Monday 23 December 2013

Evolution's Greatest Gift


Gifts can arrive carefully concealed or straight out of the box. Either way their consequences can only be envisaged but never seen from the outset.


I recently had a discussion with a philosopher in which he brought up the Wollheimian concept of "seeing-in." He claimed that an expert talent-scout working for a modelling agency would be able to see - literally perceive - potential in the face of a prospective model. Sadly the discussion was interrupted and I didn't get a chance to explore his view further.

It may be the case that he was using the term "perception" figuratively as an equivalent of "appreciation", "evaluation" or "inference". Potential in this sense is a projective notion, specifically a notion about a possible or even likely future. It's an anticipatory account, image or expectation. If I say that someone has potential, I don't mean that I literally see anything. If the person in question leaves the room, the potential I see doesn't walk out of the door with them. So to see something in this sense is not to perceive it in anything like the sense in which we might readily see the shape or size of objects around us.

Expertise may sometimes involve a degree heightened perceptual awareness developed through practice and experience. But to see indications, signs, suggestions, evidence etc. involves skills beyond those of mere perception. Even when we see dark clouds looming and say "It looks like it's going to rain" we can only say this because we know from past experience that dark clouds often precede rain and we can use this knowledge to inform our judgements about the future.

Expertise is thus at least partly a condition of being acquainted with certain causal regularities. Experience and education about these regularities furnishes experts with exemplars that allow them to make more accurate predictions. But these predictions are not properties of the world. This is why even experts are often wrong, especially about long-term events.

So, to say that we see potential in a student is to hedge a bet based upon their previous achievements. It is certainly not a kind of mysterious emanation that only experts can sense. It is not a perceptible thing or energy of any kind. It is a supposition, based upon evidence and supported by experience without which the determination of potential would be impossible.

If beauty were a perceptible property of a prospective model then every perceiver - other life forms included - would have to be capable of perceiving it. I don't think anyone would be confident of that view. Beauty and meaning etc. are socially negotiated attributions. They are ideas we closely associate with certain kinds and configurations of perceptible attributes and objects.

When we give gifts we often try to conceal their identity by wrapping them. It is never the point of gift giving that the recipient should be able to predict the contents. Such a skill would render the ritual of wrapping meaningless. One of the great pleasures of wrapped gifts is the expectation they elicit, an expectation that is often most pronounced in childhood. This has two important consequences. Firstly, it encourages self control; a vital life-skill. Secondly, it encourages powers of imagination that are of inestimable value to us as a species.

-------------------

Thanks to Brian for his contributions to a previous discussion on the subject of talent that was an important prompt for this post. Major revisions, July 2021.

Tuesday 10 December 2013

Sense and Sensibilities (distinguishing and discriminating)



When discussing such things as taste, discrimination and discernment, there is always a risk of sounding elitist.  It's an occupational hazard in the arts. But, as I hope to show, this is due to two quite different ways in which we are capable of responding to experiences, one of which we all have access to, whilst the other has to be learned.

"But though there be naturally a wide difference in point of delicacy between one person and another, nothing tends further to increase and improve this talent, than practice in a particular art, and the frequent survey or contemplation of a particular species of beauty." —David Hume (1757)

It could be said that an art school education is largely directed towards the cultivation of sensibilities, to the sharpening and refinement of tastes and to the subtleties and nuances of method, media and process. In order for us to become conversant with and to develop our appreciation of the fine details and qualities of experience it is often necessary to attend closely to the minutiae of sensations and to form distinctions between subtly different materials, gestures, expressions etc. But does this skill emerge through sensory experience alone or do language, and the distinctions that language enables, play a vital part in what Hume called "the delicacy of taste"? He writes: "Where the organs are so fine, as to allow nothing to escape them; and at the same time so exact, as to perceive every ingredient in the composition: This we call delicacy of taste."

Hume observes that our first experiences of things are often "obscure and confused; and the mind is, in a great measure, incapable of pronouncing concerning their merits or defects." But as experience increases, "The organ [of taste] acquires greater perfection in its operations; and can pronounce, without danger or mistake, concerning the merits of every performance." But if it is the case that experience and exacting discrimination are sufficient to explain this skill then we are left with an unsettled question: do animals develop a delicacy of taste through experience also? And if experience and fine discriminatory capacities are fundamental to a delicacy of taste then old dogs and innumerable other mature animals must possess sensibilities that far exceed our own.

In the late 1980’s and early 90’s Timothy D. Wilson and colleagues published the results of several frequently cited studies which explored the relationships between feelingful or “affective” judgements and verbal criteria. In the most famous of these studies test-subjects were asked to rank a selection of 5 jams in order of quality. The average rankings determined by these novice jam tasters turned out to be fairly closely correlated with rankings determined by expert jam tasters. Another group of novice jam tasters were also asked to take the test but in this case they were asked to make a note of the reasons for their judgements before ranking each jam. In this case the average rankings were significantly unlike those of the expert tasters.

Wilson et al conclude that what differentiates novices from experts is that novices have not yet integrated their affective states into a conceptual system, with the consequence that their attempts to verbalise their feelings fail to do them justice. Experts, on the other hand, have a much more stable grasp of their conceptual system and the ways this describes and frames their underlying affective states.

What this research reveals is that experience alone is insufficient for the development of our sensibilities. No matter how many jams I try - and I like jam a lot - I'll never become an expert simply by triggering my affective states, by tasting my way. What makes the crucial difference is my capacity to structure and articulate my feelings through language. 

Savouring a sensation then, can be thought of in two quite different ways. We can prolong an experience by deliberately lingering over it - sustaining and consolidating the associated feelings. Or we can contemplate an experience by carefully distinguishing between its component parts. This would be impossible if not for the categories and concepts of language. The ability to discriminate, on the other hand, is enabled by our sensory capacities, capacities that we share, though obviously to differing degrees of resolution and acuity, with other animals. 

Sensibilities, it turns out, are what language enables us to derive from the affective responses of sensory discrimination - from our senses. Without language we might still be able to savour our sensations but we would be unable to contemplate them. Not only is contemplation linguistically derived, it is also inherently social, precisely because of these linguistic roots. Perhaps this is why much of the most memorable and pleasurable savouring is that which is shared.

"No pleasure has any savor for me without communication." —Michel de Montaigne

Sunday 24 November 2013

Signification and Sensibility: Notes on the Photographic Index



Is smoke a representation of the fire that produces it? Do shadows represent the things that cast them. Do the rings of a tree represent its age? Is the inverted view seen in a dewdrop a representation? Is a footprint? Or an echo?

For Charles Sanders Peirce (1839 – 1914) each of the above would constitute what he defined as “Indices.” For Peirce "An Index [singular] is a sign which refers to the Object that it denotes by virtue of being really affected by that Object.” Indices “represent their objects independently of any resemblance to them, only by virtue of real connections with them.”

This notion of a causal connection between the index and its object has often been thought to confer a special status upon indexical signs, and amongst such signs none has figured more prominently, or generated more discussion, than that of the photograph. Photographs are the result of electrochemical changes manifested in a photosensitive surface in response to exposure to light, and this causal relationship is believed to contribute to, or constitute even, the unique - and some would say “objective” – identity or “truth” of photographic images.

Perhaps most notable of all commentators upon the indexical nature of photography has been the French literary theorist and philosopher Roland Barthes whose 1964 Essay “The Rhetoric of the Image” proclaimed that the photograph is a “message without a code.” (other theorists argue something similar when they claim that photography is “transparent” to its subject) For Barthes this unique feature of photography is a consequence of its “absolutely analogical nature.” Barthes argued that photography should thus be differentiated from “drawing which, even when denoted, is a coded message.”

Since 1964 many theorists have followed in Barthes’ footsteps, variously arguing for the unique status of photography, its “truth claims”, its essential “objectivity”, “immediacy” or “reality.” But rather than explore the intricacies of these theories I’d like instead to make a point that may well have been obvious right from the outset.

The smoke rising from a chimney is not a representation and nor is a shadow, an echo or the miniature view spied in a dewdrop. It is true that these things have direct causes but the fact that these are implied, suggested, connoted, denoted, indicated, signified or otherwise made evident does not bring about a transformation of their subsequent effects into representations. An obvious or inevitable connection between an effect and its cause does not yield a representation. Representations are the products of intentional acts of substitution.

And what then of the unique status of photography? Certainly it can’t be anything to do with its absolute analogical nature. And if it is, we are simply left with the very obvious question raised by digital photography. It might then be argued that the analogical nature of photography, to which Barthes was referring, was in fact closer to what C. S. Peirce meant when he wrote:

“Photographs, especially instantaneous photographs, are very instructive, because we know that they are in certain respects exactly like the objects they represent. But this resemblance is due to the photographs having been produced under such circumstances that they were physically forced to correspond point by point to nature.”

Notice though that Peirce states that photographs “are in certain respects exactly like the objects they represent.” Did he mean to say that they are exactly the same as the objects they represent? Presumably not. To be exactly like is precisely not to be exactly the same. If Peirce had been more exacting he should perhaps have written: photographs are in certain respects and in certain circumstances perceptually indiscriminable from the things they represent. In other words, if the circumstances of presentation are very tightly controlled (in a lab experiment for example), we can find that we are unable to discriminate the difference between a photograph and whatever it represents in a variety of respects. It is nonetheless likely that we could easily distinguish between the representation and its object but it is important to point out here that conceptual distinguishability and perceptual discriminability are distinctly different. We are capable of distinguishing between things as a consequence of our ability to use language, whereas our capacity to discriminate perceptually is genetically acquired. Infants don’t have to learn to see photographic representations any more than they have to learn to see. Distinguishing, on the other hand, requires the development conceptual understanding.

It should be apparent then, that our genetically acquired ability to perceptually discriminate or, more precisely, to fail to discriminate, is the decisive factor that enables photographic representation. And it should also be noted that this includes all photorealistic depictions, despite Barthes’ logically unsupportable contention that all drawings are coded.

It turns out then, that Peirce’s theory of the index – at least as it pertains to photographs – is incapable of throwing much light upon the nature of  photography. It both lacks analytical precision and places too great an emphasis on the immediate causal history of photographs. As every digital photograph clearly attests, the causal history – even the immediate causal history - of the photographic image is often little more than a mystery to most of us. The same can be said of analogue photography in fact. This isn’t to say that the causal history of photographs is unimportant. Of course, some makers of photographic images deliberately choose to emphasise the indexical aspects of their craft; to draw attention to the processes of production, to the methods, qualities and materiality of the print and so on. In this sense they rely on, or seek to encourage, our abilities to make distinctions, to recognise, to compare and to distinguish. We might also say that these often subtle qualities draw upon our sensibilities, but this word demands a little care. If by sensibilities we mean some kind of genetically inherited ability to sense the finer details and qualities of things then we might have reason to be sceptical. Are our sensibilities genetically acquired? To some small degree perhaps they are, but for the most part our sensibilities are developed and refined through experience and most especially through considered attention, assisted significantly by our abilities to use language. Sensibilities are cultivated.

It might also be argued that photographs function by way of reference, signification or meaning. There is no doubt at all that photographs indicate, connote, denote, signify and allude to all kinds of things but they are by no means reliant upon these forms of signification because meaning, like sensibility, derives from our skills as language users. Although photography and other forms of non-verbal representation are enormously influenced, assisted and informed by language – by our abilities to distinguish - they are made possible by circumstances that must certainly have preceded our capacity to use language. They are derived moreover from a felicitous vulnerability in our perceptual capacities, a systematic perceptual weakness that we have discovered how to exploit to our profound and extraordinary advantage.



Sunday 17 November 2013

Representing the Mind



Researchers in the fields of cognitive neuroscience and Artificial Intelligence are fond describing the brain as a highly sophisticated computational device in which data is encoded and processed. Prior to the invention and development of computers many of the most prominent theories of the "workings" of the brain assumed either that it was minutely mechanical in its functioning (tiny cogs and pulleys, weights and levers) or else – or also - that the physical matter of the brain was simply some form of interface between the brain and the non-physical mind. This “Dualistic” theory, as it came to be known, is most closely associated with the 17th Century French philosopher Rene Descartes.

Since the 17th Century, advances in scientific understanding have enabled us to be a great deal more certain that the mind is entirely physical in origin, yet vestiges of Cartesian Dualism continue to linger in many quarters despite the overwhelming evidence that consciousness – the mind - is a product of electrochemical processes occurring in – or largely in - the brain.

Throughout history, many differing conceptions of the origins of thought and intelligent agency have stocked our vocabulary with a rich variety of metaphors borrowed from a range of objects, activities and occupations. We talk of the mechanisms of the mind, of streams of consciousness and information flows, of brain cells lighting up, sparking and firing, of cascades, filters, causal chains and neural triggers.

Such terms are extremely useful and often unavoidable in formulating theories and characterising barely understood processes. But characterisations can also present significant difficulties, especially if taken too literally as conceptual models. When attempting to understand complex and currently unexplained phenomena we are inevitably restricted both by what we know and what we don't know. Our knowledge and the terminology we have to describe it frames our thinking in ways that do not always help to illuminate the issues or processes we seek to understand. The same was true for Descartes, and is no doubt the reason he rejected the mechanical model of mind in favour of a non-physical account. It is perhaps possible then, to see how the versatility and character of the analogies available to us can significantly influence the concepts we are likely to formulate and the conclusions we are likely to draw during our exploration of unfamiliar processes and phenomena.

Since the 1970's Deirdre Gentner’s research has focussed on the use of analogical reasoning in the domains of cognitive science and artificial intelligence in particular. Gentner’s findings show that “analogical access and analogical inference are governed by very different rules.” In other words, when we make comparisons we have a tendency to pick out superficially accessible analogies instead of analogies that might offer greater predictive potential. I think this is precisely what is happening when the functioning of the brain is explained by way of computational analogies. Even more worrying though is the assumption – frequently made by theorists and scientists - that brains necessarily utilise encoded representations. I would argue that the failure to recognise the incoherence of this analogy is having a significantly detrimental influence on the development of both cognitive neuroscience and artificial intelligence.

In their 1995 book “Foundational Issues in Artificial Intelligence and Cognitive Science” Mark Bickhard and Loren Terveen argue that the fields of Cognitive Science and Artificial Intelligence are “in the midst of a programmatic impasse.” They make the vital point that this impasse cannot be overcome by making changes – no matter how insightful - at the “project level.” That is to say; if the foundations aren’t sound, no amount of work above ground is ever likely to resolve the initial errors - in fact it is far more likely to compound them.

Bickhard and Terveen coin the term “Encodingism” to describe the assumption that cognition involves the use of representational encodings. They write:

“It is assumed that the symbol represents a particular thing, and that it - the symbol - somehow informs the system of what that symbol is supposed to represent. This is a fatal assumption in spite of its seeming obviousness.”

The point they are making is that it is by no means straightforward what any symbol is actually intended to symbolise. As C. S. Peirce theorised at the end of the 19th Century, symbols represent their denoted objects as a result of rules or interpretive habits that are independent of any properties that may or may not be shared between symbols and the objects they represent. Symbols are things that require interpretation. They do not lead immediately and without complication to the things they denote. The rules and habits of symbol use have to be learned. If this is true, then it is extremely doubtful that an intelligence could emerge capable of interpreting symbols without a prior non-symbol-using evolution. And if intelligence is possible without symbol use, then no encoded symbols are required for cognition.

Friday 8 November 2013

Knowing Absence

Towards an alternative theory of the role of representation in perception.



“You don't merely think that the tomato has a back, or judge or infer that it is there. You have a sense, a visual sense, of its presence. In what does the visual sense of the presence of the hidden parts of the thing consist, if it does not consist in the fact that we see them? This is the problem of perceptual presence—or better: the problem of presence in absence. The object shows up for visual consciousness precisely as unseen. […] The thing that needs explaining is not that we mistakenly take ourselves to see something that we don't see. The puzzle is that we take ourselves to have a sense of the presence in perception of something that is manifestly out of view." —Alva Noë

Here contemporary philosopher of perception, Alva Noë, outlines the the issue of what he terms "presence in absence." Whilst I do not agree with all of Noë’s theories, his avoidance of any reliance upon cognitive representational states is very much to his credit. Nonetheless, in rejecting these unsubstantiated phantoms, these spectral darlings of much scientific and philosophical presupposition, he pays little regard to the role of other forms of representation (the only kind for which proof exists) that are so vital for an explanation of our ability to conceive of absence and, arguably, of perception itself.

Noë is right, I think, to identify a significant philosophical issue with the question of absence, of our capacity to account for the parts of the things that are currently out of view. For him the answer to this paradox lies in our "sensorimotor understanding” and “conceptual knowledge" of the things that we see that "brings the world into focus for perceptual consciousness" and that "discloses the world to us." He writes:

"To perceive something, you must understand it, and to understand it you must, in a way, already know it, you must have already made its acquaintance.
There are no novel experiences. The conditions of novelty are, in effect, the conditions of invisibility. To experience something, you must comprehend it by the familiarizing work of the understanding. You must master it. Domesticated it. Know it."

But how should we interpret this “understanding”, this “perceiving”, this "knowing"? If I reach for what I know is a fresh tomato only to find that it is fake then quite evidently there was something I didn’t know that I now know. Moreover, if the new knowledge contradicts the assumed knowledge – the preconception - then I think it would be right to say that I have encountered an unexpected, if not novel, experience. If someone pops their head around my door to say hello, I know that I’m not looking at a disembodied head. But I’m not convinced that the best explanation for what I perceive is provided by reference to my sensorimotor understanding or even to my conceptual knowledge. What the explanation requires is my capacity to answer the question – should it ever arise – “To what is the head attached?”

For an organism to be surprised - for novelty even - there has to be a capacity for expectation. Without an expectation of presence it is inconceivable that an absence could be registered. A vixen doesn't register the absence of an inexistent fox cub any more than we might detect the absence of the future. Or, to paraphrase Noë: to expect something, you must have already made its acquaintance. But, as we have already seen, to be acquainted with something, does not necessitate knowing it in any substantive sense. To be acquainted with something may be to know nothing more than what to expect when becoming reacquainted with it – to be aware of what features, attributes or properties one would miss if they were absent. In order for this to be possible we need not assume representational mental states nor require sensorimotor understanding but merely a capacity for registering a disagreement between the current perception and whatever residual cognitive capacities (memories) remain of the thing as previously encountered.

But to conclude the explanation here would be to neglect the most important part of the story. What needs explaining is the nature of this capacity to register absence. Advocates of representational states will simply assert that we possess cognitive representations that allow us to perceive absence. But as Noë rightly argues, an absence is not something that can figure as a mental state, let alone a representational mental state because "What they [neurons] can’t do is fire in such a way as to signal that a hidden feature is present”

The reason that our capacity to produce representations is so vital to an understanding of absence is because perception is conditional upon our capacity to represent things. For sophisticated representation users like ourselves this capacity most often manifests itself as one or other of a vast repertoire of representational actions: an exclamation, a description, a raised eyebrow, a shrug, a song, a dance, a movie, a poem. But it need not be the case that we actually produce a representation. We need only be capable of doing so.

So, when Noë says that we have a “visual sense” of the hidden parts of the tomato he is not implying – although he might be mistaken as implying - that we literally have a sensory capacity to detect absence. Not even the Large Hadron Collider at CERN has sensors capable of detecting absence. Absence is undetectable. What we “sense”, or better still, what we are capable of doing when we look at the world is representing the things that we see. Furthermore, we know how to offer representational best guesses about the things that we can’t see. But this knowledge isn’t so much a knowledge of the fullness of things - it is a knowledge of how to represent them in whatever partial or limited form they are encountered. Very often the accuracy of our representational capacities is borne out by further investigation. Sometimes not. Sometimes the back of the tomato turns out to have a bite out of it.

Wednesday 30 October 2013

Pretending



“Pretending to growl like a bear, or lie still like a corpse, is a sophisticated performance, where the bear’s growling and the corpse’s immobility are naïve.” –Gilbert Ryle

When he was around the age of 2 my son loved playing games of roaring with anyone willing to indulge him. Sometimes, if my performances were too convincing, he would become genuinely fearful - once or twice ending up in tears. It might be assumed that his fascination with roaring was simply the result of a desire to conquer his fear of facing up to a threatening adversary. I’m sure this is partially true but perhaps it’s not the full story.

In 1958 two British philosophers, Elizabeth Anscombe and J. L. Austin, were involved in a debate over the nature of pretending. Simply put, the central problem went like this: how can we tell the difference between pretend anger and anger proper? There is a lot of detail to the arguments presented which I cannot hope to distil here but one particularly interesting aspect that neither philosopher fully addresses concerns the representational character of anger.

To make a display of anger, as many animals do when threatened, is to produce a representation of a violent intention; it is to make a symbolic representation that other creatures, even other species, are easily able to recognise and as such it has enormous evolutionary efficacy.

To make a convincing pretense of anger is to match all of the physical features of anger in what would be best described as a matching representation. When well executed, matching representations are identical to the things they seek to represent. Indeed, get a young child to feign anger and you’re likely to get a real bite, scratch or thump for your trouble – though it might be argued – and I’m not too keen on testing the theory – that the bite etc. is not as determinedly forceful as it might otherwise be. I suspect this is why my son, and so many other boys (being the more violent members of the species) of a similar age are so fascinated with displays of mock anger because it is vital for them to be able to distinguish a mere representation of anger from the far more serious representational prospect of immanent violence, i.e. genuine anger.

As inordinate representation users we humans are extremely sensitive to and adept at employing contextual cues in ways that other animals simply aren’t and we have developed highly sophisticated and sometimes extremely subtle ways of alerting one another to the representational status of our behaviour. The next time you encounter a dog, try making a matching representation of anger towards it and I’m sure you’ll see what I mean.


Wednesday 23 October 2013

Imagining Itself (Part XX: Conclusion)



When I first set out to explore the subject of human imagination, more than 18 months ago, I hadn’t the faintest idea that I would be led on the journey that has culminated in these last 20 entries. I had expected to produce a total of perhaps 5 or 6 entries covering the limits of imagination, its possible evolution and some thoughts about its relation to learning and education. Instead I have encountered theories, research and discussions that have radically altered my understanding, not just of imagination but of the nature of consciousness and most especially of the indispensible role that our capacity to produce representations plays in perception, imagination and conscious action.

Complex fields of enquiry are often a source of contrasting opinions and interpretations and this is undoubtedly the case where imagination is concerned. Disagreements of this kind can be an important catalyst for closer scrutiny of the issues and more careful assessments of the available evidence. However, there are instances where the weight of opinion on a subject becomes so overwhelming that dissenting voices are simply drowned out of the conversation. In such circumstances the importance of discourse is easily overlooked and even the most questionable ideas can sometimes appear to be unassailable truths. One such theoretical presupposition that continues to command a significant amount of credence amongst scientists and philosophers is the notion that mental states necessarily involve the utilisation of representations of one kind or another, whether they be mental images, computational data structures or mental content. Nonetheless, a handful of scientists and philosophers interpret the lack of evidence for such representational states as grounds for serious doubt regarding what is sometimes called “representationalism”. As yet, no scientific study has been successful in identifying the coordinates, structure or encoding of any form of neural representation, yet for many influential researchers and thinkers the supposition is all but proven.
"The most successful tyranny is not the one that uses force to assure uniformity, but the one that removes awareness of other possibilities, that makes it seem inconceivable that other ways are viable, that removes the sense that there is an outside." —Allan Bloom
The desire to press ahead in any form of enquiry is always tantalizing, but the temptation to do so in the absence of sufficient evidence should probably be resisted, especially where the theoretical foundations are as uncertain as they clearly are regarding mental representations. Where there is no need for doubt though regarding the available evidence is in our genetically acquired and culturally evolved capacities to produce publicly available representations. As I have tried to clarify at various points throughout this series of essays, as well as in associated comments, brains have evolved as essential components in complex responsive organisms and it is these processes of response, and the dispositions that underlie them, to which we should be looking for insights into the workings of cognition, not to some incompletely theorised and scientifically questionable assumptions about representational states.

With the help of various theories I have tried to expose some of the obvious flaws in representationalism and to show that important alternatives exist that provide significant explanatory potential and scope. I have also tried to show that these alternatives offer important conceptual tools for resolving problems that have stymied thinkers throughout history. No doubt some will find these claims to be overly assertive or immodest on my part. Perhaps this is the price one pays for taking an unorthodox stand on any issue. But what I hope the reader will appreciate above all else is that the subject has been considered carefully and that genuine shortcomings have been identified within the mainstream theorisation of imagination and mental states.

Through my research I have come to the conclusion that a comprehensive understanding of genetically acquired and culturally evolved representational practices is fundamental to the furtherance of our understanding of imagination, perception and consciousness. Without the ability to recognise, create or use representations it is doubtful whether consciousness of any form would be possible and certainly there would be nothing to say or even to wonder about regarding the nature of imagination.



Sunday 13 October 2013

Imagining Itself (Part IXX: Learning Skills)



If it were the case that imagination could initiate neural growth (learning) to the same degree as experience then the question arises: how could we ever distinguish between cognitive development drawn from experience and that drawn from entertaining but infeasible flights of fancy? If there were no difference between the ways we learn from experience and the ways we learn from imagination then how could our cognitive architecture ensure that imagination didn’t simply fill our minds with a deluge of impractical nonsense? To set too much store by the inventions and indulgences of the imagination would surely risk laying down all kinds of meaningless neural networks and worse still, this would probably happen without the slightest conscious awareness. But how is this not the case and how do our minds avoid such self-deception?

It might be suggested that there must be some kind of limiting mechanism or filter at work that inhibits neural development on the basis of imaginative thought, but this need not be the case. There need be no such mechanism because - as Sartre, Ryle and others have surmised - imagination is the exercise of already formed knowledge – of prior memories, ideas and beliefs. Just as perceptual experience leaves its most indelible imprint when it is most repeated, most profound or most surprising, so too, we might suppose, does  imagination. But unlike perception, which frequently presents us with the surprising and the unpredictable, our thoughts and imaginings are rarely truly unexpected. If imagination consists of the interplay of what we already know, then the only means by which we might encounter the unexpected – and from which we might learn – is through the previously untested recombination of this knowledge and the careful pursuit and resolution of incompatible ideas and beliefs. As it happens, we are not especially skilled at this - as research into Cognitive Dissonance clearly shows.

Skills are acquired either by genetic inheritance or by active engagement with the world and never simply by imagining. So for example, it is not possible to become capable of drawing something by the agency of imagination alone because a capability is not something that imagination can ever bestow or confer. Imagination is the means by which we sharpen our skills of representation but it is not the means by which we acquire such skills – or skills of any description. Indeed, imagination would be better understood as being nothing other than the means by which we contemplate, anticipate and refine our skills of representation.

It is because representation is so fundamental to learning that imagination plays such a vital role in the improvement of skills. To acquire a skill is to be able to demonstrate it - it is to be capable of representing it to others. And these capabilities of demonstration and mimicry - of teaching and learning - have been of inestimable use to us as a species. It might be argued that to carry out a learned action is not necessarily to represent it. This is true. However, to carry out an action in the presence of onlookers is for it to be available in representational terms; as the means by which an action is ‘performed’ and as the procedure by which something might be accomplished.

To watch an action is to become capable of representing it. But, crucially, it is not necessarily to be able to represent it either faithfully or fully. Why else would we ever ask: "Can I have a go?" if the capacity to represent something (to have committed it to mind so to speak) was sufficient in and of itself?

Despite the fact that imagination cannot enable skills it can, and frequently does, allow us to improve them. It has long been known that the use of visualisation techniques in sports training can bring about measurable improvements in performance. To imagine how you could better swing a tennis racket is to imagine how you would represent the motion as a performative act. It is to deliberately engage many of the same cognitive structures as would be exercised if you were to actually perform the action. And, as many studies have shown, repeatedly exercise of these cognitive structures through visualisation causes them to become increasingly consolidated and increasingly stable.

In 2011 the results of a study were published evaluating the benefits of a variety of visualisation techniques for the rehabilitation of stroke patients. The study found “No evidence of the benefit of mental practice with motor imagery in stroke.” In other words, no matter how hard or regularly the stroke patients tried to imagine improvement in their impaired motor skills they were unable to bring about any measureable improvement. On the other hand, when visualisation techniques were combined with physical therapy, then measurable improvements did occur and visualisation helped significantly. This research lends strong support to the idea that imagination is not an enabler of capacities but is instead an important means for reinforcing and focussing skills that have already been acquired.

Imagination is an enormously important capacity – not just for capabilities that we think of as being mental, like thinking for instance - but also for our abilities to act deliberately and skilfully in the world. In fact it might be said that imagination is the very precondition of our being able to act in such skilful ways.