You are in: Home » History of the Unit
ATTENTION Project 57 - Control and Function of Selective Attention
General Notes. This material has been scanned from the original typescript. While we have done our best to remove errors, some may well remain. You can access other parts of this particular Progress Report either from the menu at the bottom of this entry, by using your browser's back function, by navigating back to the Unit History Timeline, or by accessing the relevant section of the electronic archive. Reference for this report are indexed by number and these can be found in a dedicated section also accessible from the menu at the bottom of this entry.
57.1 Visual Attention (Duncan)
In this reporting period there has been substantial progress in understanding the stimulus principles governing the efficiency of visual selective attention. The theory outlined in the last Report [7] has seen several major developments. In its present form [13] this theory integrates a wide range of experimental findings, relates sensibly to the requirements of normal vision, and suggests a possible neurophysiological implementation which is to be investigated in a new, collaborative project with N.I.M.H. in Washington D.C.
Much of the experimental work in this period has involved visual search [10, 11, 13]. Alternative accounts of search efficiency have rested on distinctions between (a) serial vs. parallel search; (b) single visual features vs. feature conjunctions; (c) stimulus identification vs. local mismatch detection. Duncan's work suggests none of these is crucial. For all search materials, efficiency increases with (a) decreasing similarity of targets to nontargets, and (b) increasing similarity between nontargets. The interaction between these has been explored in detail. We must also distinguish similarity between possible stimulus alternatives, and similarity between stimuli within a display.
The findings suggest the following theory. A parallel stage of perceptual grouping and description is followed by competitive interaction between inputs, guiding selective access to awareness and action. An input gains weight to the extent that it matches an internal description (the attentional template) of the information needed to control current behaviour (hence the importance of target-nontarget similarity). Perceptual grouping encourages input weights to change together (weight linkage; allowing spreading suppression of similar nontargets). The idea of attentional templates allows control of visual selection to be indefinitely flexible. Depending on current concerns, inputs with any visual property (e.g. location, level of scale) can be relevant to behaviour.
Because of weight linkage, it should be easy to divide attention between visual inputs as long as there is strong perceived grouping between them. Preliminary evidence was described in the previous Report, and the prediction has now been confirmed in a long series of studies involving absolute identification of shape, orientation, spatial frequency, location, colour, brightness and texture. In each case it is very easy to identify two properties of the same perceived object in a brief display, but more difficult to identify properties of two different objects. This work also shows that, when attention is divided between objects, it does not matter whether they vary in the same or different attributes. (Something of an exception is colour, whose unique behaviour is still not understood.) Despite partial specialization of separate, parallel cortical areas for analyzing different visual attributes, there is no evidence for correspondingly parallel perceptual function.
Perceptual grouping may also be the key to the problem of tag assignment, or keeping together parts of the same whole in a distributed perceptual representation. Several connectionist models achieve accurate tagging by dealing serially with separate input locations, but the problem need not be solved serially, nor is common location always the important tag [26]. When attention is diverted, it is true that some stimuli (e.g. words) tend to produce percepts that are incorrect combinations of parts actually present, but others (e.g. single letters) do not [10]. Even when explicit identification of an unattended shape is at chance [6], the way that its parts are combined can influence its effects on a second, attended shape [10].
Project 58 - Executive Processes and their Pathology
58.1 Executive Functions and Intelligence (Duncan, Bourke)
An attempt is under way to link findings in neuropsychology, individual differences, and dual task performance into a unified account of executive function. In neuropsychology the "frontal lobe syndrome" reflects widespread disorganization in many aspects of behaviour. In individual differences, "intelligence tests" (even those with restricted content) show broad positive correlations with many different sorts of laboratory and real world activity. In dual task studies, interference is partly content-specific but also occurs between very dissimilar tasks. Duncan has suggested that all these phenomena reflect the widespread importance of a cognitive executive dealing with choice of goals in a complex environment [9, 12].
A common element between general intelligence (g) and interference between dissimilar tasks has been confirmed in a study of driving skills (see Section 73.1 [409]). Across a range of such skills, profiles of correlation with g and of dual task decrement are in excellent agreement. That frontal lobe damage reflects an impairment in g is generally denied; the reasons, however, seem poor, and new evidence suggests a link. A test has been developed to mimic, in normal subjects, one classic aspect of frontal disorder: neglect of a goal despite the fact that it is understood and the failure is appreciated. This test correlates about .5 with g, and is performed very poorly by frontal syndrome patients. Another new test which correlates even more strongly with g, though it measures RT rather than accuracy, uses fairly simple stimuli and contains no element of problem solving. In common with standard tests of frontal function this task requires frequent shifting of attention between different stimulus attributes, and again patients with a frontal syndrome perform very poorly.
Behavioural definition of the frontal syndrome is very loose, and despite the name, the underlying neuropathology is poorly understood. A grant from the Regional Health Authority has allowed us to assess a broad range of frontal impairments in 100 head injury patients, collaborating with Roger Johnson from the Rehabilitation Unit at Addenbrooke's Hospital. We use a structured clinical interview, a relative's questionnaire, and neuropsychological tests to study emotional, social, motivational and cognitive changes. Multivariate analyses will be used to define one or more "frontal syndromes", allowing principled selection of patients for future work.
In a final project Bourke is asking whether a single, general factor is sufficient to explain interference between different, dissimilar task pairs. By broad and systematic sampling of such pairs, this work assesses whether the order of primary task demands on the (hypothetical) general factor is consistent across secondary tasks. Results to date in fact suggest considerable consistency.
58.2 Executive Processes and the Frontal Lobes (Burgess, Shallice, Wilkins)
In an article covered in the last Progress Report, Norman and Shallice put forward the position that the action and thought selection process should be divided into two levels. 'Contention scheduling', the lower-level of the two, is a process which we held carries out routine selection of routine action or thought schemata/procedures. This selection process is, we argued, modulated by a higher level Supervisory System which can provide extra activation or inhibition to individual schemata in contention scheduling (see [1], chapter 14 for the application of the theory to frontal lobe disorders]. Three empirical studies have developed from this theoretical perspective:-
58.2a Incidental Utilisation Behaviour.
Lhermitte (1983) has described patients with frontal lobe lesions who when presented with a familiar object (e.g. glass, jug, comb, glasses) and no instructions will start to use the object. This behaviour is what one would expect from the operation of contention scheduling if Supervisory modulation of the system was grossly impaired as schemata specific to the objects would be triggered. However an alternative interpretation is that the frontal patient merely misinterprets the task demands of the situation. Shallice, Burgess, Schon and Baxter described a patient with bilateral medial frontal lesions who manifested utilisation behaviour when tested clinically using Lhermitte's procedure [20]. However he also exhibited it when familiar objects were present on the testing desk while he was being instructed to carry out clinical neuropsychological tests. A demand characteristic artefact could therefore be excluded; an interpretation in terms of an impaired Supervisory System fitted the findings.
582b Disorders of Numerosity Judgements.
Concentration requires the operation of the Supervisory System to prevent the type of distracting triggering shown in the previously discussed syndrome. Performance on a test of concentration was found to be impaired following right frontal lesions [23].
58.2c Strategy Application Disorder Syndrome.
Shallice and Burgess argue that damage to the Supervisory System should produce problems for the efficient organization of a number of unrelated activities that have to be carried out over a period of time. A case-study of three head-injury patients with frontal lesions who had preserved intellectual skills (IQs of 120+) showed that they had severe difficulties on two tasks of this type. On the other hand, two of the patients were virtually unimpaired when other tasks sensitive to frontal lobe lesions were carried out.
These findings provide quantitative support for earlier clinical claims that a cognitive impairment producing severe judgement and organisation problems in everyday life can occur in certain frontal head injury patients who show no deficit on standard intelligence or frontal lobe problems. They also suggest that the Supervisory System fractionates with these patients having particular difficulties in the interruption of on-going behaviour by stimuli linked to intentions.
58.3 Frontal Lobe Memory Disorders (Burgess, Shallice)
In an extension of the above theory ([1] - chapter 15) it was argued that the description and verification processes involved in episodic memory retrieval would be evolutionarily linked to the functioning of the Supervisory System. A deficit of such processes can account for the characteristics of a highly selective memory problem analysed in a single case study [5]. The patient had suffered from a ruptured anterior communicating artery aneurysm; he had relatively preserved performance - measured in terms of items correctly retrieved - on a range of recall tests but performed in the amnesic range on many recognition tests and confabulated in both formal testing and on autobiographical recall. In a second study Burgess, Baxter, Rose and Alderman have described a new form of circumscribed delusional misidentification syndrome [4]. Neuropsychological, neurological and neuroradiological evidence supports frontal lobe involvement in the syndrome and a model related to the previously mentioned one was proposed which accounts for the paramnesic aspects of the phenomenon. Further evidence of an association between frontal amnesia and confabulation is provided by a study by Baddeley and Wilson [3].
58.4 Inference from Neuropsychological and Dual-Task Evidence to 'Normal' Functional Architecture (Shallice)
Shallice gave a detailed analysis of the assumptions necessary and the supporting evidence for drawing inferences from neuropsychological data for theories of normal cognitive functioning. No equivalent analysis has been carried out except by Caramazza, whose assumptions and conclusions - which are very different - are open to a number of criticisms ([1] chaps. 2, 9,10,11).
In conjunction with McLeod and Lewis, Shallice has studied two sub¬components of the language system which, on neuropsychological grounds can be considered isolable; the dual tasks of reading aloud and detecting a word from a target category in an auditory stream of words, which should require these two isolable sub-systems, can be carried out simultaneously with only a small loss in overall accuracy [22]. However two tasks (e.g. detecting a word in a target category on one ear while shadowing the input on the other ear), which on neuropsychological grounds would be assumed to require a common subsystem are strongly mutually interfering in normal subjects. That neuropsychological and dual task methodology independently lead to similar conclusions supports both approaches.
58.5 Rehabilitation Following Severe Neurological Damage (Burgess)
Two cases of behavioural dyscontrol following frontal lobe injury were successfully treated by Burgess in collaboration with Alderman of St Andrews ITospital in Northampton [25]. Detailed analysis of the symptoms and neurology of the patients demonstrated the utility of a cognitive neuropsychology characterisation of frontal lobe function - related to that discussed in section 1 -for the planning of appropriate treatment procedures. A second study [24] demonstrated that whilst treatment that focuses on a patient's action routines may have short-term success, to achieve long-term success and generalisation once the treatment framework is removed may require that the programmes focus on 'cognitive schemata'. Six single case studies are described where traditional behavioural methods failed to maintain improvement but once combined with a cognitive component the improvement was maintained at follow-up.
Project 59 - Analysis and Function of Conscious Awareness
59.1 Theoretical Treatment of Consciousness (Marcel, Shallice)
Marcel has edited (with Bisiach) a book on approaches to consciousness in philosophy, psychology and neuroscience [2] containing several contributions from the APU (see [34] and [368]). An introductory chapter by Marcel and Bisiach [32] reviews the problems of the topic for functionalist and biological science and the possible approaches. Marcel's own chapter [30] argues for the centrality, legitimacy, and causal status of subjective phenomenal experience, and also discusses some of the currently controversial issues in empirical studies of nonconscious processing (see also [28]).
Two publications [31, 38] discuss the philosophical issues and the problems for information-processing models of Intentionality - the content or "aboutness" of mental states such as images or propositional attitudes.
59.2 Consciousness and Information-Processing Theory (Shallice)
Shallice has argued that the standard method of attempting to relate consciousness with information-processing models is inappropriate [34]. On that approach it is assumed that consciousness corresponds to some aspect of the input, storage or output of some particular subsystem. Instead it is argued that consciousness corresponds to information multiply represented over a very short period of time in a number of different control systems.
59.3 Blindsight, Visual Masking and Dissociation of Indices of Awareness (Marcel, Price)
Following work reported in the last progress report [17], studies of blindsight and normal visual perception have become more integrated. If unilaterally cortically blind subjects showing 'blindsight' (nonconscious visual processing) are asked to indicate when they 'feel' that a light has been presented, there is a dissociation in sensitivity, independent of criterion differences, between voluntary eye-blink, finger button-press, and oral responses. Blink responses become more accurate without feedback and transfer to the finger but not to verbal responses. There are separable effects of delay and modality of response. This has been replicated in normal subjects with threshold-luminance stimuli. The implied dependence of perception on response system or intention (under investigation), casts doubt on the assumed unity of phenomenal experience and the assumed equivalence of psychophysical indices.
This dovetails with work conducted by Price with Marcel on the measurement of awareness thresholds for stimuli under backward visual masking in normal subjects. When target-mask onset asynchrony is reduced, there are dissociations (a) between different indices of confidence, (b) between confidence and performance, and (c) between performance of different target discrimination tasks. Present results also suggest that evidence of nonconscious processing depends on the relation between stimuli and probe task.
REFERENCES Al - Authored Books
1. SHALLICE, T. (1988) From Neuropsychology to Mental Structure. New
York: Cambridge University Press.
A2 - Edited Books
2. MARCEL, A.J. and Bisiach, E. (Eds.) (1988) Consciousness in Contemporary Science, Oxford: Oxford University Press.
B - Refereed Journal Articles
3. BADDELEY, A.D. and Wilson, B. (1988) Frontal amnesia and the dysexecutive syndrome. Brain and Cognition, 7, 212-230.
4. BURGESS, P.W., Baxter, D.M., Rose, M. and Alderman, N. The role of the frontal lobes in delusional paramnesic misidentification. Manuscript submitted.
5. Delbecq-Derouesne, J., Beauvois, M-F. and SHALLICE, T. Preserved recall versus impaired recognition: A case study. Brain, in press.
6. DUNCAN, J. (1985) Two techniques for investigating perception without awareness. Perception and Psychophysics, 38, 296-298.
7. DUNCAN, J. (1985) Visual search and visual attention. In M.I. Posner and O. Marin (Eds.), Attention and Performance XI. Hillsdale, N.J.: Lawrence Erlbaum Associates, pp.85-106.
8. DUNCAN, J. (1986) Consistent and varied training in the theory of automatic and controlled information processing. Cognition, 23, 279-284.
9. DUNCAN, J. (1986) Disorganisation of behaviour after frontal lobe damage. Cognitive Neuropsychology, 3, 271-290.
10. DUNCAN, J. (1987) Attention and reading: Wholes and parts in shape recognition. A tutorial review. In M. Coltheart (Ed.), Attention and Performance XII: The Psychology of Reading. London: Lawrence Erlbaum Associations, pp.39-61.
11. DUNCAN, J. (1989) Boundary conditions on parallel processing in human vision. Perception, 18, 457-469.
12. DUNCAN, J. Goal weighting and the choice of behaviour in a complex world. Ergonomics, in press.
13. DUNCAN, J. and Humphreys, G.W. (1989) Visual search and stimulus similarity. Psychological Review, 96, 433-458.
14. GROEGER, J.A. (1988) Qualitatively different effects of undetected and unidentified auditory primes. Quarterly Journal of Experimental Psychology, 40A, 323-339.
15. McLeod, P.D., SHALLICE, T. and WATSON, F. Semantic activation without explicit identification: A demonstration with Rapid Serial Visual Presentation. Manuscript submitted.
16. MARCEL, A.J. A tactile illusion produced by vision. Manuscript submitted.
17. MARCEL, A.J. and WILKINS, A.J. Cortical blindness and blindsight: A problem of visual function or visual consciousness. Manuscript submitted.
18. MARCEL, A.J. How do you feel? Bodily sensation as representation inference and symbolism. Manuscript in preparation.
19. SHALLICE, T. and BURGESS, P.W. Strategy application disorders in frontal lobe patients. Manuscript in preparation.
20. SHALLICE, T., BURGESS, P.W., Schon, F. and Baxter, D.M. (1989) The origins of utilisation behaviour. Brain, 112, 1587-1598.
21. SHALLICE, T. and Jackson, M. (1988) Lissauer on Agnosia. Cognitive Neuropsychology, 5, 153-192.
22. SHALLICE, T., McLeod, P. and Lewis, K. (1985) Isolating cognitive modules with the dual task paradigm: Are speech perception and production separate processes? Quarterly Journal of Experimental Psychology, 37A, 507-532.
23. WILKINS, A.J., SHALLICE, T. and McCarthy, R. (1987) Frontal lesions and sustained attention. Neuropsychologia, 25, 359-365.
C - Invited Chapters and Commentaries
24. Alderman, N. and BURGESS, P.W. Treatment success and failure: A pragmatic approach to rehabilitation following severe brain injury. In R.L. Wood and I. Fussey (Eds.), Cognitive Rehabilitation in Perspective. London: Taylor and Francis, in press.
25. BURGESS, P.W. and Alderman, N. A cognitive neuropsychological approach to the treatment of dyscontrol syndromes following frontal lobe injury. In R.L. Wood and I. Fussey (Eds.), Cognitive Rehabilitation in Perspective. London: Taylor and Francis, in press.
26. DUNCAN, J. (1989) Parallel processing: Giving up without a fight.
Behavioral and Brain Sciences, 12, 402-403.
27. GROEGER, J.A. (1987) On not knowing the meaning of words we can detect: Crucial qualitative differences. Behavioral and Brain Sciences, 10,765-766.
28. MARCEL, A.J. (1986) Consciousness and processing: Choosing and testing a null hypothesis. Commentary on Holender (1986). Behavioral and Brain Sciences, 9, 40-41.
29. MARCEL, A.J. (1988) Electrophysiology and meaning in cognitive science and dynamic psychology - Comments on "Unconscious conflict: A convergent psychodynamic and electrophysiological approach". In M.J. Horowitz (Ed.), Psychodynamics and Cognition. Chicago: The University of Chicago Press, pp. 169-201.
30. MARCEL, A.J. (1988) Phenomenal experience and functionalism. In A.J. Marcel and E. Bisiach (Eds.), Consciousness in Contemporary Science. Oxford: Oxford University Press, pp.121-158.
31. MARCEL, A.J. Intentionality, consciousness and functionalism. Invited paper to appear in Mind and Language. Manuscript in preparation.
32. MARCEL, A.J. and Bisiach, E. (1988) A cautious welcome: An introduction and guide to the book. In A.J. Marcel and E. Bisiach (Eds.), Consciousness in Contemporary Science. Oxford: Oxford University Press, pp.1-15.
33. Norman, D.A. and SHALLICE, T. (1986) Attention to action: Willed and automatic control of behavior. In R.J. Davidson, G.E. Schwartz and D. Shapiro (Eds.), Consciousness and Self Regulation: Advances in Research and Theory, Vol.4. New York: Plenum Press, pp.1-18.
34. SHALLICE, T. (1988) Information-processing models of consciousness: Possibilities and problems. In A.J. Marcel and E. Bisiach (Eds.), Consciousness in Contemporary Science. Oxford: Oxford University Press, pp.305-333.
D - Conference Proceedings
35. ELLIS, J. (1988) Memory for future intentions: Investigating pulses and steps. In M.M. Gruneberg, P. Morris and R.N. Sykes (Eds.), Practical Aspects of Memory: Current Research and Issues, Vol. 1: Memory in Everyday Life. Chichester: John Wiley, pp.371-376.
36. LOGIE, R., BADDELEY, A.D., Mane, A., Donchin, E. and Sheptak, R. (1988) Working memory in the acquisition of complex cognitive skills. In A.M. Colley and J.R. Beech (Eds.), Cognition and Action in Skilled Behaviour. Elsevier Science Publishers B.V. (North-Holland), pp.361-377.
E - Technical Reports, Theses and Tests F - Dissemination
37. DUNCAN, J. (1985) Attention. In A. Kuper and J. Kuper (Eds.), The Social Science Encyclopaedia. London: Routledge and Kegan Paul, p.50.
38. MARCEL, A.J. Intentionality. Invited contribution to Blackwell's Dictionary of Cognitive Psychology. Manuscript in preparation.
39. MARCEL, A.J. Consciousness. Invited contribution to Blackwell's Dictionary/ of Cognitive Psychology. Manuscript in preparation.
40. SHALLICE, T. (1985) Consciousness and its disorders. In A. Kuper and J.Kuper (Eds.), The Social Science Encyclopaedia. London: Routledge and Kegan Paul, pp.151-153.
Other sections in the 1985-1989 report
3. LANGUAGE, SPEECH, READING AND WRITING