by Guy Claxton
I have suggested that one way of expressing this disparity between conscious and unconscious is in terms of two thresholds, a lower one, above which the undermind becomes active, and a higher one, above which information enters consciousness. The closer together these two points are, the more ‘in touch’ with the unconscious we are, and the more complete is our conscious awareness of what is happening across all the mental realms. The further apart they are, the more our conscious perception is impoverished. This quantitative notion of thresholds is rather crude, but it enables us to formulate an important question: what is it that determines how near or how far apart the two thresholds are? More generally, is the relationship between conscious and unconscious forms of awareness a dynamic one, subject to change, and if so, what are the forces that control it? In cases such as Masling’s, it seems clear that some information received unconsciously can cause a considerable amount of nonverbal discomfort, and that, as a result, it is gated out of consciousness. The therapist’s remark occasions a very rapid raising of the conscious threshold. Perhaps it is specifically things that are threatening that cause the conscious threshold to shoot up.
Corroboration for this supposition comes from a phenomenon called ‘perceptual defence’, which has been known to experimental psychologists since the 1940s. In the classic version of these studies, a subject is repeatedly flashed a word very briefly, and the exposure duration is gradually increased until she is able to identify the word correctly. Some of the words used are neutral, while others are vulgar or disturbing in some way. The charged words do not become consciously visible to the subjects until they are exposed for a considerably longer duration than the neutral words. If recognition and consciousness are the same thing, this result is simply incomprehensible. How could one selectively raise the perceptual threshold for things that have not yet been recognised? Unconscious perception provides the only explanation: the taboo word is recognised unconsciously, and the upper, conscious threshold is immediately raised in order to try to protect consciousness from the threat or emotional discomfort that the word has generated.2 Jerome Bruner, one of the instigators of research on unconscious perception back in the late 1940s, used to use the analogy of the ‘Judas eye’, the peephole used by the doorkeeper at a ‘speakeasy’ to distinguish between bona fide members, for whom the door opens, and undesirables, such as the police, who are shut out. Without the Judas eye, one could only tell friend from foe by opening the door – and then it was too late.
Conversely, we can demonstrate that access to information in the undermind that is dubious, not because it is directly threatening, but because it is faint or ephemeral, can be increased by making subjects feel more relaxed and ‘safe’. One way to do this is to ask subjects to express this weak information without feeling that they are under pressure, or being judged in any way. Normally when people are asked to recall something previously shown, they feel that they are being tested.3 Psychologists’ experiments are designed to be hard enough for people to make some errors: if everybody got everything right, the data would not differentiate between different conditions, and it is these differences that tell us interesting things about how the mind is working. And nobody likes making mistakes. It may be that the normal type of memory test, for example, underestimates how much people really do know, because the feeling of being on trial makes them adopt a cautious attitude.
Experiments by Kunst-Wilson and Zajonc and others have managed to demonstrate this effect. First they show people a sequence of complicated nonsense hieroglyphics. When these are subsequently mixed up with some new shapes of the same kind, subjects are rather poor at picking out the ones they saw before. However, if, instead of asking people in the second part of the experiment to recognise the ‘old’ squiggles, they are simply asked to point out the ones they prefer, they tend quite reliably to choose the ones they saw before. When self-esteem is at stake, delicate unconscious forms of information and intelligence seem to be disabled or dismissed, and the way we act becomes clumsy and coarse. When we are less ‘on our best behaviour’, the glimmerings of knowledge from the undermind are more available to guide perception and action. Sometimes reception is good, and we are able to pick up and use the undermind’s faintest broadcasts. At other times, when we are stressed, only the strongest stations get through.4
The same kind of release from pressure can be achieved by presenting the ‘test’ as if it were a guessing game, rather than a measure of achievement. When we treat something as a ‘pure guess’, we do not feel responsible for it in the same way. We are freed to utter things that come to us ‘out of the blue’, because there is no apparent standard of correctness or success against which they, or we, will be judged. One method that has been used to investigate this idea was first devised to assess the memories of people with severe retrograde amnesia. Such people seem to have completely lost their ability to remember what has happened more than a few seconds ago. If you meet one of these people, then go out of the room and come back five minutes later, they will greet you as if you were a stranger. Give them a short list of words to study, take the list away, and after a short delay ask them to recall the words, and they will look at you blankly and say ‘What words?’ Yet there has for many years been a suspicion that these patients do have some memory; it is just that they are unable to access it deliberately.
The nineteenth-century French physician Claparede, for example, concealed a pin between his fingers when he was introduced to one of these amnesic patients, giving him a prick as they shook hands. On leaving the room and reappearing a few minutes later, Claparede was, as expected, treated by the patient as if they had never met before – yet the patient was curiously reluctant to shake his hand. When queried about this antisocial behaviour, he rather vaguely explained that ‘you never know with doctors; sometimes they play tricks on you’.5 It is no coincidence that it was a painful stimulus that was unconsciously registered, for the undermind is particularly concerned with things that are of significance for our survival and wellbeing.
The suspicion that amnesiacs have more memory than it appears has recently been confirmed in the following way. Subjects are given some time to study a list of words which they are asked to remember, and then the list is removed. A little later, instead of asking the subjects to recognise or recall the words in the conventional manner, they are shown the first two or three letters of a word and asked to respond with the first word beginning with those letters that comes into their heads. As far as the patients are concerned, this a completely new exercise. But the prompts they are given have been selected so that they can be completed with one of the words on the original list – and this to-be-remembered word has been chosen to be less common in the language than some alternative words that could also be used to complete the frame. So if one of the words on the original list was CLEAT, the patients are asked to think of a word that completes the frame CLE—. Without any memory of the list, people would respond to the cue with a more common word like CLEAN or CLEAR, but in fact the amnesic patients tend to produce the rarer word which they recently saw, but cannot ‘remember’.
The words must have been recorded, but the memories only reveal themselves in the way they influence the (apparently) ‘free’ association. It begins to look as if the ‘amnesia’, in these cases at least, is more to do with an inaccessibility of memories to consciousness than with an inability to register what has happened. This same effect has now been reproduced in people with undamaged memories, using subliminal perception techniques. Subjects are presented with a number of words on a screen, one by one, too briefly for conscious perception. If they are subsequently asked to recall the words, they will, like the amnesiacs, say ‘What words?’ Yet, if they are tested with the same ‘free association’ game, the words they did not even ‘see’ are found to have made a significant difference to how their minds are spontaneously working. They behave exactly like the memory-impaired patients.6
The same kind of distrust of faint informat
ion by consciousness has recently been demonstrated by Cambridge psychologist Tony Marcel in a neat study that focused on perception rather than memory.7 His (unimpaired) subjects were presented with very weak flashes of light, so weak that it was hard to tell whether there was anything there or not, and asked to indicate each time they thought they saw a flash. But Marcel asked them to indicate in any of three different ways: by blinking their eyes; by pressing a button; or by saying ‘Yes [I see the light]’. He discovered that these three different methods of answering the same question were not equivalent but gave quite varied results. When people were blinking, they ‘saw’ many more of the weak flashes than when they were replying verbally, with the button-pressing somewhere in between. When the subjects were asked to respond to the light by both blinking and reporting, there were many occasions on which the eyes said Yes while the voice said No. (By measuring the time intervals between stimulus and response, Marcel was able to exclude the possibility that these results could have occurred simply because of reflex blinking.)
Marcel points out how these results challenge our commonsense view of the mind. Our normal assumption is that we have a unitary consciousness which registers ‘what’s there’. If something is ‘there’, you can sense it and report it, and the way you report it ought not to make any difference to whether you ‘see’ it or not. Responding is, we assume, ‘downstream’ of perception, and what comes later in the processing chain should not affect what comes earlier. Under ordinary conditions, this assumption seems to work. But with stimuli that are ambiguous as to whether they are ‘there’ or not, that model of ourselves begins to break down. The method of report now appears to have a retrospective effect on what we see. The medium that is most closely tied to normal consciousness – verbalising – turns out to be the one that ‘sees’ least well, while the one that is most automatic, most engrained, most unconscious, turns out to be the most sensitive of all. There is corroboration for the idea that the more the self is involved, the more cautious consciousness has to be, for fear of ‘getting it wrong’. Clearly a verbal report, ‘Yes, there was a flash’, or ‘No, there wasn’t’, feels like more of a personal commitment than the mere blink of an eyelid – an act that one does not usually think of as requiring close personal supervision or involving much ego investment.
There was another aspect to the experiment which directly replicated the beneficial effect of ‘guessing’, and its superiority over ‘trying’. In each of Tony Marcel’s studies, regardless of which modes of response were being used, subjects’ ability to detect the weak light was less than perfect. However, when Marcel asked people not to try to report the presence of the flash accurately, but simply to guess, their performance magically shot up to almost 100 per cent! To ‘try’ is to have some kind of investment in the outcome. You care, you bother - and therefore you cannot help but be ‘bothered’ if your effort proves unsuccessful. With a pure guess, on the other hand, you feel as if you are plucking an answer out of thin air. And, as we have seen before, when the pressure is reduced, you are able to allow your choice to be guided by unconscious promptings that are adequate for the task, despite consciousness’s lack of faith.
Twenty-five years ago, when I was beginning my graduate work in psychology at the University of Oxford, one of my fellow students was a lanky, bearded Australian called Geoff Cumming. Geoff was investigating the effects of ‘backward’ and ‘lateral masking’ on perception, using much the same kind of procedure as Tony Marcel. A faint image of a letter was projected on to a screen for a brief period, and then followed, after a short, variable delay, by another image, such as a checkerboard pattern. Geoff was looking at the effect of various characteristics of the second stimulus on people’s ability to detect the first event: under some conditions the latter would wipe out conscious awareness of the former. During the course of his experiments, he noticed a rather curious phenomenon. When his subjects were responding at their own speed, they would, under particular combinations of conditions, regularly fail to detect the target letter. But when they were urged to respond as quickly as they could, under the same conditions, they would often make a fast response that correctly detected the target – but would then, a moment later, verbally apologise for having made a mistake! It would seem that they could break through their self-consciousness by responding very fast, and thus were enabled to make use, unconsciously, of the faint information from the first stimulus. But because this information had not been strong enough to make it through into consciousness, they retrospectively concluded that their response had been in error – and incorrectly corrected themselves.8
Some of the deleterious effects of self-consciousness on performance have been vividly demonstrated in certain types of brain disorder. Patients with neurological damage may show a dramatically increased disparity between their ability to function and what they are consciously aware of. Tony Marcel has recently reported the clinical case of a woman suffering from hemiplegia with anosognosia – she has lost the use of one side of her body as the result of a stroke, but seems curiously unaware of her deficit.9 If she is asked to describe herself she will not mention the paralysis. If she is asked to rate her ability to do something that requires the use of two hands – catch a large beach-ball, for example – she will give herself eight or nine out of ten. When asked about herself directly, her consciousness of her condition is remarkably low. However, if questions are put to her in such a way that her self-image is not so directly involved, she then gives quite different answers. If, instead of asking ‘How well can you do these things?’, the questioner says ‘If I were like you, how well would I be able to catch the ball?’, she will give a rating of only one or two out of ten.
When the form of the question allows her to distance herself from her condition, to ‘disown’ it, she is able indirectly to acknowledge it. And this is not just a matter of ‘not wanting to say’, or feeling consciously embarrassed. The evidence is that her reluctance to admit her condition operates ‘upstream’ of consciousness, in the underground departments of the mind where the decisions about what to allow into consciousness are being made. Interestingly, one can create this ‘disinhibition’ of consciousness not only by allowing her to project her disability on to someone else, but by inviting her to talk in a regressive, childlike mode. If you hunker down by the side of her chair and whisper rather conspiratorially, ‘Tell me, is the left side of your body ever naughty?’, she will join in the game and whisper back, ‘Yes, terribly.’ We could speculate that there is a childish sub-personality that needs to exercise less tight control over what can be given access to consciousness – after all, children are used to not being in control of things. Much of the world for them is refractory, and ‘naughty’ is the childhood word par excellence for things that misbehave or will not do what they are told.
We know that, in extremis, people are capable of extraordinary feats of tactical unconsciousness. People suffering from ‘hysterical blindness’, for example, have cut off their consciousness of vision as a result of witnessing something horrific – to protect themselves from further trauma.10 It is possible to raise the conscious threshold on one sensory modality, apparently, in a way that is both non-selective (unlike the ‘perceptual defence’ case) and extreme. Yet such people, though they have no visual experience, manage to navigate their way around obstacles uncannily well. Analogously, functionally deaf people show no response to sounds – they fail to show the normal startle reaction to a loud, sudden noise, for example – yet may respond ‘No’ when quietly asked if they can hear anything. Just as the patient with no memory nevertheless knows, at some level, not to shake hands with the doctor who has hurt them, so it is possible for people with no vision to be able to find their way about, and for people with no hearing to respond to questions.
We are all familiar with more mundane versions of the same phenomena, such as ‘unconscious driving’, which I used as an example in the last chapter. Though the idea of ‘unconscious seeing’ may strike us
at first as weird or paradoxical, it does so only because it conflicts so strongly with our implicit beliefs about how the mind works, and this dissonance simply makes us unaware of how much of the time we respond appropriately in the absence of conscious awareness. If we were ‘seeing unconsciously’ in the absence of any conscious experience at all – if we were able magically to find our way around a strange house in pitch darkness, as it were – we would indeed be stunned. But our reliance on the undermind is conveniently masked by the fact that the ‘hole’ in consciousness is almost always plugged by some other content that is drawing our attention. It is only because our conscious mind is occupied with something else that we usually fail to observe the extent to which the visual information on which we have been relying has bypassed conscious awareness.
We are familiar too with the effect of ‘self-consciousness’ on behaviour as well as perception. Think of someone who is being interviewed for a job they desperately want, or a child who has been specifically enjoined to carry a full cup of tea very carefully. In such situations there is a sense of vulnerability, of a precarious balancing act, the successful execution of which depends on a degree of skill or control that we do not confidently possess. Thus there is anxiety and apprehension. And this leads to a coarsening of motor control, making us clumsy, in addition to the constriction of attention. Under pressure, we seize up, or ‘go blank’. The interviewee fails to understand a perfectly straightforward question. The child concentrates so hard on not spilling the tea that her coordination goes, and she becomes graceless and gauche. It was the very day, in 1984, when my long-standing (and long-suffering) partner had finally finished our relationship that I dived – without any conscious suicidal intention – into the shallow end of a swimming pool and split my head open on the bottom. Losing keys, breaking plates and denting the car are similar symptoms of stress. It is when consciousness is most fiercely preoccupied, usually with a difficult and emotionally charged predicament, that it disregards information (like a large sign saying DEPTH ONE METRE) most egregiously, and siphons off from the unconscious the resources that it needs to function well.