Psychologist logo
BPS updates

One on one... with Anne Treisman

…with Anne Treisman (including online-only answers)

01 November 2010

One inspiration
Richard Gregory was my tutor during my one-year undergraduate degree in psychology. He made tutorials exciting, partly through his own excitement in the ideas. They were all completely new to me (my first degree was in French literature), and he bypassed the then current topics of behaviour theory (which would have left me cold) and introduced the new ideas of information theory, vision as cognition, and the relation of brain to mind. And he made it all wonderful fun.

One moment that changed the course of your career
The decision to move to North America. It put me in touch with a much wider range of contacts and openings, although I still miss the British psychological scene.

One book that you think all psychologists should read
Stephen Pinker’s How the Mind Works.

One impact of my feature integration theory
The binding problem has challenged philosophers, neuroscientists and computer scientists (as well as me), and the empirical evidence I reported contributed relevant findings to the discussion.

One challenge you think psychology faces
Understanding consciousness. Francis Crick believed the problem would melt away as the science advanced and more data were discovered, just as the problem of life did.  But I am not even sure of that.

One regret
Not having any background in neuroscience or mathematics. These would have increased my understanding of psychology. The school system in England forced a choice of just three subjects at age 15 and I chose French, Latin and History. Still, I can’t regret my first degree in French literature because I enjoyed it and was enriched by it.

One memory of Wakefield
None, since I was only two when we left. My memories start with the war years in Kent, the air raids that we spent in the cellar, the barrage balloons at the bottom of the garden, the blackout, and the map of Europe that my father updated with flags to mark the advance of the Allied Forces.

One great thing about being married to a Nobel Prize winner’ (Daniel Kahneman)
The interesting people we meet – not least my husband himself.

One nugget of advice for aspiring psychologists
Enjoy the excitement of finding out about the mind and testing your own hunches within the framework of?current research. But if one idea doesn’t work out, move on to the next rather than throwing good time after bad.

One cultural recommendation
Among movies I loved both Les Enfants du Paradis and Annie Hall.

One alternative career path
I did apply for a job teaching French at a high school, and have been deeply grateful to them for turning me down ever since. If I had to choose again, a good possibility would be neurology, but I am pretty happy to be an experimental psychologist

One thing I’ve learnt from Buddhist monks
A completely different view of the human condition. I am not able to share it, but it has enlarged my notion of human possibilities.


Online extras

One proud moment


The one week when my husband won the Nobel prize and my younger daughter became the fiction editor at the New Yorker.

One hero/heroine from psychology past or present
Donald Broadbent. He inspired my early work on selective listening by setting up a framework in which questions about human attention could be fruitfully asked and answered, and in which their practical implications could be spelled out.

One great thing that psychology has achieved
Ways to link the mind and the brain, not just by finding out where things happen but by illuminating how. This is a quest that is still just beginning.

One thing that you would change about psychology
Nothing about the science of Psychology. I could do without many of the phony applications and popularisations.

One hope for the future
That psychology doesn’t get swallowed up by neuroscience but remains a distinct approach, using all the new methods and data but asking its own questions.

One way our understanding of visual perception and attention has changed since your early days
I have lived through three big transformations in Psychology. When I was a student, we were right on the cusp of the Cognitive Science revolution. In the Cambridge department in my year there were 12 students altogether. By the luck of the draw, most of them had tutorials on Behaviourism. Its explanatory tools were limited to stimuli and responses.


If you wanted to speculate about what intervened between them in the head, you were limited to ‘little sgs and little rgs’, which were the internal equivalents of observable stimuli and responses. I was lucky because I was assigned a tutor, Richard Gregory, who wasn’t interested in that approach He encouraged me to read about information theory – thinking of stimuli not as mechanically evoking responses but as reducing uncertainty, and so interacting with an internal state of prior knowledge and expectancies  – and about vision, really neat illusions and the mechanisms that might explain them.

I went on to work for my PhD on attention, which is the antithesis of the passive switchboard of evoked responses. The model of the mind was changing from the telephone switchboard to the flowchart of processing stages, and more generally to the computer. Cognitive Science was born, and of course it is still flourishing. In the early days of Cognitive Psychology we asked questions about functional mechanisms, independently of their neural embodiment, using behavioral experiments to rule out some possibilities and narrow down progressively onto a model, which was consistent with the data. Psychology came up with black box models like Broadbent’s filter theory, and I think that we learned a considerable amount using only behavioral experiments. In some cases we could test which stages seemed to be independent, and the order in which they acted. We did learn about brain-damaged patients too, and I even did an EEG experiment on discrimination in sleep with Ian Oswald. So I was aware that there was a brain generating the behavior we observed, but the flow charts could also stand on their own as a functional level of description. One of the problems was that ruling out alternatives didn’t prove easy. As experiments and models multiplied, the goal became to find constraints – things that the brain did not seem to do, links between the black boxes that were not present.

The second big revolution, some 30 years later and which is also still well under way, has been Cognitive Neuroscience. Obviously Cognitive Neuroscience has transformed Psychology, but not by making it irrelevant. It has given us new tools and new sources of data, but I think the original goals remain intact. Of course neuroscience does not necessarily solve the problem of finding constraints on possible models.

Another big problem was that it turned out that a lot of mental life is implicit, not consciously accessible, so just because a participant could not report seeing or hearing something did not mean that it wasn’t happening. That made behaviour a less reliable guide to the mind than we had hoped. Here Cognitive Neuroscience has been more helpful, actually even as early as my own EEG experiments with Oswald where we found that the EEG showed arousal signs (K complexes) to the person’s own names without waking him or her up.

But with the advent of fMRI, the scope of these tests has greatly increased. Just to give a couple of examples from the field of attention, which illustrate what I think are clear answers to some problems which had been giving us trouble for many years before. One example was “what does attention select?” Places, objects or features? O’Craven et al. (1999) used overlapping face and place stimuli and asked people to report either the direction of motion of one picture or the location (slightly shifted from center) of the static one. They found that the picture with the relevant attribute also produced increased activation in the brain area specialized for its (irrelevant) category – the Parahippocampal Place area for houses and the Fusiform Face area for faces. So selection of one attribute here (for example motion) resulted in enhanced attention to the object as a whole. By spying on a specialized brain area, we can see things that the subject herself doesn’t. On the other hand, Downing et al., (2001) reported equally strong evidence that selecting the object in one location automatically enhances activation to another object in the same location. The picture that emerges is that attention can be biased to select either objects or places, depending on the requirements of the task.

Another example of a long-lived classic problem in attention research was whether selection occurs early or late in processing, before complete identification or after. This was a controversy that raged for a long time when psychology was restricted to behavioural evidence. Again being able to spy on the brain has given us new insight. Marois, Yi and Chun (2004) used an attentional blink experiment to throw light on the early versus late selection problem. They showed a sequence of scrambled scenes in which they embedded two targets.

The first was a face and the second was a scene. Usually when two targets occur close to each other in time people miss the second. Detecting the first target temporarily overloads attention, causing a so-called attentional blink. The question was what happens in the brain to the blinked second target when people fail to consciously see it?  They found that it still activated the parahippocampal place area, showing that they did register the fact that a scene had appeared. The activity was a bit less than when they did see it.

These are examples where we learn answers to theoretical questions. We don’t just learn where in the brain something happens, but WHAT is the mechanism, what is actually going on. It’s interesting that they depend on our having previously discovered WHERE something happens. The face and place areas have been used in many different clever ways in attention research, beyond the interest of knowing where those particular processes occur. Sceptics sometimes accuse fMRI advocates of just telling us where in the brain something happens, which they say they don’t particularly care about, but these examples and many others clearly also tell us about what is happening.

These various sets of findings have produced fairly profound changes in our understanding of the mind since I was first a student in Psychology. Instead of a sequence of discrete stages that could be directly probed as well as inferred from behaviour, we have a more interactive set of processes, with multiple reentry loops, parallel processes, with some parts of the mind knowing things of which other parts are ignorant. The mind may be more like an organization than like a unified entity. When would we say that the CIA or IBM knows something? When the chief executive knows it? Or when some lowly clerk in an office somewhere knows it? Knowing, or cognition itself, has become a tricky state to define. We need, and we are getting, more ways to probe the complexities and more complex ways to model them. It has been an exciting time to live through!