Discovering the Social Mind

Selected works of Christopher D. Frith

Christopher D. Frith

• Publisher: Psychology Press
• Year: 2016
• ISBN:
  9781317247227
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

In the World Library of Psychologists series, international experts themselves present career-long collections of what they judge to be their finest pieces - extracts from books, key articles, salient research findings, and their major practical theoretical contributions.

Christopher D. Frith has an international reputation as an eminent scholar and pioneer in the fields of schizophrenia, consciousness, and social cognition. A specially written introduction gives an overview of his career and contextualises the selection in relation to changes in the field during this time.

This collection reflects the various directions of Frith's work, which has become increasingly philosophically oriented throughout his career, and enables the reader to trace major developments in these areas over the last forty years. Frith has had his work nominated for the Royal Society Science Book Award and, in 2009, was awarded the Fyssen Foundation Prize for his work on neuropsychology. He has also been awarded several prestigious prizes for his collaborative work with Uta Frith.

This book is an essential read for those students and researchers engaged in the fields of social cognition, cognitive psychology and consciousness studies.

Section 1
Schizophrenia

1
Consciousness, information processing and schizophrenia

C. D. Frith

Introduction

The three principal positive symptoms of schizophrenia, hallucinations, delusions and thought disorder, are all disorders that manifest themselves in the consciousness of the patient. Indeed, two of these symptoms, delusions and hallucinations, can only be assessed on the basis of the patient’s introspections concerning his conscious experience. The third symptom, thought disorder, is observed in the patient’s speech; however, it is inferred that this language disorder is also a reflection of another disorder of consciousness, i.e. the stream of thought. Even though some important symptoms associated with schizophrenia, such as motor retardation, flattening of affect and muteness, can be assessed on the basis of the patient’s behaviour rather than of his reported conscious experience, none of these symptoms would be sufficient on their own to allow an unequivocal diagnosis of schizophrenia. The only two symptoms which are both necessary and sufficient for a diagnosis of schizophrenia, according to some widely used standardized procedures (Wing, Cooper and Sartorius, 1974; Feighner, Robins, Guze, Woodruff, Winokur and Munoz, 1972), are hallucinations and delusions.

In this paper I shall try to show that a basic disorder of consciousness can account not only for the three principal symptoms of schizophrenia but also for a number of cognitive disorders associated with the illness. In order to throw some light on the nature of this disorder it will first be necessary to discuss some recent experiments concerning the differences between conscious and preconscious processing.

In the last 20 years experimental psychologists have made considerable advances in understanding the various cognitive processes that must be involved when a person interacts with the environment. These studies have provided a vocabulary and a methodology for studying the hypothetical processes carried out by the brain. For example, Broadbent’s (1958) suggestion that human information processing involves a limited capacity system and the ‘filtering’ of information has had a considerable influence. This has lead to various models of schizophrenic cognitive deficit in terms of a defective filtering system (e.g. Payne, Matussek and George, 1959; Cromwell, 1968) and to a number of ingenious experiments on perception in schizophrenic patients.

This model has had success in explaining some of the cognitive disorders found in schizophrenia, but how it accounts for the principal symptoms of the disorder is much less clear. A more serious criticism of the theory is its assumption that stimulus selection in general has become defective. Perception is so dependent on selection that such a defect would result in a reduction in cognitive abilities to a level approaching severe subnormality. In contrast, many schizophrenic patients can function and communicate remarkably well in spite of continuous hallucinations and all-pervasive delusions. Thus a defect in filtering cannot be of a general kind, and it is necessary to specify at what level of processing a breakdown occurs. In addition, the defective filter model does not account for deficits in response selection, although such a defect is clearly an important component in the cognitive abnormalities found in schizophrenia (Broen and Storms, 1967; Chapman, 1966).

The model of schizophrenic cognitive deficit to be presented here, while being closely related to these defective filter theories, attempts to overcome these problems. It incorporates recent evidence on the nature and function of conscious awareness and specifies more precisely the nature and extent of the cognitive deficit. In addition, the model considers deficits in output as well as input processes.

The properties and functions of conscious and preconscious process

1 Preconscious processes

It is by no means a novel idea that certain processes operate outside consciousness (i.e. cannot be observed by introspection). There is now good evidence that these unconscious processes are multitudinous and extremely sophisticated.

Among these unconscious processes, I shall be particularly concerned with those which have been or can become conscious, that is, what Freud termed the preconscious. This restriction excludes various complex control systems, such as those carried out by endocrine or spinal mechanisms. A simple demonstration of one of these preconscious processes is given by Sperling’s (1960) partial report technique. When people are shown very briefly a large array of letters, they can report only about four of these. If, however, they are asked after presentation to report letters from a particular row or a particular column of the array, they can still report about four. Thus for a brief period of time virtually the whole array of letters must have been available in the preconscious. Only about four letters from any part of the array can actually become conscious, although subjects are aware that other letters were present. This is taken as evidence for the existence of a brief, high-capacity ‘iconic’ store below the level of consciousness.

Preconscious processes include not only perceptual inputs but also motor outputs. Almost any repetitive skilled movement, such as typing or riding a bicycle, is carried out without conscious attention, and indeed rapid skilled movements can only be achieved by processes operating below consciousness since the reaction time to consciously perceived signals is too slow to provide the necessary control (Poulton, 1966; Frith, 1973).

Perhaps the best example of a highly sophisticated preconscious process is that involved in word recognition. In order to recognize a written or spoken word the sensory input must undergo graphical or phonological analysis; and, probably at the same time, semantic properties of the word must be processed. The final result of the interaction between these processes is that some model of the actual word presented (its meaning and form) reaches consciousness (Allport, 1976; Marcel and Patterson, 1977).

There are a number of studies indicating that much of this processing occurs at the preconscious level. For example, Mewhort (1967) showed that some processing occurred in iconic memory before items reached awareness. Subjects were asked to report one of two rows of letters; which row was to be reported on was indicated after presentation. More letters could be reported from the target row if the letters in the other row obeyed the laws of English orthography (e.g. CERNALIT) than if they were random, even though the subject was unable to report what was in this second row. Thus subjects must have processed this second row to the extent of making use of the regular sequence of the letters, even though the contents of this row never reached awareness.

Preconscious and conscious processes do not differ from one another simply in their availability for awareness. There are a number of additional properties on which the two kinds of process differ.

2 Capacity

As we have already seen, there is evidence that some preconscious processes involve the storage of large numbers of items (Sperling, 1960). It is not only the iconic store that has a large capacity: the long-term store has an even larger capacity. Material in this latter store is clearly preconscious in the sense given above, and the processes by which material (such as one’s telephone number) is retrieved from this store are largely unavailable to introspection. We also have an enormous preconscious store of English words and their meanings, but we can only be conscious of a very few such words at any time. Atkinson and Shiffrin (1971) and Erdelyi (1974) have argued that the contents of consciousness may be equated with the contents of a short-term memory system of limited capacity (roughly seven items, Miller, 1956). Thus, in contrast to preconscious processes, the capacity of conscious processes is very small.

3 Multiple vs single meanings

If a stimulus is ambiguous and has more than one meaning, only one of these interpretations can be in consciousness at one time. Marcel (1976) has demonstrated experimentally this difference between preconscious and conscious processes, using a word recognition paradigm. The perception of a word is facilitated if a word of similar meaning has just been seen (e.g. BREAD—BUTTER). This also occurs with ambiguous words. Thus PALM will be facilitated if it is preceded by either TREE or HAND. However, once one meaning of PALM has become conscious the other meaning is inhibited. Thus, WRIST will be facilitated if preceded by HAND—PALM, but not facilitated if preceded by TREE—PALM. If, however, by a masking technique, PALM is prevented from reaching consciousness and is only processed in the preconscious, then alternative meanings are not inhibited and WRIST is facilitated when preceded by TREE—PALM.

4 Serial vs parallel processing

We have great difficulty in attending consciously to more than one task at a time. On the other hand, many different preconscious processes can be carried on simultaneously without difficulty. This difference between the two systems is perhaps a consequence of, or another way of describing, the differences in capacity. In conscious processing the relevant items must be dealt with one at a time, in other words serially. If items from more than one task are present, they will interfere with one another (Posner and Klein, 1973; Posner and Snyder, 1975). In a number of experimental situations, such as the dichotic listening task (Broadbent, 1958), people function as if all information had to be processed through a single ‘channel’ of limited capacity. These are all tasks demanding conscious attention. When a task does not demand conscious attention (as when a trained pianist or typist is required to touch a particular key), the person behaves as if the task were controlled by one of many channels simultaneously available for processing information (Shaffer, 1975).

5 Automatic vs strategic processing

Turvey (1974) and Erdelyi (1974) have suggested that preconscious processing is regulatory or automatic, whereas conscious processing is strategic, implying that the crucial feature of conscious processing is flexibility. Automatic or regulatory processing does not involve flexibility; in such processing the repertoire of appropriate stimuli and responses is already known, and there is a fixed relationship between stimulus and response. Such tasks are those that can be carried on in spite of the fact that conscious attention is directed elsewhere, and they can be maintained in parallel with other tasks. Novel tasks in which the repertoire of stimuli and responses is not known and for which there is no fixed relationship between input and output have to be carried out by conscious processes. The learning of a skill involves the development of knowledge about the repertoire of stimuli and responses and the construction of fixed relations between them. Thus the performance of a skill is gradually transferred from a conscious process to an automatic preconscious process (Eysenck and Frith, 1977). To the extent to which our environment is continually presenting novel situations which require novel reactions, conscious processing is crucial for successful interactions with the environment.

6 The role of consciousness

Shallice (1972) has suggested that the role of consciousness is that of a high-level executive system. Such an executive system must select one course of action from among many and then carry it out without hesitation or interruption unless the need for change becomes very strong. To achieve this aim there must be an unambiguous representation of the current situation and the desired goal. In addition, the system must have a certain amount of inertia so that a chosen course of action will not be abandoned simply because an alternative action seems momentarily more desirable. Thus, crucial for the successful functioning of such a system are the various limitations of consciousness discussed above: small capacity, intolerance of ambiguity, and serial processing. What we are conscious of is the selection and carrying out of the chosen course of action (the dominant action system, in Shal-lice’s terminology). The chosen course of action is dominant both in relation to other possible courses of action and in relation to any automatic processes that may be going on simultaneously. In particular, the dominant action system has priority in the interpretation of perceptions and the selection of responses if there is any conflict with other on-going processes.

7 A mechanism for consciousness

Shallice has shown that a simple mechanism can have the properties desired for this executive system. Although this mechanism is not of relevance to the symptoms of schizophrenia, it is of interest in relation to the possible causes of the breakdown, as it suggests the manner in which consciousness may lose its normal role. If each of many possible actions is represented by a single hypothetical neurone, then a network in which each neurone inhibits all the others will have the property that when many of the neurones are stimulated only one will finally become activated. In addition to this capability of selection, the network also has the property of inertia. Once one neurone has ‘captured’ the system it is relatively more difficult for others to take over.

Schizophrenia as a disorder of consciousness

I suggest that the symptoms of schizophrenia occur when the selective capacity of consciousness breaks down. In terms of the filter theory, the filter that has become defective in schizophrenia is the one that determines which items in preconsciousness shall enter awareness. A...

Table of contents

• Cover
• Title
• Copyright
• Contents
• Acknowledgements
• Introduction
• SECTION 1 Schizophrenia
• SECTION 2 Will & consciousness
• SECTION 3 Social cognition
• Index
• Plates