Neurological Evidence On Concepts

Throughout this chapter we have concentrated on attempts to understand the nature of the "normal" knowledge organisation. However, in the last 20 years a parallel research stream has examined impairments in knowledge that arise after neurological damage. This research has revealed a number of interesting and important findings.

First, people with a variety of neurological damage develop specific impairments of their semantic memory. When the cognitive systems involved in reading and speaking remain intact, there is evidence that the storage of knowledge or access to it, or both, can be disrupted. For example, Schwartz, Marin, and Saffran (1979; Schwartz, Saffran, & Marin 1980) studied a patient, WLP, suffering from a severe dementing disease. WLP's ability to read was intact but her comprehension was poor. For example, when she was asked to indicate which one of a set of words a picture represented (using basic-level words, like "spoon", "apple", "cigarette"), she was poor at selecting the correct word for the picture. Furthermore, when she chose the wrong word, she tended to choose one that was related semantically to the correct choice. So, for example, for a picture of a fork she chose the word "spoon" and for a picture of a brush she chose "comb".

A second noteworthy finding is the way in which knowledge about superordinate concepts seems to be less susceptible to damage than more subordinate information (Warrington, 1975). Warrington (1975; see also Coughlan & Warrington, 1978) studied a patient, EM, also with a dementing illness, using a forced-choice decision task; that is, the patient was given a question like "Is cabbage an animal, a plant, or an inanimate object?" or "Is cabbage green, brown, or grey?" and had to choose one of the alternatives. It was discovered that EM was only wrong in 2% of cases on the former type of question but was wrong 28% of the time on the latter type of question. The point being that more subordinate attribute information about the cabbage concept was lost, even though the superordinate classification of a cabbage as a plant was retained (although see Rapp & Caramazza, 1989, for a challenge to this finding). Similar evidence has been found by Martin and Fedio (1983) in the naming errors made by Alzheimer's patients. These patients tend to give superordinates when they name objects wrongly. So, for example, asparagus is named as a vegetable and a pelican as a bird.

A third, and perhaps most surprising, finding from the neurological literature is evidence that patients have deficits in their knowledge of specific categories of objects. For example, Dennis (1976) has reported a patient who had difficulties only with the category "body parts". Warrington and Shallice (1984) have studied patients with similar deficits following damage to the medial temporal lobes, arising from herpes simplex encephalitis. These patients were very good at identifying inanimate objects by either verbal description or picture but were considerably poorer on objects that were living things or foods. More specifically, Hart, Berndt, and Caramazza (1985) have reported a patient, MD, with a deficit specific to the naming and categorisation of fruits and vegetables. Even though lesioned connectionist models manifest these sorts of deficits (see Plaut & Shallice, 1993), some researchers have argued that they may be artefactual (see Funnell & Sheridan, 1992; Parkin & Steward, 1993; Steward, Parkin, & Hunkin, 1992). Funnell and Sheridan (1993) proposed that studies showing category-specific effects did not control for important variables like the familiarity of the objects and their name frequency. In a patient they examined, they found that when they controlled for such factors there was no evidence of category-specific deficits. At this point, although the phenomenon is not at issue, the interpretation of it is the subject of considerable debate (Caramazza, 1998; Forde & Humphreys, 1999).

Many of these effects can be found together in Alzheimer's disease, where there is considerable degradation of patients' abilities in semantic memory tasks (see Nebes, 1989). How are we to understand these findings in the light of the previous theories of concepts? Shallice (1988) suggests that the salience of superordinate information indicates that the Roschian basic level is less important than previously thought. In terms of specific psychological models he suggests that these results favour later network models (e.g., Collins & Loftus, 1975) and distributed memory schemes (McClelland & Rumelhart, 1985). In distributed memory models, patterns of activation encoding superordinate information are less disrupted than patterns of activation representing exemplars.

All About Alzheimers

All About Alzheimers

The comprehensive new ebook All About Alzheimers puts everything into perspective. Youll gain insight and awareness into the disease. Learn how to maintain the patients emotional health. Discover tactics you can use to deal with constant life changes. Find out how counselors can help, and when they should intervene. Learn safety precautions that can protect you, your family and your loved one. All About Alzheimers will truly empower you.

Get My Free Ebook

Post a comment