The Neurology of Anticipation

In the previous post, I asked what is the difference between having a painful experience myself and feeling sympathy for someone else’s pain.  The answer seems pretty clear.  My experience of my own pain is neurologically ‘hard-wired’, but there is no direct neural connection between other persons and myself which exposes me to their pain.  Things could have been different.  If we, like the Na’vi of the movie Avatar, had the physiological equivalent of USB-ports which allowed us to connect our nervous systems at will, then we could experience the pains, pleasures, and other sensations of other people while we were connected to them.  The ‘privacy’ of our minds is an anatomical limitation, not a metaphysical necessity.

A parallel question can be asked about future experience.  What is the difference between anticipating my own pain and having sympathy for a friend whose pain I can foresee?  As in the case of present pain, there is a vivid difference in my experience between anticipating having pain, and expecting you to have pain.   There is nothing resembling a direct neural connection between myself, now, and myself in the future, to explain this difference.  Nevertheless, neuroscience can help us understand it: why it is so vivid, why expectation of my own pain makes me anxious rather than just sympathetically concerned, why foreseeing pain in my own future feels like an unavoidable problem for me in a way that foreseeing your pain does not.

Another difference is that sympathy for others, unlike self-concern, seems at least partially under conscious control.  When we feel we cannot afford it, we tend to dial it down or switch it off altogether.  It feels optional, in contrast to the sense of unavoidability that clings to self-concern.

In his book, Hardwired Behaviour, Laurence Tancredi of the New York University School of Medicine offers some clues to an answer from a neurological perspective.  The engine of self-concern is a cyclical interaction between the limbic system, which generates our emotions (giving emotional ‘meaning’ to experience), and the prefrontal cortex, which makes plans and decisions.

Emotion and Memory: The Limbic System’s Role in Self-Concern

The amygdala, which belongs to the limbic system, assigns emotional significance to perceived events.  It is part of the neural apparatus of motivation. (p 124)

The amygdala generates powerful motivating emotions, including fear and anger.  It will induce fear when a person perceives similarity between her present experience and traumatic or frightening experiences in her past.

When my family crossed Sutil Channel in B.C.’s Desolation Sound by canoe on a windy day over fifteen years ago, my wife grew afraid.  The wind blew up whitecaps; some waves broke over the bow.  Our two children, young then, huddled under a tarp.  The high waves and deep troughs made it hard for me to steer.  It was all Claudia and I could do to make headway.  We were not sure we were strong enough to make the crossing; but turning back did not seem like a safer option.  We were committed.

Eventually we made it to the Subtle Islands and collapsed on the beach.  We were approached by a man who told us we were on private property; he had bought the islands since our last camping trip in the area, and was planning to build a resort.  I invoked the right of mariners to seek refuge from a storm, and informed him that in Canada, unlike the U.S., ocean beaches are public property.  We weren’t moving that night – we were safe.

All four of us were anxious that windy day.  The kids retreated from the experience, under the tarp.  Claudia and I paddled with all our strength.  In the bow, Claudia took the brunt of the spray; she also had less control over the situation than I had as steersman.   (But only slightly; the waves determined our course more than I could.)  She was vividly aware of the danger to our two young children.

Since that day, Claudia has not felt the same about wind and small boats.  To her credit, she still goes kayak-camping with the rest of us; but winds make her afraid.   That day’s experience left a lasting record in her brain, not just memories, but an emotional response to similar experiences.  Ocean winds that would not have worried her before that crossing now trigger – not panic, because Claudia is made of sterner stuff – but still, acute anxiety.

This is a fairly extreme example of what Tancredi is talking about.  The neural mechanisms involved are quite generalized, responding not only to physical threats to survival but also to highly socialized events.  Long known for its role in motivating fight-or-flight behaviour, the amygdala has more recently been shown to respond to gains and losses.  In a chapter titled, “The Biology of Money,” Tancredi describes the particularly strong effect that monetary gains and losses have on the amygdala.  Tancredi’s example  of someone deliberating how to invest a substantial amount of money invites us to compare the phenomenology of a investment decision with its neural correlates.

When a person experiences a significant investment loss, the amygdala “revs up, creating apprehension, anxiety, and fear, and these in turn cause the hypothalamus to accelerate, resulting in a full range of parasympathetic bodily symptoms – increased breathing, racing pulse, and profuse sweating.”  (p 125) The expectation of loss, as opposed to an actual loss, also affects the amygdala, “creating that unpleasant feeling of anxiety in the pit of the stomach that we associate with uncertainty and the fear of losing something valuable.”

Tancredi discusses the conditioning effect of repeated losses.  When the first experience of loss occurs, the hippocampus, stimulated by emotional pain, stores it in memory.  A later situation is recognized as similar to the earlier one, prompting the amygdala to create more fear.  “Thus, for some interested in investing in stocks now, the fear and anxiety experienced during the 1990’s downturn may stop them from taking subsequent risks on otherwise rational investments.” (p 126)

The danger of capsizing at sea and the hazard of an investment loss are both threats to self, activating self-concern – which, in these cases, includes a visceral emotional response from the amygdala.  But not all threats are threats to self; we hold other values that can be threatened.  For example, most people would agree that the floods now going on in Pakistan are a bad thing.  We don’t like to hear news of other people’s suffering and loss of life, even if we don’t know them; and many of us have been moved to contribute to the relief effort.   Selfless values as well as self-concerned ones can be threatened by events.  When they are, is our neurological response the same, or similar?

Tancredi’s list of parasympathetic symptoms suggests that it is quite different.  Hyperventilation, racing pulse, and sweating are not normally triggered by one’s fears for the well-being of others.  Those symptoms occur when our own interests are on the line.  (An exception, or borderline condition, may be found in cases where there is strong emotional attachment to others, such as the attachment of parent to child, child to parent, or lover to lover.)

Another important limbic structure is the hippocampus, which lays down long-term memories.  Only a tiny portion of our experience is remembered for more than a day or two.  Tancredi suggests that the hippocampus creates an association between emotions and perceived events.(p 125) It selects experiences that have emotional significance – that ‘mean something’ to us – for retention.  At a later time when we perceive ourselves to be in a situation similar to one that had emotional significance, the hippocampus retrieves the earlier memory and its emotional colouring, thereby providing the executive portions of the brain with guidance in dealing with the new circumstances.

Elizabeth Phelps’ review of recent research reveals an emerging picture of an even closer relationship between the amygdala and hippocampus, between emotion and memory.   The amygdala “modulates” the hippocampal function of  encoding and storing memories.   When emotion-inducing stimuli are encountered, the hippocampus influences the amygdala’s response.  Phelps describes both the amygdala and hippocampus as “independent memory systems” which “act in concert when emotion meets memory.”

Studies of specific neurological deficits throw the roles of the two structures into sharp relief.

In a classic fear conditioning paradigm, during which a neutral blue square is paired with an aversive shock to the wrist, patients with amygdala damage fail to show a normal physiological fear response to the blue square, even though they are able to report that the blue square predicted the shock. Patients with damage to the hippocampus show the opposite pattern. That is, they demonstrate a physiological arousal response to the blue square, but are not able to consciously recollect that it was paired with the shock.

Although the patient knows the blue square is always followed by a painful electric shock, he does not learn to fear the square if his amygdala is knocked out.  Conversely, if his hippocampus is not functioning, he will not cognitively learn that the blue square is predictive of pain – nevertheless, the amygdala, unaided, learns to fear the square.   The amygdala, it appears, is close to the emotional heart of self-concern.

The Role of the Prefrontal Cortex

Tancredi describes the prefrontal cortex as a command-and-control centre.  “Virtually every operating system of the brain has direct communication with the prefrontal cortex,” including limbic structures: the amygdala, the hippocampus, the hypothalamus, and the thalamus.(p 128) The cortex “organizes purposeful control and concentration of thinking to achieve well-defined objectives.” (p 127) It holds information relevant to the task at hand in working memory, and excludes what is irrelevant.

The prefrontal cortex also modulates an organism’s emotional response to events.  “When the limbic structures overreact to fear, the prefrontal cortex is often able to intervene and put reactions into perspective through careful evaluation of the data at hand.”

By the same token, the prefrontal cortex and frontal lobes in general depend on the limbic structures for guidance in making “rational” decisions.  Patients with early childhood injuries of the prefrontal cortex, especially the orbitofrontal cortex…show striking thought abnormalities resulting from an inability to benefit from information sent to limbic structures such as the amygdala. (p 128)

The use of the word “rational” in the above passage is telling.  In a 1995 paper, “Affective Computing,” R. W. Picard summarizes some of the central findings in A.R. Damasio’s landmark book, Descartes’ Error: Emotion, Reason and the Human Brain.

Damasio’s patients have frontal-lobe disorders, affecting a part of the cortex that communicates with the limbic system. Otherwise, the patients appear to be intelligent, and unusually rational.  However, these same patients suffer from an impaired ability to make decisions.  Years of studies with frontal-lobe patients indicate that they spend inordinate amounts of time trying to make decisions that those without frontal-lobe damage can make quite easily.  For example, the mere task of choosing a date to schedule an appointment can lead these patients through abnormally long chains of decisions, perhaps without ever reaching a decision, until a date is imposed upon them by someone who is tired of waiting for their response.

Emotion, it seems, provides rationality with a vital motivational kick-in-the-pants.  Picard cites Johnson-Laird and Shafir, who pointed out:

…the inability of logic to determine which of an infinite number of possible conclusions are sensible to draw, given a set of premises.  Consider: how do you decide which path to take given some evidence? There is not time to consider every possible logical constraint and associated path. Emotion does not merely play a tie-breaking role in making certain decisions.  Rather, it appears to be essential in learning the biases required to construct rational responses.  Damasio’s findings provide neurological support that there is no “pure reason” in the healthy human brain – emotions are vital for healthy rational human thinking and behavior.

Although the roles of the prefrontal cortex and the limbic system are very different, both are essential in producing appropriate self-concerned behavior.  Tancredi describes an experiment by Bechara, Damasio and others in which subjects were allowed to choose cards from four stacked decks.  Two of the decks were biassed towards the interests of the subjects, two against.  After a number of trials, normal subjects developed a correct ‘hunch’ as to which decks were biassed in their favour, and began to choose accordingly.  Subjects with bilateral prefrontal lobe damage failed to detect the bias, and chose disadvantageously. (p 129)

The subjects’ anticipation of a favourable or unfavourable result, before turning over a card, was measured by skin conductance responses (SCR’s).   After picking a few ‘punishment’ cards, during what the authors call the ‘pre-hunch’ stage, normal subjects all began to show anticipatory SCR’s, but none of the six injured patients did.  Around this time, the normal subjects also began to preferentially choose the ‘good’ decks.  Despite their SCR and behavioural responses, the normal subjects said they “did not have a clue about what was going on” until they had worked through about 50 cards.  Then they began to report accurate ‘hunches’ about which decks were riskier, although they could not say why.

The authors describe a ‘conceptual period’ which was reached by seven of the ten normal subjects and by three of the six prefrontal patients, in which, after turning 80 to 100 cards, they had ‘figured out’ the bias in the stacked decks and were able to report on it accurately.

Remarkably, the three normal participants who did not reach the conceptual period still made advantageous choices.  Just as remarkably, the three patients with prefrontal damage who reached the conceptual period and correctly described which were the bad and good decks chose disadvantageously.  None of the patients generated anticipatory SCRs. Thus, despite an accurate account of the task and of the correct strategy, these patients failed to generate autonomic responses and continued to select cards from the bad decks. The patients failed to act according to their correct conceptual knowledge.

It appears not only that emotional experience helps the rational mind reach correct factual conclusions, but, even when we know the truth, emotion is needed to prod us into rational action.

The prefrontal cortex moderates our emotional responses, bringing ‘reasoned judgement’ and ‘second thought’ to bear against the impulses of the moment.  It has a reputation for rationality.  But in light of the neurological evidence, it seems better characterized as the origin of just one component of rationality: ‘cool rationality’.  The sensible self-interested behaviour which we call “rational” depends at least as much on ‘warm rationality’ – the  emotional affect arising from the limbic system.

Dopamine: the Motivator

All emotion, positive and negative, depends on the neurotransmitter dopamine.  Tancredi writes:

The brain at its most basic is designed to obtain rewards and avoid punishment. … Motivated behaviours, such as remembering and learning, and the cognitive and emotional tasks of processing reward-directed behaviours all do their work primarily through the dopamine system.  This is very simply all the connecting cells that use the neurotransmitter dopamine to send and receive signals – that is, to communicate with each other.

The brain’s reward circuitry consists of the frontal lobes and limbic structures, working together as a highly integrated network to produce feelings and behaviours. (p 133)

Rewards and punishments reinforce not only behaviour but feelings.  To human beings, money is a reward – one that acts “on more different pathways of the reward system than do natural rewards such as food and water.”  Gambling for money is as addictive to some people as cocaine.  Dopamine plays a key role in addiction.

The brain’s fundamental reward mechanism is the production of dopamine; it fuels our emotions and hence motivates our behaviour.  The levels of dopamine released into our synapses determine the salience of our experiences – whether they interest or bore us – and the affect we feel towards them – whether we like or dislike them.

What Does it All Mean?   Reinforcement of the Temporally Extended Self

Neuroscience has much to say about what happens in the brain when we feel self-concern.  But does it deepen our understanding of the questions which I set out to address in the Phantom Self project?  When (if ever) is self-concern appropriate?  What is the relationship between self-concern and personal identity?   What differences (if any) are there between self-concern and concern for other people, and for the non-human world?

I think it does.  First, the fact that special physical symptoms are associated with emotions of self-concern – parasympathetic symptoms such as sweating and elevated pulse – may partly explain why those emotions are phenomenologically distinct from others.  Those emotions involves distinctive brain structures which are not usually, or not so strongly, involved in other-directed emotions.

Second, the neuroscientific account of anticipation highlights the importance of reinforcement cycles.  Anticipation of pleasant or painful experience is a result of a person’s reinforcement history.  Patterns of perceived similarity of experience create expectations of rewards and punishments, modifying the brain so as to produce emotionally-charged anticipation.  Rewards and punishments stimulate the dopamine system, eliciting attention and memory more strongly than do affect-neutral experiences.  Memories trigger affect-tinged anticipation.  When expectations are met, in the form of new rewards or punishments, positive reinforcement of anticipation occurs; pattern recognition and the corresponding emotional affect will be stronger next time.

In short, causal mechanisms that preserve memories of anticipation also preserve self-concern, and give us a sense of persisting through time.

If, for example, I take a Hawaiian vacation by means of information-based teleportation,  my expectations of a tropical vacation experience will be met, and I will be reinforced in my inclination to anticipate future experience.

The account of my expectations being met can be expressed neutrally with respect to the question whether the person who enters the teleportation booth in Vancouver (GordonV) is or is not the same person who emerges in Hawaii (GordonH).  It is not necessary to suppose that I actually remain the same person, in order to understand how my self-concern (and my corresponding conviction that I have survived) is sustained.

The facts are: GordonV, in Vancouver, anticipates the experience of a Hawaiian vacation; and GordonH, in Hawaii, actually has the experience of being in Hawaii.  Moreover, GordonH remembers anticipating the experience while he was still in Vancouver.  That can be restated without the suggestion of identity, by saying that GordonH quasi-remembers GordonV’s anticipation.   GordonH’s experience is that of having achieved his goal of a Hawaiian vacation.  His expectations are fulfilled – or to speak more cautiously, the effect on his brain is just as if they were.

The anticipation GordonH (quasi-)remembers strikes him as eminently rational.  If GordonV had any doubts as to whether or not he would survive the teleportation process, GordonH cheerfully lays them to rest.   He does so because he has the normal experience of comparing his memories of anticipation of experience to the actual outcome.  Nothing, I think, gives us a solider sense of persisting through time than comparing expected experience to the actual outcome.  I was afraid it would be too hot, I say to myself, but today’s actually coolish.  I think it’s going to rain.  I wish I’d brought a jacket. In such banal thoughts, we affirm our identity through time and the rationality of our self-concern. 

In the midst of such thoughts, the rationality of self-concern seems not just permissible, but required.  Standing on that beach in Hawaii, I think I would have been deluded if, in Vancouver, I had not had normal self-concern for my vacation experience.  That was dumb, not packing my jacket!

A causal mechanism that preserves the memory of anticipation will preserve a sense of self.  It need not be the ‘normal’ cause, that is, the continued existence of a functioning brain. Any such causal mechanism will do.

To see this, suppose that I ‘anticipate’ your Hawaiian vacation.  By that I mean that I know you are planning a Hawaiian vacation, and that I imagine your experiences in Hawaii vividly, from a first-person perspective, and that I experience qualitatively similar emotions when I imagine you surfing, enjoying the sun, having daiquiris, as I would if I expected to visit Hawaii myself and were anticipating my own experiences.  My experience when contemplating your incipient Hawaiian experience might be called ‘quasi-anticipation’.

I think I could work up quasi-anticipation of your vacation.   I don’t even need to know you very well.  It is not important that the content of the quasi-anticipation be accurate.  I might quasi-anticipate your terror as you attempt to surf a twelve-foot wave, and your pain and confusion as you wipe out in a life-threatening tangle of board, wave, and rocks.  I might be far off the mark: if you’re a confident, expert surfer, you would not be terrified, and the probability of a wipe-out would be low.   What makes it quasi-anticipation is (1) I imagine the experiences ‘from the inside’, i.e. from a first-person perspective, and (2) the emotions attached to these imagined experiences ‘feel like’ – are  phenomenologically similar to – the emotions I feel when I anticipate my own future experiences.   Even in the ordinary case when I anticipate having experiences, I often get them wrong, both in details of external events and in how I will feel about events that I predict accurately.

For my quasi-anticipation of your experience to actually ‘feel like’ anticipation of my own experience, it must be similarly motivating.  If I were not motivated to take action that would affect the quality of experience I expect you to have in Hawaii, my attitude would not be quasi-anticipation, but something less.  If I were unwilling to contribute to the cost of your plane ticket, for example, my expectation of your trip should not count as true quasi-anticipation.  (This is not an offer!  I refuse to quasi-anticipate your vacation unless I know you well, and like you too.)

Why do we all anticipate our own experiences, but rarely quasi-anticipate the experiences of other people?   One explanation is that, in quasi-anticipation of other peoples’ experiences, the normal neurological reinforcement mechanisms are lacking.   If  I quasi-anticipate your vacation – make arrangements, buy your ticket, drive you to the airport – the moment never arrives when I’m standing on the Waikiki strip saying to myself, “Here I am at last in Hawaii!  It’s as great as I expected, if not better!”, or alternatively, “This is a Hawaiian vacation?  I hate these hotels – what a disappointment!”

If it is my own vacation, not yours, then either outcome – satisfaction or disappointment – reinforces my conviction that, when I planned the trip, I was right to assume that the experiences I imagined having in Hawaii would be my experiences, not someone else’s – that my anticipation was rational.  Even if the reality does not match my expectations, I am reinforced in my belief that I was justified in anticipating having some experiences in Hawaii.  The next time it occurs to me to plan a vacation, I will again anticipate the trip with as much interest as I did before.

No such reinforcement mechanism exists if I quasi-anticipate your Hawaiian vacation.  No one will find himself on the sand remembering (or quasi-remembering) my anticipation and thinking, I’m glad I came.  The reinforcement mechanism for my quasi-anticipation is missing.  If, a year from now, you announce that you would like to take another vacation, I am likely to take less interest, because my quasi-anticipation was not reinforced.  In general, because of lack of reinforcement, quasi-anticipation of  other peoples’ experiences is likely to be extinguished.

Conversely, where such reinforcement mechanisms exist, anticipation of experience, and, more generally, self-concern, have what they need to flourish.  Transformations that reinforce anticipated experience, and are beneficial in other ways are likely to find satisfied customers.  Such transformations include information-based teleportation, fission, and fusion.  (With some qualifications, they also include post-death reconstruction from backup information: the ‘life insurance’ application described in the story, “Phoenix”.)

In my post on Ray Martin’s book, ­Self-Concern, I wrote, “Many people, if contemplating undergoing a transformational process that will result in more than one coexisting successor, would regard those successors as ‘mere duplicates’, and the transformation as one that would end their life.”   A basic understanding of the reinforcement mechanisms of self-concern tells us that they are likely to get over it.  Although to try such a radical transformation the first time may require courage, once expectations are met it will quickly become a matter of routine.  Full-blown, normal self-concern will thrive if psychological continuity is preserved by any cause.  This is an empirical claim; I believe it will be borne out as advancing technology makes new opportunities for transformation  available.


Bechara A, Damasio H, Tranel D, Damasio AR (1997): “Deciding Advantageously Before Knowing the Advantageous Strategy”,  Science, vol. 275.

Damasio, AR (1994): Descartes’ Error: Emotion, Reason and the Human Brain, Grosset/Putnam, New York.

Johnson-Laird, PN and Shafir, E (1993): “The interaction between reasoning and decision-making: an introduction,” Cognition vol. 49.

Martin, Raymond (1998), Self-Concern, Cambridge Studies in Philosophy

Phelps, Elizabeth (2004): “Human emotion and memory: interactions of the amygdala and hippocampal complex”, Current Opinion in Neurobiology 2004, 14:198–202

Picard, RW (1995): “Affective Computing”, MIT Media Laboratory

Tancredi, Laurence R (2005), Hardwired Behaviour, Cambridge University Press.

Click here to view other posts on the Phantom Self home page.

One Response to “The Neurology of Anticipation”

  1. […] some person in another dimension is also me.” Implicitly, we care about whether we can anticipate future experiences. Not, unfortunately, what the ultimate truth of identity really […]

Leave a Reply