Riding a bicycle may or may not involve the same areas of the brain in different people. We do know, though, that aspects of our ability to spell (and specifically, to spell regularly and irregularly spelled words) are encoded in the same areas. But writing has not been around long enough for evolution to have encoded in our DNA anything at all that could be specific to spelling, let alone the distinction between regular and irregular spellings. And anyway, not all languages have irregular spellings, and not all languages are alphabetic (see Chapter 11). So the relevant neuroanatomy cannot have been laid down with spelling in mind.
The knowledge we have about something-its meaning-will necessarily reside across many different parts of the brain, so as to include visual information, information about the sounds associated with that thing (or the word for that thing), information about its feel, its smell or taste, its function, and other aspects of the contexts in which it occurs. Some things may require stronger associations with one part than with others-the visual areas of the brain, for instance, or the motor areas, which mediate movements and manipulation. Not surprisingly, therefore, the deficits that arise following damage to these different parts can be very varied indeed.
It is surprisingly difficult to identify the form these deficits take. One problem is that there are two quite distinct ways in which these deficits can be described. A classification based on symptoms alone would hide the fact that any one symptomatology could arise from a number of causes. If a patient was unable to say the word `tiger' when shown one, they might have lost the ability to recognize tigers. Or perhaps they could recognize tigers and knew the word `tiger', but were simply unable to say it. Why they might be unable to say it is another matter. So going by gross symptoms alone is not enough. An alternative would be to classify the deficits according to which components of the language faculty have been impaired. In effect, that is what we just did when distinguishing between the two `tiger' deficits. But identifying which components have been impaired is rarely easy. Patients who have suffered some form of brain damage rarely exhibit a deficit that is restricted to just one mental process. Often, different processes are affected to differing degrees, making it hard to disentangle which processes have been affected, and to what extent. Most patients with a deficit in production will have some kind of deficit in comprehension and vice versa. Many patients who have a deficit in their spoken language will also have a deficit in their written language. Do these patients have a single underlying deficit, or do they suffer from multiple impairments that have co-occurred only coincidentally? Often, it is hard to know.
In the descriptions that follow, some of the main kinds of deficit will be described, in terms of both their symptoms and the components that appear to be affected. Deficits which affect spoken language are called aphasias. Deficits affecting the ability to understand written language are called dyslexias. These two kinds of deficit will be described separately, although as just mentioned, they rarely occur one without the other. And any one patient described as having one particular deficit probably has some other deficit too. But first, a final word on what `a deficit' means. It is rare for a deficit to completely abolish a particular ability. A patient who is described as being able to read aloud written words whilst being unable to understand them may be able to understand some words, but many fewer than he or she could have been expected to before the event that caused the damage. Generally, the published descriptions of these patients do include information about the severity of the deficit.
Failures of understanding
One of the rarest deficits that manifests itself as a failure of understanding is certainly one of the oddest. Patients suffering from pure word deafness are unable to understand words spoken to them. But they can read, write, and speak normally. Some patients report that what they hear sounds as if it is in some unrecognizable language. Some (but not all) of these patients can repeat or write down the words they hear even though they do not understand them. In one such case, the patient would write down what she heard, and only by reading what she had written could she understand what had been said to her. In another case, the patient could, despite not understanding the words spoken to him, guess which were real and which were made up. Because the real and made up words differed by just one phoneme, the `deafness' could not have been due to any failure in recognizing speech sounds (such failures do sometimes occur). These cases suggest that at least part of the neural circuitry that was activated on hearing a word was intact, but that the connections to the circuitry that encoded the meaning had been lost. The meaning was still there, though, otherwise these patients would not have been able to read, write, or speak.
An equivalent deficit has been found in reading-one patient was unable to understand many of the written words shown to her, but she could read them aloud perfectly. When she did, she would only then understand the word she had just seen. It did not matter whether the word was regularly or irregularly spelled-her pronunciation would be perfect. When not reading aloud, she could tell which written words were real and which were made up, even though she understood neither. This is another case where the associations from the form of a word (in this case, its visual form) to the circuitry encoding its meaning have been lost.
So being deaf to the meaning of a spoken word, or blind to the meaning of a written word, does not mean that the meaning itself has been lost. These patients have no problem, when wanting to express that meaning, in finding the right words with which to do so.
Failing to find the right words
These patients with word deafness or blindness could not make the link between the name of an object and the object itself (that is, its meaning). This inability was quite specific to either seeing, or hearing, that name. An equivalent deficit, in production, occurs when a patient cannot make the link between an object they are looking at, or perhaps thinking about, and the object's name. Patients suffering from this naming deficit will recognize an object if it is shown to them, and will know that they are being shown a picture of a tiger, for instance, but they will be incapable of saying `tiger'. Instead, they will either say nothing, or will provide a description rather than the word itself, or will come out with the wrong word, which, depending on the deficit, might be related in meaning-for example, `lion' for `tiger'. Some of these patients have absolutely no problem saying a word when it is written down for them, or when it is defined for them-they just cannot say it when shown the object to which it refers. One patient could even write the names of the objects that she could not say. How might these naming deficits come about?
When we look at something, the information we receive via our eyes is just a kind of sensory trigger that sets off some neural circuitry associated with the meaning of whatever we are looking at (this was covered in more detail in Chapter 9). A written word is also a trigger, as is a spoken word, or a definition. But each of these triggers is likely to stimulate a different neural circuitry-a definition will evoke a very different kind of experience, for instance, than whatever is evoked when the physical form of the object is experienced. The taste of an egg, for example, evokes a very different kind of experience from the sight of an egg, and yet each is clearly `egg', as is the description `it is laid by hens'. In effect, these triggers evoke different aspects of the meaning of the same thing. They evoke different meaning circuits. And each of these circuits needs to be connected to the circuitry responsible for saying the appropriate word. These connections constitute the association between the different aspects of meaning and speaking.
Naming deficits result when some of the connections have been lost between the neural circuitry that is stimulated by seeing an object (the aspect of meaning concerned with what the thing looks like) and the neural circuitry associated with saying the name of that object. Other connections might still be intact. So if patients experience an object in a different way, perhaps by feeling it, or hearing a description of it, they may be able to name it if the neural circuitry that has been stimulated by th
at experience is still connected to the circuitry associated with saying the appropriate name. And because only some connections are affected, these patients may be able to name some things, but not others.
The fact that only some of the connections may be affected explains why deficits are rarely all or none-much depends on how extensive the damage has been. It might also explain the surprising finding that some patients are worst at naming certain kinds of thing-living things, man-made things, or fruit and vegetables, for instance. In all likelihood, these selective deficits arise because the meaning of an object (what it looks like, what it is for, how it should be used, and so on) requires many different associations across many different parts of the brain-parts that encode physical form, or function, or movement and manipulation. Certain kinds of thing may rely for their meaning more on one part than another (man-made objects, for example, tend to have function, whereas animals do not). So we store everything we know about an object across different parts of the brain, with each part encoding some different aspect of that object. If connections to one of these parts are lost, whichever aspect of an object is stored in that part will not be able to trigger a naming response. For example, the function of something (a potato-peeler, perhaps) may not be enough, now, to trigger the appropriate naming response. The result may then be a deficit that affects the patient's ability to name certain kinds of object, or even, depending on which connections have been damaged, to recognize those objects.
The deficits described so far have all been quite specific, affecting the connections either between the form of a word and its meaning, or between the meaning of a word and its pronunciation. But the meanings themselves have been spared. Imagine the frustration of knowing what you want to talk about, but being unable to find the words with which to do it. But these are the rarer cases-it is more common for patients to exhibit a range of fairly general deficits, some of which are due to damage to the connections to and from meaning, and some of which are due to damage to the neural circuitry that encodes meaning itself.
Beyond individual words I
Whatever the cause of the deficit, it is inevitable that patients who have problems understanding single words will also have problems understanding sequences of words. Similarly, patients who have problems saying individual words will inevitably have problems producing sequences of them. However, there are deficits in sentence production and comprehension which go beyond what could be expected on the basis of just single-word deficits.
Traditionally, these deficits have been divided into those that affect production, and those that affect comprehension. However, it is quite rare to find a deficit in one and not in the other. Broadly speaking (there is no other way when it comes to the complexities of acquired language disorders!), many patients with disordered spoken language (aphasia) produce slow hesitant speech. Some have particular problems with grammatical function words ('the', `by', `was', and so on) and inflections ('-ing', `-s', `-ed', and so on), tending to miss them out completely. Here is one example:
Two ... ah, doctors ... and ah ... thirty minutes ... and yes ... ah . . . hospital. And, er Wednesday . . . nine o'clock . . . and er Thursday, ten o'clock ... doctors. Two doctors ... ana ah ... teeth. Yeah ... fine.
More often than not, these patients also have a comprehension deficit. One of the diagnostic features of this comprehension deficit is particularly interesting. Whereas these patients would understand sentences such as `The dog bit the boy' or `The boy was bitten by the dog', they would have problems understanding `The dog was bitten by the boy'. It seems that they tend to interpret such sentences according to what would be a plausible scenario. Dogs tend to bite boys, not the other way around. Basically, it looks as if these patients' brain injury has severely limited their ability to use grammar.
Things are not quite this simple, because a few of the patients who are severely impaired in their ability to use grammar in production have no corresponding comprehension deficit. Similarly, some patients with this particular comprehension deficit do not have the corresponding production deficit. In short, it looks as if the production and comprehension of grammatical structure can be affected independently.
Reconciling the findings with earlier theory
This last claim is controversial and, if true, it would compromise the account in Chapter 10 of how grammatical conventions are applied during the understanding and production of language. There, it was suggested that the associations that form when learning to understand language serve the production of that language also. In one case, sequences of certain kinds (including words and their inflections) would evoke meanings of certain kinds. In the other, meanings of certain kinds would evoke sequences of certain kinds. But if the same set of associations could both decode grammatical information and, when run in the reverse direction, encode it, we should find that patients who have a decoding deficit also have an encoding one. But we have just seen that this is not always the case.
This is an example of how research into what can go wrong with language can feed into theories of how language works normally. We now need to modify the theory to take into account the fact that different associative links are used during the encoding and decoding processes. Somehow, these two processes, which are basically just mirror-images of one another, would have to become embodied, during learning, in two distinct sets of associative links. But how?
Imagine that each time we hear a sentence, we simultaneously and unconsciously produce it. Associations will develop between the input sequences and their corresponding meanings. But if the meanings must become associated with the covertly produced output, then these associations will necessarily become a kind of mirror image of those between input and meaning. In effect, going from input to meaning will require one kind of process embodied in one set of associations, whilst going from meaning to input-as-covert-output will require the embodiment of the reverse process but on another set of connections. Technically, the two sets of associations will not be exact mirror-images because the neural activity in response to the input will spread through the neural circuitry, and in doing so will change. This means that the neural activity that has to be associated with the input-as-covert-output will not be quite the same as the neural activity which that input evoked to begin with. But, despite this, there is a sense in which the two sets of associations are complimentary.
This description is very speculative, but in the next chapter we shall come across networks of artificial neurons which do the equivalent. They associate input sequences with patterns of (artificial) neural activity corresponding to meaning, and associate those meaninglike patterns with output sequences that are just copies, more or less, of the input.
Beyond individual words II
A second broad category of deficit that goes beyond individual words is marked by very fluent and grammatical speech, complete with grammatical function words and inflections, but often containing completely inappropriate content words (the words that convey the main meaning, such as the nouns, verbs, and adjectives). Sometimes, patients with this deficit produce nonsense words instead of real words, a little like Lewis Carroll's Jabberwocky. In one case, for instance, a patient was asked to describe a scene which included a bull chasing a boy-scout past a table on which was a saucepan and a knife, and next to which was a stool. The description of this scene included the following:
A buk is cherching a boy ... ged stroe a table with orstrum and a three-leg
These patients are often severely impaired in their understanding of spoken sentences, and so are unaware of what they themselves have uttered. The deficit suffered by these patients is the converse of the deficit described earlier in which patients could retrieve content words, but could not organize them into grammatical sentences. Here, patients cannot retrieve the appropriate content words, but can organize what they erroneously retrieve into grammatical sequences.
It looks, then, as if the application of grammatical conventions and the selection of the words to which
those conventions should apply are functionally separable (see Chapter 10). Unfortunately, this state of affairs is muddied somewhat by the fact that the symptomatology of any individual patient is not always as straightforward as the descriptions given here would predict. As soon as one stops making gross generalizations, and starts looking at individual cases, things become much less clear-cut.
This last complication is true of all acquired language disorders. Which is why it is not easy to categorize the deficits into anything but the broadest of categories, each of which represents a range of deficits and a range of severities. This is true irrespective of whether we are dealing with impaired spoken language, or impaired written language, to which we turn next.
Deficits in reading: acquiring dyslexia
The deficits described so far have been aphasias-predominantly affecting spoken language. Dyslexias affect the ability to comprehend written language. To distinguish the kind of dyslexia due to brain damage from the kind that has its origins in childhood, the first kind is often referred to as an acquired dyslexia, and the second as a developmental dyslexia.
The Ascent of Babel: An Exploration of Language, Mind, and Understanding Page 23