You are in: Home » History of the Unit
THINKING
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.
Project 69 - The Study of Modelling of Cognitive Skills and Expertise
The general framework of this research is based on two main ideas: the notion that high-level cognition calls for the construction of mental models of the states of the world [371], and that the cognitive processes of thinking, reasoning, and imagination can best be understood in computational terms [368, 367, 333, 366].
69.1 Comprehension of Discourse (Anderson, Johnson-Laird)
According to the theory of mental models, discourse can be understood at two levels — one superficial level that is similar in structure to the sentences of the discourse, and one more profound level that is similar in structure to the states of affairs described by the discourse [364, 340]. The meanings of words must provide the information necessary to construct models of the situations that they can be used to describe [369]. This theory predicts that discourse will be harder to understand if the order of the sentences makes it difficult to retrieve the links between co-referential phrases [336]. It also predicts that discourse will be harder to understand if the events that are described are difficult to integrate using general knowledge. The first factor is peculiar to discourse, but the second applies to the interpretation of any domain, e.g. perceived events. It follows that the two factors should have independent effects, and two recent experiments have confirmed this prediction [388].
69.2 Studies of Inferential Processing
69.2a Informal everyday inferences (Anderson, Johnson-Laird). Reasoning in everyday life often occurs in situations in which there is insufficient information from which to derive a valid deduction. If the process depends on the construction of models of the premises and a search for putative conclusions, then we can predict that such a search will be influenced by several variables [370, 389]. Contrary to strict Skinnerian principles, people search more assiduously when they are told that their previous conclusions are wrong — as opposed to being told that their previous conclusions are possibly correct. Ordinary individuals establish putative conclusions, even between quite disparate events, very rapidly [365].
69.2b Deduction (Byrne, Johnson-Laird). The theory that comprehension and reasoning depend on the construction of mental models of the situations described in the discourse [392, 390, 374, 391] has been extended. Five experiments on propositional reasoning, which depend on such connectives as "if", "and", and "or", have shown that the number of situations that a reasoner has to hold in mind determines the difficulty of making a deduction. This phenomenon was predicted by the theory that reasoning depends on constructing mental models of the situations described by the premises [361, 380], and runs counter to predictions derived from theories which assume that people are equipped with a mental logic containing formal rules of inference. Predictions of the two theories have been compared in two other domains: spatial reasoning, where the difficulty of a deduction was related to the number of different layouts that have to be kept in mind [332]; and reasoning with premises that contain either single quantifiers, such as "only" (four experiments reported in [341], see also [384]), and premises that contain pairs of quantifiers, such as "none" and "all" ([382], three experiments in [342]). In all three domains, the complexity of formal derivations had no relation to the difficulty of the inferential task.
Project 70 - Risk Assessment, Judgement of Causality and Decision Making
70.1 Judgement of Causality (Shanks)
Shanks has investigated the way in which people detect causal relationships, particularly between their own actions and the events caused by them. The processes involved are implicated in many theories in psychopathology, decision¬making, and perception. In this project, a new causality judgement procedure has been developed, but the main concern has been to test a normative theory which proposes that people base their causality judgements on the metric dP, which is the difference between the probability of an outcome given an action [P(0/A)] and the probability of the outcome in the absence of the action [P(0/-A)]. This theory assumes that people maintain subjective representations of these probabilities and take the difference between them as the basis for their causality judgements.
The project has highlighted a wide range of difficulties inherent in this view. For example, it says little about the role of temporal contiguity, yet a series of experiments [356] has shown that as the time interval between the action and outcome increases, judgements of causality decrease. The normative account can only explain this by saying that, as a result of the delay, the action and outcome are no longer perceived together as a pair; the data do not seem to be consistent with this view [355].
A second problem with the normative theory concerns acquisition profiles in causality judgement experiments. As people receive more experience of a causal relationship, their judgements of causality, and their confidence in those judgements, become progressively greater [355]. The normative model has difficulty explaining the details of these learning curves.
The final, and most important, problem for the theory is the demonstration of selectional processes in causality judgement. A cause never occurs in isolation, but in the presence of other potential causal events (the "causal context"). It has been shown [351] that the extent to which a target cause is associated with an outcome depends critically on the extent to which the causal context predicts the outcome, even though this does not alter the normative measure, dP.
In general, the data seem more consistent with an associative model of causality judgement in which an association is formed between the mental representations of the cause and effect. Such a model has been implemented as an OPS-5 production system [385].
70.2 Categorization by Connectionist Networks (Shanks)
Shanks is attempting to determine whether connectionist networks can reproduce the results of a series of experiments [352, 353] on people's ability to deal with novel instances of some category. Learning in the network consists of finding a matrix of weights which will map input patterns (the stimulus to be categorized) onto output patterns (the category). A series of experiments has examined the predictions of this model concerning the way in which the features of an instance become associated with the category. In the experiments, subjects have to diagnose the illness (which constitutes the category) of a series of hypothetical patients on the basis of their symptoms (which constitute the features).
Feature-category associations are formed to the extent that the feature is a good predictor of that category, even though this may affect neither the probability of the category given the feature [P(C/F)] nor the probability of the category in the absence of the feature [P(C/-F)]. It is just such probabilities, according to cue-validity models of categorization, which determine assignment of objects to categories.
For example, in one experiment subjects saw patients with symptoms A and B, who all had disease 1, and a smaller number of other patients with symptoms A and C who all had disease 2. When asked to categorize a new patient with symptoms B and C, the subjects chose disease 2, quite contrary to the base rate of the two illnesses. This phenomenon, which has caused much perplexity in the literature, is readily explained by a connectionist network. In the network, a stronger association is formed between symptom C and disease 2 than between symptom B and disease 1.
The Rescorla-Wagner and Mackintosh theories [352], developed to account for conditioning experiments using animals, are similar to the network model and seem to provide good descriptions of the mechanism responsible for determining the strength of the feature-category associations. The first of these theories is formally equivalent to the delta rule, a widely-used algorithm for adjusting the weights in connectionist networks. Specific predictions from computer-simulations of this theory have been upheld, and so it appears, at present, that the data are consistent with a connectionist model of categorization.
70.3 Classical Conditioning (Levey)
Classical conditioning represents a very simple form of information processing and learning. When studied in human subjects it affords an opportunity to examine the interaction of higher cognitive processes (e.g. decision and planning strategies), with primitive learning mechanisms which lie largely outside conscious awareness and probably function automatically. Several studies (in collaboration with Martin, Institute of Psychiatry, London) based largely on human eyelid conditioning paradigms have suggested that this interaction involves a degree of independence between verbalisable knowledge of the learning task and its actual performance [375, 377, 344]. The classical conditioning technique has also been applied to the modification of preferences. A series of parametric studies has shown that acquisition of preferences for picture materials follows the general principles of simple automatic conditioning [376].
REFERENCES
Al - Authored Books
328. JOHNSON-LAIRD, P.N. (1988) The Computer and the Mind: An Introduction to Cognitive Science. London: Fontana Press.
329. JOHNSON-LAIRD, P.N. and BYRNE, R.M.J. Deduction. Hillsdale, N.J.: Lawrence Erlbaum Associates, in press.
330. POULTON, E.G. (1988) Bias in Quantifying Judgments. Hove, Sussex: Lawrence Erlbaum Associates.
Refereed Journal Articles
331. BYRNE, R.M.J. (1989) Human deductive reasoning. Irish Journal of Psychology: Special Issue on Cognitive Science, 10, 216-231.
332. BYRNE, R.M.J, and JOHNSON-LAIRD, P.N. (1989) Spatial reasoning. Journal of Memory and Language, 28, 564-575.
333. BYRNE, R.M.J, and Keane, M.T.G. (Eds.), (1989) Cognitive Science: A Special Issue of the Irish Journal of Psychology.
334. BYRNE, R.M.J, and Keane, M.T.G. (1989) An introduction to cognitive science. Irish Journal of Psychology: Special Issue on Cognitive Science, 10, i-vi.
335. Gentner, D. and GRUDIN, J. (1985) The evolution of mental metaphors in psychology: A 90-year retrospective. American Psychologist, 40, 181-192.
336. JOHNSON-LAIRD, P.N. (1986) How is meaning mentally represented? Versus, 44/45, 99-118.
337. JOHNSON-LAIRD, P.N. (1986) Human and computer reasoning. Trends in Neurosciences, 8, 54-57.
338. JOHNSON-LAIRD, P.N. (1987) Connections and controversy. Nature, 330, (No.6143), 12-13.
339. JOHNSON-LAIRD, P.N. (1987) The mental representation of the meaning of words. Cognition, 25, 189-211.
340. JOHNSON-LAIRD, P.N. (1988) How is meaning mentally represented? International Social Science Journal (Special Issue Feb. 88) Cognitive Science, 115, 45-61.
341. JOHNSON-LAIRD, P.N. and BYRNE, R.M. (1989) Only reasoning. Journal of Memory and Language, 28, 313-330.
342. JOHNSON-LAIRD, P.N., BYRNE, R.M.J, and Tabossi, P. (1989) Reasoning by model: The case of multiple quantification. Psychological Review, 96, 658-673.
343. JOHNSON-LAIRD, P.N., Oakhill, J.V. and Bull, D. (1986) Children's syllogistic reasoning. Quarterly Journal of Experimental Psychology, 38A, 35-38.
344. LEVEY, A.B. and Martin, I. (1989) Propositional knowledge and mere responding. Biological Psychology, 28, 149-155.
345. Martin, I. and LEVEY, A.B. (1988) Human Pavlovian conditioning. Biological Psychology, 27, 203-206.
346. Oakhill, J., JOHNSON-LAIRD, P.N. and Garnham, A. (1989) Believability and syllogistic reasoning. Cognition, 31, 117-140.
347. POULTON, E.C. (1986) Why unbiased numerical magnitude judgments of the loudness of noise are linear in decibels: A rejoinder to the Teightsoonians. Perception and Psychophysics, 40, 131-134.
348. POULTON, E.C. (1987) Bias and range effects in sensory judgements. Chemistry and Industry, 1, 18-22.
349. SHANKS, D.R. (1986) Selective attribution and the judgment of causality. Learning and Motivation, 17, 311-334.
350. SHANKS, D. (1987) Acquisition functions in contingency judgment. Learning and Motivation, 18, 147-166.
351. SHANKS, D.R. (1989) Selectional processes in causality judgment. Memory and Cognition, 17, 27-34.
352. SHANKS, D. Connectionism and the learning of probabilistic concepts. Quarterly Journal of Experimental Psychology, in press.
353. SHANKS, D. Connectionism and human learning: Critique of Gluck and Bower (1988). Journal of Experimental Psychology: General, in press.
354. SHANKS, D. Contingency awareness in evaluative conditioning: A comment on Baeyens, Eelen and van den Bergh. Cognition and Emotion, in press.
355. SHANKS, D. and Dickinson, A. (1987) Associative accounts of causality judgement. In G.H. Bower (Ed.), The Psychology of Learning and Motivation, Vol. 21. New York: Academic Press Ltd., pp.229-261.
356. SHANKS, D.R., Pearson, S.M. and Dickinson, A. (1989) Temporal contiguity and the judgement of causality by human subjects. Quarterly Journal of Experimental Psychology, 41B, 139-159.
357. WASTELL, D.G. and NIMMO-SMITH, I. (1986) The polarity coincidence correlator: Significance testing and other issues. Bulletin of the Psychonomic Society, 24, 211-212.
C - Invited Chapters and Commentaries
358. BYRNE, R.M.J, and JOHNSON-LAIRD, P.N. Remembering conclusions we have inferred: What biases reveal. In J.-P. Caverni, J.-M. Fabre and M. Gonzalez (Eds.), Cognitive Biases: Their Contribution for Understanding Human Cognitive Processes. Amsterdam: Elsevier (North-Holland), in press.
359. JOHNSON-LAIRD, P.N. (1985) Deductive reasoning ability. In R.J. Sternberg (Ed.), Human Abilities: An Information-Processing Approach. New York: Freeman, pp.173-194.
360. JOHNSON-LAIRD, P.N. (1985) Logical thinking: Does it occur in daily life? Can it be taught? In S. Chipman, J. Segal and R. Glaser (Eds.), Thinking and Learning Skills, Vol. 2: Research and Open Questions. Hillsdale, N.J.: Lawrence Erlbaum Associates, pp.293-318.
361. JOHNSON-LAIRD, P.N. (1986) Conditionals and mental models. In E.C. Traugott, A. ter Meulen, J.S. Reilly and C. Ferguson (Eds.), O n Conditionals. Cambridge: Cambridge University Press, pp.55-75.
362. JOHNSON-LAIRD, P.N. (1986) Reasoning, imagining and creating. Geest, Computer, Kunst, 7, 298-316. (See also APU/2055)
363. JOHNSON-LAIRD, P.N. (1986) Reasoning without logic. In T. Myers (Ed.), Reasoning and Discourse Processing. London: Academic Press Ltd., pp.13-49.
364. JOHNSON-LAIRD, P.N. (1987) Grammar and psychology. In S. Modgil and C. Modgil (Eds.), Noam Chomsky: Consensus and Controversy. Lewis, Sussex: Falmer Press, pp.147-156.
365. JOHNSON-LAIRD, P.N. (1987) Reasoning, imagining, and creating. Bulletin of the British Psychological Society, 40, 121-129.
366. JOHNSON-LAIRD, P.N. (1988) A computational analysis of consciousness. In A.J. Marcel and E. Bisiach (Eds.), Consciousness in Contemporary Science. Oxford: Oxford University Press, pp.357-368.
367. JOHNSON-LAIRD, P.N. (1988) A taxonomy of thinking. In R.J. Sternberg and E.E. Smith (Eds.), The Psychology of Human Thought. New York: Cambridge University Press, pp.429-457.
368. JOHNSON-LAIRD, P.N. (1988) Freedom and constraint in creativity. In R.J. Sternberg (Ed.), The Nature of Creativity: Contemporary Psychological Perspectives. Cambridge: Cambridge University Press, pp.202-219.
369. JOHNSON-LAIRD, P.N. (1988) On opening the dictionary. In W. Hirst (Ed.), The Making of Cognitive Science: Essays in Honor of George A. Miller. Cambridge: Cambridge University Press, pp. 186-196.
370. JOHNSON-LAIRD, P.N. (1989) Analogy and the exercise of creativity. In S. Vosniadou and A. Ortony (Eds.), Similarity and Analogical Reasoning. New York: Cambridge University Press, pp.313-331.
371. JOHNSON-LAIRD, P.N. (1989) Human experts and expert systems. In L.A. Murray and J.T.E. Richardson (Eds.), Intelligent Systems in a Human Context. Oxford: Oxford University Press, pp.35-46.
372. JOHNSON-LAIRD, P.N. Human thinking and mental models. In K. Said et al. (Eds.), Modelling the Mind. Oxford: Oxford University Press, in press.
373. JOHNSON-LAIRD, P.N. Jazz improvization: A theory at the computational level. In P. Howell, R. West and I. Cross (Eds.), Representing Musical Structure. Academic Press, in press.
374. JOHNSON-LAIRD, P.N. The development of reasoning ability. In G. Butterworth and P. Bryant (Eds.), Causes of Development: Interdisciplinary Perspectives. Hemel Hempstead, Herts.: Harvester Wheatsheaf, in press.
375. LEVEY, A.B. and Martin, I. (1987) Knowledge, action, and control. In H.J. Eysenck and I. Martin (Eds.), Theoretical Foundations of Behavior Therapy. New York: Plenum Press, pp.133-151.
376. LEVEY, A.B. and Martin, I. (1987) Evaluating conditioning: A case for hedonic transfer. In H.J. Eysenck and I. Martin (Eds.), Theoretical Foundations of Behavior Therapy. New York: Plenum Press, pp.113-131.
377. Martin, I. and LEVEY, A.B. (1987) Learning what will happen next: Conditioning, evaluation and cognitive processes. In G. Davey (Ed.), Cognitive Processes and Pavlovian Conditioning in Humans. Chichester: John Wiley, pp.57-81.
378. POULTON, E.C. (1989) Uncertain size of exponent when judging without familiar units. Behavioral and Brain Sciences, 12, 286-288.
379. SHANKS, D.R. and Dickinson, A. (1988) The role of selective attribution in causality judgement. In Contemporary Science and Natural Explanation: Commonsense Conceptions of Causality. Sussex: The Harvester Press, pp.94-126.
D - Conference Proceedings
380. BYRNE, R.M.J, and JOHNSON-LAIRD, P.N. Models and deductive reasoning. In K.J. Gilhooly, M.T.G. Keane, R.H. Logie and G. Erdos (Eds.), Lines of Thinking: Reflections on the Psychology of Thought. Chichester : John Wiley and Sons, in press.
381. BYRNE, R.M.J, and Keane, M.T.G. Reasoning. In K.J. Gilhooly, M.T.G. Keane, R.H. Logie and G. Erdos (Eds.), Lines of Thinking: Reflections on the Psychology of Thought, Vol. 1. Chichester: John Wiley & Sons Limited, in press.
382. JOHNSON-LAIRD, P.N. (1988) Reasoning by rule or model? In Proceedings of the Tenth Annual Conference of the Cognitive Science Society, Hillsdale, N.J.: Lawrence Erlbaum Associates, pp.765-771.
383. JOHNSON-LAIRD and BYRNE, R.M.J. Mental models. In H. Yoshikawa and T. Holden (Eds.), Proceedings of IFIP WG 5.2 Second Workshop on Intelligent CAD II. CAD. Elsevier Science Publications, B.V. (North-Holland), in press.
384. Oakhill, J., Garnham, A. and JOHNSON-LAIRD, P.N. Belief bias effects in syllogistic reasoning. In K.J. Gilhooly, M.T.G. Keane, R.H. Logie and G. Erdos (Eds.), Lines of Thinking: Reflections on the Psychology of Thought, Vol. 1. Chichester: John Wiley and Sons, in press.
385. SHANKS, D.R. and Pearson, S.M. (1987) A production system model of causality judgement. In Proceedings of the Ninth Annual Cognitive Science Society Meeting (Seattle, WA, July 1987). Hillsdale, N.J.: Lawrence Erlbaum Associates, pp.210-220.
E - Technical Reports, Theses and Tests F - Dissemination
386. GROEGER, J.A. (1987) Computation: The final metaphor? An interview with Philip Johnson-Laird. New Ideas in Psychology, 5, 294-304.
387. JOHNSON-LAIRD, P.N. (1987) How could consciousness arise from the computations of the brain? In C. Blakemore and S. Greenfield (Eds.), Mind Waves. Oxford: Basil Blackwell Ltd., pp.246-257.
388. JOHNSON-LAIRD, P.N. (1987) The comprehension of discourse and mental models. In M. Nagao (Ed.), Language and Artificial Intelligence. Elsevier Science Publishers, B.V. (North-Holland), pp.253-261.
389. JOHNSON-LAIRD, P.N. (1988) Editorial: A case for imagination in human reasoning. Logic, 3, p.2.
390. JOHNSON-LAIRD, P.N. Human thinking and mental models. In R. Viale (Ed.), Human Mind, Artificial Mind. Turin: Editori La Rosa, in press.
391. JOHNSON-LAIRD, P.N. (1989) Mental models. In M. Posner (Ed.), Foundations of Cognitive Science. Cambridge, MA: MIT Press, pp.469-499.
392. JOHNSON-LAIRD, P.N. Modeles mentaux en science cognitive. Bulletin de Psychologie, in press.
393. POULTON, E.C. (1987) Quantifying judgements. In R.L. Gregory (Ed.), Oxford Companion to the Mind. Oxford: Oxford University Press, pp.667-670.
Other sections in the 1985-1989 report
3. LANGUAGE, SPEECH, READING AND WRITING