A model for the sequential organization of behaviour.

Abstract

At present, there does not appear to be a satisfactory theory of the sequential organization of behaviour. A number of experiments on tachistoscopic recognition and dichotic listening, in which it is necessary to respond sequentially to simultaneously presented stimuli, are reviewed. The data from these experiments indicate that the order of responses is determined by the spatio-temporal arrangement of die stimulus elements, by the experimental set, and by learned associations between the individual elements. On the basis of these data, a model for serial order is developed and applied to other situations with some success.IN GENERAL, PSYCHOLOGICAL THEORY and experimentation have been concerned with the relation between simple stimulus events and responses. Typically, however, a response to a stimulus is not an isolated bit of behaviour, but rather a single incident in a long series of responses. In speech or piano playing, for example, the important behaviour is not the single response, but the response in its temporal relation to other responses. It is clear that a complex temporal integration of neuronal activity must underlie such behaviour. To date, few attempts have been made to work out the basis of serially ordered behaviour, perhaps because of the complexity of the behaviour involved and the difficulty of analysing it experimentally.The most systematically developed theories of the temporal integration of behaviour are the associative chain theories. Perhaps the best statement of this position has been made by Hull (1930), although essentially similar ideas have been proposed by Bechterev (1932), Washburn (1916), and Watson (1924, among others. More contemporary versions of the associative chain theory have been presented by Berlyne (1965) and 1 The preparation of this paper was facilitated by grants to the author from die National Research Council of Canada (APA-95) and the Defence Research Board of Canada (9401-26), and by D.R.B. Grant No. 9401-11 to D. O. Hebb. I am indebted to Professors Woodburn Heron of McMaster University, D. O. Hebb of McGill University, and R. V. Thysell of the University of Waterloo for their stimulating and critical discussion of an earlier draft of this paper, and to many of my colleagues and students at the University of Waterloo for their comments on the 2

The psychology of intelligence analysis: Drivers of prediction accuracy in world politics.

“…In the demonstrations reported here, we have explored the use of item (word) and associative (word-to-word) information; we acknowledge, however, that word-to-context associations may also be used in the special circumstances explored by Brown et al (2000) or by Bryden (1967). In addition to the information stored in the hologram proper, the theory proposes control mechanisms designed to help extract all useful information from the hologram.…”

“…Indeed, one of the desirable characteristics of our holographic model is the ability to combine sources of information, in particular, item-to-item information and item-to-context information. Bryden's (1967) account is likely the earliest of modern contextbased theory. Based on tasks in which space and time were put in conflict (e.g., dichotic-listening or split-span tasks), he proposed that order of report reflected ranking of items in space or by time, and he provided a neurophysiological model to explain the ranking.…”

“…An easy extension to the mechanism would be to add temporal and/or spatial environmental information (in the form of an association between each word and a corresponding context vector). Such an extension would allow the ordering mechanism to combine internal and environmental context to handle the temporal-spatial conflicts addressed by Bryden (1967), the uneven timing cases explored by Brown et al (2000), and list discrimination studies of the sort pioneered by Jacoby (1991). At issue, is a fundamental computing-science question of data versus process.…”

Scanning strategies and differential sensitivity in a visual signal detection task: Intersubject consistency

“…White (1970) found that targets exposed for 100 msee were gene rally reported better if they were to the right rather than to the left of fixation, though targets at the left of a display were reported first. Following previous suggestions (Heron, 1957;Bryden, 1967), he proposed that there is a postexposure direetional scanning mechanism which operates sequentially upon a stored information trace. Crovitz & Sehiffman (1965), in contrast, found fewer errors for the leftmost target in a line but agreed that such tasks should be eategorized as involving memory rather than perception.…”

Retinal locus and the letter-span error function

“…1a In this experiment, the analysis of data was based on the correct localization of letters. The correct identification of a letter, irrespective of its serial positioning, mechanism which sequentially operates on the stored information trace, proposed by Heron (1957) and Bryden (1967).…”