically organized, and that feelings of self awareness emerge when higher order, metacognitive processes such as planning or deliberation integrate bodily information which signals relevance. On Seth’s and my view the Anterior Insular Cortex (AIC) is in some ways specialized for that function in view of its architecture: it does not merely relay first order bodily information but is involved in the representation of the significance of that information. Thus it is well placed to be the source of some of the metacognitive feelings identified by Proust (2013) as serving crucial indicator functions.[3]
Affective processes represent the relevance of information for an organism and initiate suitable action tendencies and autonomic responses. The bodily consequences are sensed and summarised by specialised systems that inform the organism how it is faring in the world: this is affective information (Prinz 2004). This affective information is made available to other cognitive processes, which operate at different time scales, from instantaneous and automatic, to reflective and controlled. We are able to think and behave as continuing entities because the salience of information for different organismic goals is represented by affective processes at different time scales and levels of explicitness. An organism that can use that affective information in the process is a self.
This suggests that if the ability to access affective information is lost then self-awareness would also be diminished. Thus as we suggested above a key to the experience of depersonalisation in the Cotard delusion is the profound loss of affect associated with extreme depression. This suggestion is almost correct but it ignores another stage in the production of depersonalisation. After all, from what we have said so far affective processes represent the self-relevance of information. If the consequent feelings are unavailable the world should feel not significant for the subject. That is to say the subject might feel detached from the world or as if the world was emotionally inert. But it seems an extra step from a lack of affective experience to the feeling or thought of non-existence. Of course the step might be a small one. This was the idea of Gerrans in his pioneering work at the dawn of the millennium. He suggested that there was such an intimate connection between affective experience and the self that any profound involuntary change in affect would be felt as a change to the self. However since then interesting work on depersonalisation disorder has provided a deeper understanding of the phenomenon. That work draws on the predictive coding theory of cognitive function.