Michael I. Posner

Introduction

Michael I. Posner (born September 12, 1936) is an American psychologist who did more than anyone else to make attention a respectable object of study, and then to make it a subject with an anatomy. He began as a mental chronometrist measuring milliseconds, and he ended as one of the founders of cognitive neuroscience, having taken positron emission tomography into the study of normal human cognition while most of the field still thought of it as a clinical tool.

His name is attached to a paradigm — the Posner cueing task, in which a cue draws attention to a location before a target appears — and to a framework: the claim that attention is not one thing but three, implemented by separable brain networks for alerting, orienting and executive control. The paradigm is used in thousands of laboratories. The framework is the standard vocabulary of the field, and it is also contested in ways that matter.

Early Life and Education

Posner took a bachelor's degree in physics at the University of Washington in 1957, then moved to the University of Michigan for graduate work in psychology, taking an MS in 1959 and a PhD in 1962. The physics background shows in the work: he has always preferred designs in which a quantity can be isolated by subtraction or added by design, and in which the result is a number in milliseconds rather than a description.

Academic and Professional Career

He began as an assistant professor at the University of Wisconsin in 1962 and moved to the University of Oregon in 1968, where he stayed until retiring in 2000 and where he remains professor emeritus. At Oregon he built the programme that produced much of the modern literature on spatial attention and on the development of attentional control. He also holds an adjunct appointment in psychiatry at Weill Cornell Medical College in New York.

His awards include the National Medal of Science (2008), the Grawemeyer Award in Psychology (2001, shared with Marcus Raichle and Steven Petersen), the John J. Carty Award for the Advancement of Science (2012) and election to the National Academy of Sciences (1981, per the Academy's own member record). He is unusual in having been honoured both by psychology, for the experimental work, and by science generally, for the imaging work that came after.

Further details of his career are available from his University of Oregon page and his Weill Cornell profile.

Major Contributions to Psychology

Mental chronometry

Posner's first research programme used reaction time to take apart mental operations. The logic, inherited from Donders and sharpened by Sternberg's additive factors method, is that if two variables affect different stages of processing, their effects add; if they affect the same stage, they interact. Posner applied it to questions about memory scanning, letter matching and the time course of visual coding, and summarised it in Chronometric Explorations of Mind (1978). It established the standing assumption of his later work: that attention is something you can time.

The cueing paradigm and covert orienting

In 1980 Posner published "Orienting of Attention" in the Quarterly Journal of Experimental Psychology, introducing the design that bears his name. A cue — a brightening box, or a central arrow — directs attention to one of two locations; a target then appears, usually at the cued location but sometimes not. Two results follow, and they are different things: valid cues make responses faster (benefits), and invalid cues make them slower (costs). The existence of both is taken as evidence that attention has actually moved, because a change in decision criterion would produce one but not the other.

Crucially, the effect occurs without eye movement, which is why the phrase covert orienting entered the literature: attention can be directed somewhere the eyes are not looking. Later work distinguished voluntary, or endogenous, cueing from reflexive, exogenous cueing by the time course of each, and the dissociation became one of the standard arguments that attention is not a single mechanism.

The three attention networks

With Steven Petersen, Posner proposed in 1990 that the human attention system divides into three anatomically distinguishable functions:

Alerting — achieving and maintaining a state of readiness, associated with right-hemisphere frontal and parietal regions and with the brain's noradrenergic system.
Orienting — selecting information from sensory input, associated with the superior parietal lobe, the temporal–parietal junction and the superior colliculus.
Executive control — resolving conflict among responses, associated with the anterior cingulate and lateral prefrontal cortex.

The framework's power was that it turned a word with a dozen meanings into three measurable quantities. It also made neuroimaging tractable, because each network had a predicted anatomy that could be tested.

Neuroimaging with Marcus Raichle

Working with Marcus Raichle at Washington University, Posner used PET to image normal volunteers performing word-generation and attention tasks, culminating in Images of Mind (1994). The collaboration was one of the earliest demonstrations that cognitive questions — not just clinical ones — could be asked with functional imaging, and it set the template for the field that followed.

Development and the Attention Network Test

With Mary K. Rothbart he extended the framework to infancy and childhood, arguing that alerting and orienting emerge early while executive control develops slowly across the school years, and that individual differences in the efficiency of executive control predict later academic outcomes. The work is summarised in Educating the Human Brain (2007).

That programme also took a genetic turn. Posner and colleagues reported associations between the efficiency of the executive network and specific gene variants — including the CHRNA4 nicotinic receptor gene and the dopamine receptor gene DRD4 — and, more interestingly, interactions between genotype and environment, with the effects of parenting quality on attentional control differing by genotype. The molecular genetic findings of that era have not aged well; candidate-gene interaction effects of this kind have largely failed to replicate in the large samples that later became available. The more durable part of the programme is the claim that executive attention is trainable and that training effects can be measured behaviourally, which has generated a substantial and still-contested intervention literature.

To measure all three functions in one session, Fan, McCandliss, Sommer, Raz and Posner (2002) introduced the Attention Network Test, which combines the cueing manipulation with a flanker task so that alerting, orienting and conflict can each be estimated from reaction-time differences within a single run.

Major Publications

Posner, M. I. (1978). Chronometric Explorations of Mind. Erlbaum.
Posner, M. I. (1980). Orienting of attention. Quarterly Journal of Experimental Psychology, 32(1), 3–25.
Posner, M. I., & Petersen, S. E. (1990). The attention system of the human brain. Annual Review of Neuroscience, 13, 25–42.
Posner, M. I., & Raichle, M. E. (1994). Images of Mind. Scientific American Library.
Fan, J., McCandliss, B. D., Sommer, T., Raz, A., & Posner, M. I. (2002). Testing the efficiency and independence of attentional networks. Journal of Cognitive Neuroscience, 14(3), 340–347.
Posner, M. I., & Rothbart, M. K. (2007). Educating the Human Brain. American Psychological Association.

Impact and Legacy

Posner supplied the vocabulary and the instruments for a field. "Alerting, orienting, executive control" and the cueing paradigm are now standard equipment in attention research, developmental psychology, clinical neuropsychology and cognitive neuroscience, and the framework's insistence that attention is plural rather than singular is one of the most consequential claims of late twentieth-century cognitive psychology.

Three objections are worth knowing, because they are about the framework rather than about the data, and none of them has been settled.

The Attention Network Test has poor psychometrics. Because each network score is a difference between two conditions, the reliability of the score depends on the reliability of both conditions, and difference scores are notoriously unstable. The psychometric review by MacLeod and colleagues (2010) found the reliability of the ANT's component scores to be low enough to limit their use for individual assessment, and questioned whether the three scores are statistically independent in the way the subtraction logic requires. Subsequent work has shown that the scores intercorrelate, which is exactly what the independence assumption says should not happen. The ANT works well for group means and poorly for individuals.

The framework is a taxonomy, not a theory, and one of its three boxes is too big. Corbetta and Shulman's influential 2002 review divided what Posner calls orienting into two partly dissociable systems: a dorsal, bilateral, top-down system for voluntary orienting and a ventral, right-lateralised system for stimulus-driven reorienting. On that reading, "orienting" is not one network with one job. The executive control box has been divided similarly: the anterior cingulate that Posner and Petersen identified with conflict resolution is now generally placed in a broader cingulo-opercular or salience network, and the dominant account of what it does — Botvinick and colleagues' conflict-monitoring theory — is a competitor to, not a consequence of, the Posner framework.

The imaging method inherited an old assumption. The PET work with Raichle relied on cognitive subtraction: image the brain doing a task, image it doing a simpler version, and attribute the difference to the added component. That requires what is called pure insertion — the assumption that adding a mental operation adds a region without changing what the rest of the system is doing. It is the same assumption Donders made for reaction time in 1868, and it has been criticised on the same grounds ever since. Posner and Raichle were careful, and their results have largely held up, but the method's vulnerability is structural rather than accidental.

None of this diminishes the achievement. A taxonomy that forty years of research has complicated rather than replaced is a considerable thing.

Learn More

Franciscus Donders — mental chronometry and the subtraction method Posner's early work depended on
John Ridley Stroop — the interference task that supplies the ANT's measure of executive control
Charles W. Eriksen — the flanker task and the American tradition of attention research
Anne Treisman — feature integration theory and the study of visual selective attention

References

Posner, M. I. (1980). Orienting of attention. Quarterly Journal of Experimental Psychology, 32(1), 3-25.