by Guy Claxton
The classic description of creativity divides it into four phases: preparation, incubation, illumination and verification. (Martindale’s ‘inspiration’ corresponds to incubation and illumination, while his ‘elaboration’ corresponds to what in a more scientific context would be called verification.) During the preparation phase, information is gathered and analysed through focused attention or d-mode, in which the brain acts as if the neuronal clusters were relatively sharply demarcated, and trains of associations unfold in a relatively conscious, relatively conventional manner. If the problem in hand is routine this mode will suffice to generate a solution. However, if the problem is more unusual, this way of knowing will result in a series of blind alleys. Activity rushing through tightly delineated channels will not be able to spread out broadly or slowly enough to make simultaneously active the remote associations on which the creative solution rests.
But if someone is able to move into the incubation phase, the sharp inhibitory surrounds which d-mode employs to keep activity focused and corralled, and which tend to turn gentle valleys into functional canyons, are relaxed, and the wider distribution of activity across the brainscape allows a greater number of different foci to become active at the same time. The pattern of activation in the low-focus brain resembles more the one produced by a handful of gravel flung scattershot across the surface of a still lake, than the linear sequence of epicentres created by the ‘skimming’ of a smooth flat stone. Now if the residual activation from the earlier, preparatory stage remains – if the problem has been put to the back of the mind without being forgotten entirely – the neural clusters that correspond to the problem specification will still be primed. (Uncreative people, as well as having lost the knack of entering the low-focus state, may also be unable to retain this background level of priming: they do not know how to put something ‘on the back burner’ without it falling down behind the cooker.)
As creative people go about their business, the normal exigencies and incidents of daily life will keep activating thousands of concepts and clusters throughout the brain. If one of these should inadvertently facilitate a link between previously unconnected, but primed, parts of the network, there may be just enough added activation to make an image or a metaphor exceed its threshold and shoot into consciousness – producing an ‘insight’, an illumination. Finally, during the phase of elaboration or verification, the focus of activation may narrow again in order to explore in more detail the implications that have been opened up.
If creativity is associated with forms of brain activity that are dim and diffuse, and if this is because, in such a state, a greater number of different foci can be simultaneously active, then we might expect that forms of unconscious awareness – subliminal perception – would show the same increase in diversity of associations. Specifically, if an idea is activated unconsciously, its associative ripples may extend out more widely than if its activation is concentrated to a degree that produces clear consciousness. There is some evidence that this is the case. Recall the study by Bowers and his colleagues (in Chapter 6) in which subjects tried to find a single word that was a remote associate of each of fifteen words presented in a cumulative list. Spence and Holland have used the same type of materials to examine the effect of unconscious perception. Their subjects were given a list of twenty words to memorise, of which ten were remotely associated with a single word such as, in the Bowers example, ‘fruit’. The other ten words were of the same general type and familiarity, but were not linked together in this minimal way.
Prior to learning the list, some of the subjects were exposed to the word ‘fruit’ presented subliminally; some saw ‘fruit’ presented consciously; and some were shown only a blank screen. The results showed that the subjects who perceived ‘fruit’ unconsciously recalled more of the associated words on the list than did either of the other two groups. Spence and Holland interpret this result to mean that having an object clearly in conscious awareness reduces the range of associates that are active in memory to its ‘immediate family’; while a stimulus that does not quite reach the focus or intensity required for consciousness subliminally primes a wider circle of associations. Focal consciousness, we might surmise, is associated with the concentration of a limited pool of ‘activity’ within a smaller area of the memory network.
I showed in Chapter 4 that incubation also supports better thinking through allowing time for false starts and erroneous conceptualisations to fade away, and to be superseded by a different approach. We can now see how the brain makes this process of ‘reappraisal’ possible. Imagine that activity in the neural network is flowing along a channel and comes to a point of choice – a T-junction. Which way is it to go? Under normal circumstances, we can assume (for purposes of illustration) that all the activity has to follow the most well-established route. If one arm of the T is worn deeper, and/or is more highly primed, than the other, then that will be the path that is preferred. We can represent the relative facilitation of each pathway, as in Figure 10, by the thickness of its line. At each junction, the activity has to ‘choose’ the thicker line.2
Now suppose the starting point for thinking about a particular problem, given the way it has been initially construed, is at point ‘A’; and where you need to get to, the ‘solution’, is at point ‘!’. If you follow the thickness of the lines, you will see that the nature of this bit of the network is such that you can never get from A to !. You just keep going round in a circle. However, if for some reason you were to stop thinking about the problem in terms which require you to start at A, and were instead, by accident, to access the same bit of circuitry via another point, B, you would ‘magically’ find that now you are able to get from B to ! very easily. ‘BING!’, as Konrad Lorenz would say. You have an ‘insight’. When you drop out of d-mode, and just let the mind drift around all kinds of ‘irrelevant’ or even ‘silly’ associations, you may well, by chance, find yourself thinking not about A but about B – and the recalcitrant solution suddenly becomes obvious. Thus it is, for example, that the most effective antidote to the ‘tip-of-the-tongue’ state is to stop trying to recall the word that stubbornly refuses to come to mind, and to allow yourself to drift off, or to do something else. And then, at some unpredictable moment, the word comes to you, out of the blue. The head-on assault contains the seeds of its own destruction, while the sideways approach, in which you allow the undermind to sneak up on the item you want, wins the day. That is why brainstorming and daydreaming are – as creative people have always known – effective ways of knowing: they capitalise on the brain’s biochemistry.
Figure 10. A map of neural pathways, showing how a change of entry-point can make an insoluble problem soluble. Activation has to follow the thicker line at each junction.
This neural model also makes it clear why creativity favours not just a relaxed mind, but also one that is well-but not over-informed. In those parts of the neural network that represent the most familiar or routine areas of life, the continual repetition of patterns in experience may have carved out mental canyons and ravines so steep-sided that even when excitation is generally increased and inhibition relaxed, the course of activation flow will still be set. We cannot but construe the world in terms of concepts that are so engrained. However, where the brainscape is contoured enough to formulate an interesting problem, but not so deeply etched that a single approach is inescapable, then moving to the broad focus mode may well reveal novel associations.
It is widely assumed that the total amount of the brain that can be active at any one time varies only within a circumscribed range: there can be only so much activation to ‘go round’. At higher levels of arousal the pool may be increased somewhat, but it is clear that if activation were allowed to proliferate unchecked, cognition would lose any sense of direction or definition at all. We might experience an entertaining psychedelic firework display of ever-expanding associations and allusions, but we would rapidly become swamped by them, unable to discriminate the useful and pert
inent from the random and trivial. (This is exactly what happens in certain kinds of brain disorder. A famous case was described by the celebrated Russian neurologist A. R. Luria in his book The Mind of a Mnemonist.)3 As patterns of activity move through the brain, there must be inhibitory mechanisms that ‘turn off the lights behind them’4
The assumption of limited resource helps to explain why thinking in words can impede non-verbal, more intuitive or imaginative kinds of cognition, and how it is possible to become clever at the expense of being wise. Some of the conceptual hollows in the brainscape are labelled; they have been given names, like ‘Jane’ or ‘breakfast’ or ‘cat’. Names naturally pick out and focus attention on those features and patterns of the concept that are most familiar and essential: they tend to be associated with the nub of features at the bottom of the hollow, rather than with those that are on the slopes. Slightly fancifully, we might extend the landscape metaphor by planting a tall flagpole at the centre of such articulated concepts, at the top of which flutters a flag bearing the concept’s name. This image will serve provided we remember that the ‘flag’ represents another set of neural patterns, to which the concept is linked, corresponding to the way the word sounds, looks, is spoken and written.
As a child learns language, the flags proliferate, and themselves become connected together into strings of linguistic bunting that begin to create a ‘wordscape’ that overlays the experientially based brainscape. Words can be combined to ‘name’ concepts that have no underlying reality, no direct conceptual referent, in that person’s experience. Such verbal concepts are heavily influenced by the categories of a particular culture, and conveyed, moulded, through both formal and informal tuition. Different languages carve up the world of experience in different ways. The Inuit famously have dozens of words for ‘snow’. English has no concept that even remotely resembles the Japanese bushido, the warrior code that combines fighting skill, considerable cruelty and aesthetic and emotional sensitivity. The topography of each ‘plane’, the brainscape and the wordscape, and the relationship between them, represents an evolving compromise between the erosion of the brainscape by first-hand experience, and the dictates of a language about what segments and groupings are to be named.5
Figure 11. Brainscape and wordscape. Some concepts have no labels; and some labels are not directly underpinned by concepts.
This model generates a brain-based account, for example, of the fact (referred to in Chapter 6) that describing faces can make them harder to recognise. Instead of simply focusing on the face as a unique whole, and allowing a rich pattern of connections between neuron clusters to become associated together in the brainscape, energy has to be put into making the face conform to general physiognomic concepts that have verbal flags. One is forced to construct a representation that is based on stereotypes (‘bulbous nose’, ‘bushy eyebrows’); is motivated by what can be said rather than what is there; and is an accumulation of fragments rather than a holistic impression.
When a neural cluster, a concept, is activated diffusely, the intensity of activity at its focus may not be sufficient to trigger the verbal label that is attached to it. We have seen that it is possible to be aware of something, both consciously and unconsciously, without being able to retrieve its name. But when an adequate pool of excitation has concentrated at the epicentre of the concept, that activation may flow into, and activate, the representation of the name, which may, in turn, set off a chain of verbal descriptions and associations. And when verbal propositions become activated, the total reservoir of activation is depleted. The fact that the total pool of activation is limited means that activation of a portion of the wordscape must be at the expense of other possible activations and movements within the non-verbal parts of the brainscape. So the more activation that has been ‘syphoned up’ into the wordscape to subserve an attempt to construct conscious theories and explanations, the less is left to activate other areas of the brainscape. Activation that is focused and verbal tends to support concepts and ways of thinking that are more highly abstract, lacking broader detail and resonance.
In particular, widespread, low-intensity activation incorporates into the representation of a situation more of its personal resonances and connotations. The more the ripples of association spread out, the richer the pattern of meanings that is activated. Objects of perception or thought are imbued with greater significance, because their representation is shot through with a person’s felt concerns: their hopes, fears, plans and interests. Situations therefore ‘make more sense’; we know more clearly where we stand when the sensory impressions they generate are grounded and glued together by feeling. Because attention is broadly and dimly distributed, these elements of feeling may not themselves emerge into the bright light of consciousness, but nevertheless their activation ensures that perception is suffused with significance.
Felt meaning is embodied. When we sense significance deeply, it affects us physically. We do not simply understand: we are ‘touched’ or ‘moved’. Much of the warp of significance that is woven into the fabric of sensation consists of such bodily connotations. In general, the inner as well as the outer senses contribute to the overall flow and pattern of the brain’s activity, and when activity is diffused, the state of the viscera and musculature, any bodily felt emotions, needs or threats, are incorporated within the representation as a whole.6
Conversely, as the focus of activation tightens, the image of the world that is created becomes more abstract, more intellectualised, and less rich in meaning and feeling. David Gelernter, who has reviewed much of the evidence for the link between focus and emotion, concludes:
As we inch upwards, gradually raising or tightening our focus . . . thinking starts unmistakeably to grow numb. We are less and less able to feel our recollections; we merely witness them . . . Thought loses its vividness . . . In the end we are left to the cold comfort of logic alone to peg together a powerful and penetrating – [but] numb and pale – thought-stream and drive it forward.
The trains of thought that may be stimulated in the wordscape are also likely to be more rigid, more stereotypical and more defined by the conventions of the linguistic culture at large, than the patterns of the brainscape. Thus it may well be harder – as many creative people have argued – to be original in propositions than in intuitions, or to unearth and question cultural assumptions that are embodied in the very way the wordscape is constructed.
One crucial aspect of the functioning of the brain remains to be explored: how and where and why does it create consciousness? Even to pose the question in this way makes one important assumption: that consciousness is a product of the brain, rather than, for example, a universal property of all matter, or a signal from elsewhere that is picked up by the brain in the same way that a broadcasting channel is detected and transformed by a television receiver. Both of these views have lengthy histories in philosophy and religion, but I shall join the neuroscientific consensus in seeing consciousness as a correlate of certain kinds of activity that occur only in nervous systems of a particular kind and degree of complexity. The prima facie evidence for this starting point is, after all, overwhelming. We know through direct experience that members (or, if we are being absolutely precise, at least one member) of one complex species – homo sapiens – possess consciousness, whereas the idea that amoebas, daffodils or pebbles are conscious is at best a conjecture or a projection. And we know that it is damage to the central nervous system – rather than to the liver or the lungs, for example – that results in alterations to or even loss of consciousness.
But if consciousness is a property of brains, we can be sure that it is not a property of individual neurons. We see and think in terms of concepts and images that are associated with large groups of interwoven neurons, not single cells. And we may also discard the idea that consciousness can be localised within some particular area or structure of the brain. Despite many attempts to find such a specific anatomical substrate for consciousness, none has b
een discovered. Nor should we really expect it to, given that we know that a single neural cluster is itself distributed widely throughout the brain. Descartes thought that the pineal gland, situated right at the centre of the brain, was the ‘seat of the soul’ and, as Daniel Dennett describes it, the door into the mental cinema where consciousness is projected on to a screen. But we now know without doubt that there is no localised ‘headquarters’ in the brain to which all inputs are referred, and from which all orders are issued.7
We have to think of consciousness, therefore, as associated with states of the nervous system rather than places. So our question has to be: what are the conditions which are necessary and/or sufficient for the brain to make its activities conscious? The short answer is that nobody knows for sure. Finding an answer to this question is the ‘holy grail’ of neuro- and cognitive science at present – as it has been, recurrently, for philosophers and theologians over the centuries. But there are some clues. First, consciousness is associated with intensity. The strength or the concentration of activation in a cluster of neurons seems to play a role. Signals that are loud or bright or shocking grab our conscious attention, and stimuli which are being processed unconsciously can be ‘boosted’ into consciousness if their magnitudes are increased.