Michel Treisman

Introduction

Michel Treisman (1929–2023) was a British experimental psychologist, Emeritus Reader in Experimental Psychology at the University of Oxford, whose 1963 monograph on temporal discrimination became one of the most cited works in the study of time perception. He is best known for developing an information-processing model of timing built around an arousal-sensitive internal pacemaker, a framework that anticipated much of what the field later took up.

His route into psychology was unusual. Born and raised in South Africa, he qualified as a physician before turning to psychology at Oxford only in the mid-1950s, and his earliest published work was in psychophysics and sensory thresholds rather than in timing. The study of time, which came to define his reputation, began almost by accident, out of experiments on the effect of warning signals on the threshold for electrically induced visual phosphenes.

Early Life and Education

Treisman was born in South Africa in 1929 and trained in medicine there, qualifying as a physician. He moved to Oxford in the mid-1950s to take up psychology, holding the degrees of M.B., B.Ch., M.A., and D.Phil. The clinical training is not incidental: it is part of why his psychophysics is concerned with thresholds, sensory physiology, and the quantitative treatment of a subject's report, and why his later timing work stayed close to the mechanics of the nervous system rather than to the language of judgement.

His first decade of research was in psychophysics. Working with C. I. Howarth, he published a series of papers from 1958 onward on how a warning signal in one modality affects the threshold for a stimulus in another, work that led to a note on the statistical treatment of the brightness increment threshold in 1961 and, in the same year, a paper on the nature of the psychophysical law and the scaling of sensory magnitudes. He also collaborated with Ian Oswald and Anne M. Taylor, the future Anne Treisman, on discriminative responses to stimulation during human sleep in 1960, and wrote a 1962 paper on the "private data" hypothesis in psychological explanation.

During this period he married Anne Treisman in 1960; the marriage ended in divorce in 1976. The two were professional colleagues before they were married, and the sleep paper predates the marriage — a reminder that the collaboration was not an extension of the relationship but the setting in which it began.

Academic and Professional Career

Treisman spent his career at Oxford, in the Department of Experimental Psychology, and retired as Emeritus Reader. He remained scientifically active well into his eighties, attending conferences and questioning speakers with a rigour that colleagues remembered vividly, including as keynote speaker at the final conference of the TIMELY project.

The trajectory of his reputation is unusual. The 1963 monograph, for all its scope, met with little immediate response. As he noted himself, there was not much interest in the study of time at that period, nor for some time afterwards. Recognition came decades later, as time perception became a substantial field and his monograph emerged as one of its most cited works. The delay has a cost that is easy to overlook: a model that is read for the first time thirty years after it was written is read as a precursor rather than as a live option, and it is judged against theories that were built in the interval.

He returned to timing research roughly thirty years after the monograph, in a second programme of work beginning with a 1990 paper with Faulkner, Naish, and Brogan. He died in 2023 at the age of ninety-three.

Major Contributions to Psychology

The 1963 monograph

"Temporal discrimination and the indifference interval: Implications for a model of the 'internal clock'" reported seven experiments using the methods of production, reproduction, constant stimuli, single stimuli, and estimation. Treisman found that the Weber function gave a good fit to the relation between the difference threshold and the base interval, and he obtained evidence that the dip in the Weber function at short intervals is not an essential feature of time estimation but may be an artefact of method. Classic questions about the conformity of timing data to Weber's Law and Vierordt's Law were central to the work, as was his interest in the systematic lengthening or shortening of apparent durations over the course of an experimental session.

The indifference interval of the title is the duration at which estimates are on average correct — the point between the short durations people overestimate and the long durations they underestimate. It is a small, stubborn regularity, and it is the kind of thing a model of timing has to produce rather than assume. In his data it sat at well under a second, and explaining why it sits where it does, and why it moves with arousal, is what the machinery of the 1963 model was built to do.

The information-processing model of timing

The theoretical model that came out of the monograph was its most consequential part. It embedded the old idea of an arousal-sensitive internal pacemaker within a general cognitive architecture involving memory and decision processes, what Treisman later called the "clockwork." His argument was that without that surrounding machinery, performance on any real timing task cannot be properly addressed, whatever one proposes about the timekeeper itself, a point he felt the field had still not fully absorbed decades later.

Two features set the model apart. It dealt explicitly, and for the first time, with how time intervals are stored, retrieved, and compared across different timing tasks. And the diagrammatic presentation was supplemented by an elaborate algebraic specification of how the components operated, allowing far more precise quantitative predictions than had previously been possible.

The calibrated pacemaker

In his later work Treisman explored what he called a calibrated pacemaker model, in which the normally stable output of an oscillator-like pacemaker is modified by a calibration unit whose setting is determined by arousal. The method used a form of interference: trains of repetitive stimulation at different frequencies, on the assumption that external stimulation close to the oscillator's period would produce observable changes in timing performance and thereby reveal what that period was.

Psychophysics and thresholds

Though overshadowed by the timing work, his earlier contributions on sensory thresholds, cross-modal warning effects, and the psychophysical law remain part of the record of British psychophysics in the period before cognitive psychology fully displaced it.

Major Publications

Temporal discrimination and the indifference interval: Implications for a model of the "internal clock" (1963) — the monograph that established his reputation, reporting seven experiments and setting out the information-processing model of timing.
"The nature of the psychophysical law and its significance for the scaling of sensory magnitudes" (1961) — on the foundations of sensory scaling.
"Theory of the visual threshold" (1961) — a note on the statistical treatment of the brightness increment threshold.
"Psychological explanation: the 'private data' hypothesis" (1962) — a contribution to the methodological debate over introspective evidence.
"Discriminative responses to stimulation during human sleep" (1960, with Ian Oswald and Anne M. Taylor) — early collaborative work on responsiveness during sleep.
Treisman, Faulkner, Naish, and Brogan (1990) — the first of the calibrated pacemaker studies, beginning his second programme of timing research.

Impact and Legacy

Treisman's 1963 monograph is now recognised as a founding document in the modern study of time perception, and his insistence that a timing model must include memory and decision processes, not merely a pacemaker, remains a standing corrective. He published around a hundred and ten papers in total, spanning psychophysics, perception, attention, and timing, and he influenced Oxford experimental psychology through several decades of teaching and supervision.

The criticisms of his work are partly about reception and partly about substance. The first is that his influence is almost entirely retrospective: the 1963 model was ignored for years because the field had little interest in time, and by the time interest arrived, competing frameworks had been established. Scalar expectancy theory, developed by John Gibbon and colleagues and elaborated with Russell Church and Warren Meck, became the dominant account of animal and human timing. Its central result — that the variability of timed responses scales with the duration being timed, so that timing obeys a scalar property — gave it a simple, testable signature and a huge comparative literature, and Treisman's model, for all its sophistication, never became the standard reference point.

A related criticism is testability. The algebraic elaboration that made his model precise also made it hard to evaluate; a system with that many interacting components and free parameters can accommodate a wide range of data, and it is not clear that the specific architecture he proposed has been confirmed rather than merely rendered consistent with later findings. Simplicity is a scientific virtue, and scalar expectancy theory had it in a form his model did not. More recent work has also moved away from the idea of a single central clock toward distributed and state-dependent accounts, which sit awkwardly with a pacemaker-and-calibration-unit design: on those views there is no single oscillator whose period could be revealed by interference, because timing emerges from the dynamics of the circuits doing the task.

His second research programme has fared less well than his first. The attempt to identify a specific neural oscillator frequency through interference from repetitive stimulation has not produced a widely accepted result; estimates of the pacemaker period have varied across studies and across laboratories, and the interpretation of the interference effect remains contested.

There is finally a matter of visibility. Treisman spent his career at Oxford as a Reader rather than in a large, high-profile department, and his independent body of work has been partly eclipsed by the fame of Anne Treisman, whom he married in 1960, and of Daniel Kahneman, whom she married after their divorce. He is frequently mentioned chiefly as Anne Treisman's first husband, a framing that understates a substantial and original contribution to experimental psychology — and one that is compounded by the accidents of his reception: a defining paper published before its field existed, and a rival framework that arrived while he was working on something else.

Learn More

Anne Treisman — his colleague and wife from 1960 to 1976, whose feature integration theory reshaped the study of visual attention.
Donald Broadbent — whose filter model of selective attention dominated the Oxford period in which Treisman did his early psychophysics.
Daniel Kahneman — who married Anne Treisman in 1978, and whose work on attention and judgement ran alongside the Oxford tradition Treisman worked in.

Sources: The Guardian obituary (21 April 2023); J. H. Wearden (2023), “Michel Treisman (1929–2023): A Short Appreciation of his Contributions to the Study of Time Perception,” Timing, 12(1).

References

Oswald, I., Taylor, A. M., & Treisman, M. (1960). Discriminative responses to stimulation during human sleep. Brain, 83(3), 440-453.
Treisman, M. (1963). Temporal discrimination and the indifference interval: Implications for a model of the “internal clock”. Psychological Monographs: General and Applied, 77(13), 1-31.
Gibbon, J. (1977). Scalar expectancy theory and Weber’s law in animal timing. Psychological Review, 84(4), 279-325.
Treisman, M., Faulkner, A., Naish, P. L. N., & Brogan, D. (1990). The internal clock: Evidence for a temporal oscillator underlying time perception with some estimates of its characteristic period. Quarterly Journal of Experimental Psychology, 42A(1), 1-32.
Wearden, J. H. (2023). Michel Treisman (1929–2023): A short appreciation of his contributions to the study of time perception. Timing, 12(1).