Welcome to this online textbook chapter on language. The emphasis will be on a cognitive neuropsychological approach, with the data from brain-damaged individuals providing insight into how the unimpaired brain processes language. To set the scene, we will introduce you to a patient who has suffered a stroke affecting various aspects of his language ability. We will show you some of the things which are now difficult for him, and how surface appearances do not reveal those difficulties. Listen to his conversation with Dr Randi Martin, first of all. Notice how he seems to understand the questions put to him, with no trouble. Notice also how his speech production is slow, with his taking much more time than most would to express himself, even though the topic is something he has talked about many times in the past.

One of the things you may wonder about is whether he has some speech difficulty which prevented him from speaking more fluently than that. We need to look at some other task where he has to speak but does not have to construct sentences in order to rule out that possibility. One such task is repeating words. Here is another clip, showing ML's single-word repetition.

You can see that he is able to speak promptly, without hesitation, when he repeats the words. So, it is not the case that his conversation is so hesitant in conversation due to a pronunciation difficulty.

We hope you noticed how he understood the questions put to him by Dr Martin. His comprehension of sentences is not as good as it would have been before his stroke, however. To give you an idea of this deficit, take a look at the following graph, showing his performance on a comprehension task compared to that of a group of eleven controls who were round about his age and education level. There were five different sentence types and he was appreciably worse than the controls in accuracy on all five types. The pattern in his errors is qualitatively similar to the controls, however, indicating that the difficulty of the structures (from left to right, difficulty increases) had an impact on him, just as it did on the few errors the controls made.

Now watch ML performing this language task (note that he uses his index finger for YES and ring finger for NO in his judgments of plausibility). When he is wrong, you will hear a buzzer (the items are from the practice set). When he is right (on a test item), the time he took to respond will appear on the screen. Notice how slow the time on the fifth one is, which is hard because of its complex grammatical structure.

A stroke generally affects a number of different aspects of language. Just to show you more ways in which ML was affected by his stroke (on the left side of his brain, toward the front), listen to him now reading words and non-words aloud. Non-words are made up from real words by substituting one or two letters to make a pronounceable non-word which does not occur in the English language. From being able to read real words fairly well, he now shows a lot of difficulty coping with these made-up words. You will see the word written on the screen and so will be able to judge for yourself whether you could have found a pronunciation for each one.

This difficulty reading non-words while being able to read most words is curious until you understand the processes by which most of us read. What has happened with ML is that he can no longer make use of the rules he learned as a child about how to pronounce letters and letter sequences. He is, however, able to refer to his memory for the overall pronunciations of words that he has seen before. Because he never saw these non-words previously, he has no stored pronunciation of them and, without the ability to sound out the letters, struggles to pronounce them, often relying on a word he does know that is similar in visual appearance.

Another curious deficit ML has is with naming pictures of items from the category of body parts. Notice how he copes better with naming pictures of other categories, such as fruits, vehicles, furniture, kitchen implements, yet struggles to find the names of some body parts.

This curious island of difficulty is not unique to ML, though we still do not understand what brings it about. It is an illustration, however, of how it is necessary to zero in on quite particular aspects of the skills being tested. It is not good enough to have just a mix of pictures to name; they need to be selected from different categories known to the patient, across the categories the pictures need to be of similar familiarity and degree of use in the language, and they need to be of similar visual appearance, with no individual category disadvantaged by being more complex to illustrate than any other, or more uniform amongst its members than the others. Because ML could read real words well, and could name the other categories well, it is obviously not the case that he has any trouble with perceiving or recognizing pictured objects.

This little exploration of ML's deficits has given you, we hope, a bit of an idea of how the cognitive neuropsychology approach to testing is carried out. There will be many opportunities in this chapter to participate in tasks yourself and to get a score similar to what is obtained from patients in researching their difficulties. This research helps in diagnosis and helps us understand how the intact brain probably functions.

To conclude, listen to one more interview with ML in which he describes what it is like to have a stroke and some of the things he has to cope with in his daily life.