Ockham's Razor

Emeritus Professor of Neuropsychology, John Bradshaw, from Monash University, talks about experiments he undertook when investigating the brain's right and left hemispheres.

Transcript

This transcript was typed from a recording of the program. The ABC cannot guarantee its complete accuracy because of the possibility of mishearing and occasional difficulty in identifying speakers.

Robyn Williams: I think you've heard the general put-down for someone in senior management, or in politics, that they are two-car wonders. John Prescott, former Deputy Prime Minister in Britain, was known as Two Jags, quite unfairly, according to his autobiography.

Now I prefer being accused of having brains. Because I have two. But before you genuflect in wonderment, relax. Because you've got two brains as well. Only rarely, very rarely, would you be like the girl born with only half a brain, in other words, just one, who got on reasonably well in life, before they tumbled to the truth, the doctors I mean, that she only had half a brain.

We have two lumps of brain, almost autonomous, linked by a cable. If that cable is cut, they behave independently; as John Bradshaw from Monash University is about to tell us. He's Professor of Neuropsychology there. Well he is now. John.

John Bradshaw: Nearly 40 years ago, with a near-new PhD, I arrived at Monash University, then recently established in a paddock on the edge of Melbourne's suburban spread. Not any more. The paddock appeared on maps as the site of an 'epileptic colony', an unfortunate, and nowadays politically most incorrect, and some said not completely inappropriate term for the new tenants. I was soon asked to prepare a course of lectures on Memory and Learning, about which I knew little. 'Business speak' was then an abomination, way below the temporal horizon, and 'problems' had not yet been euphemised to 'issues' or 'challenges'. Nevertheless, I decided in accordance with that sorry modern cliché to make an opportunity out of a problem, to turn a necessity into a virtue. I would lecture on the Biological Bases of Memory and Learning, an approach with which I felt far more comfortable. Because I soon recognised that in no way could the currently available material fill a standard 13-week lecture series, I expanded the course to cover Biological Bases of Behaviour generally. As no-one seemed to notice, and the students seemed happy and interested, the following year I simply re-badged the lectures as Biological Bases of Behaviour, and these continued with regular updates, for more than two decades.

Research is rightly said to inform teaching; there is nothing more stimulating for a class than to tell them, breathlessly, about the latest hot new finding, especially if you have just made it. However, curiously, less is usually said about the equally important reverse process, teaching informing your research. Here I was in Australia, claiming on apparently good text-book authority, in a lecture on the neural fibres interconnecting the two sides of the brain, that these commissures are far less evident in Australian marsupials than in other mammals.

Maybe a bright young spark asked me about the possible functional consequences of such an Antipodean anomaly, or maybe I even asked myself, I forget! But I do remember wondering, while drying the dishes one morning in our rented house in the country, whether there are consequences fore us humans when a visual stimulus is obliged to cross, from its receiving hemisphere, over the commissures to the other hemisphere where a response must be initiated. Memories of what you were doing just when an important idea, or news item, such as a significant assassination hits one, are nowadays called 'flashbulb memories', but that is definitely another story for a talk on Learning and Memory!

I therefore wondered: would there be a cost in, say, response speed, when pressing a button to a visual stimulus received, by the exigencies of the experiment, on one side of the brain, when now the other side of the brain was obliged to initiate the response. Any such increase in response latency, compared with when both stimulus processing and response initiation could occur within one and the same side, or hemisphere, of the brain, should reflect transmission time across the commissures.

In theory, it was easy to arrange such a comparison. A flash of light received by either eye in the left periphery of the field of view, is initially received in the brain's right visual cortex; a right flash is similarly elaborated in the left half brain. Possible evolutionary reasons for this cross-over, and the exactly-corresponding crossed arrangement in the motor system, driving the hands and arms, are still under debate. In 1969 I set up a situation to compare button-press response times to lights briefly flashed randomly to one or other side of a central fixation point, with responses by the fingers of the hand anatomically on the same or opposite side of the body as the light flash. The latter, crossed, arrangement should be the slower by an amount equal to the time taken for information to cross the commissures, or possibly should reflect the effect of signal degradation over a channel of imperfect fidelity.

But this was in 1969, and I needed a computer to run such an experiment, with near millisecond accuracy, as only a few thousandths of a second might differentiate the stimulus-response configurations. Again I was lucky; a colleague had one of the very first such behavioural-lab computers in Australia, a PDP-8S. He agreed to collaborate with me on the project, and we got a reaction-time difference of 20 milliseconds, almost an order of magnitude greater than what might be predicted by a knowledge of the neural pathways involved. In a way, this was fortunate, as it stimulated us to an extensive program of research on brain-behaviour relationships that lasted for nearly 20 years; without such a stimulus I doubt whether we would have embarked upon such an ambitious and ultimately profitable program. It was however, also unfortunate, in a narrower sense, as many years later, after much further research in many centres around the world, it became evident that the smaller figure of a few milliseconds was closer to the mark; our values, and those of others who subsequently varied our paradigm, were partly an artefact of additional attentional processes, of which none of us were really aware.

Shortly afterwards, my colleague in this initial enterprise died under unusual and tragic circumstances, and the research grant which we had now won on the basis of early success, along with the all-important computer, became mine to do with as I liked. Indeed, many years later Digital-Decus, as I think the firm was called, bought it from me for their museum of iconic, original IT equipment. We got much more than the original purchase price, and moved the lab rapidly into the computer age proper.

It was in the early period of my work at Monash that studies became widely disseminated of war-wounded veterans, always a sad source of neurological information on the basis of discrete deficits, subsequent to localised penetrating brain lesions. It was clear that while certain language processes, at least in right-handed individuals, are largely elaborated in the left half brain, face processing may similarly be a province generally of the right. Could we demonstrate this behaviourally with healthy, intact individuals, using a similar subtractive reaction-time technique to laterally-presented face stimuli? These would have to be matched to other, memorised target faces which varied in specific ways from the test items. Could we similarly flash up numbers, letters, words, shapes, patterns, drawings, all to one or other visual field, and record our participants' discriminatory reaction times? Could we in this way get a handle on where and how in the normal brain, all such information-processing systems operate?

At this point I was approached by a graduate student, who had previously studied faces in memory experiments. We needed stimuli with ranges of features over which we had firm experimental control, as in the forensic IdentiKit system used by police for identifying possible felons by witnesses. It consisted of a huge library of varying feature overlays. We wrote to the distributors in California and hired the system. Like all equipment coming at that time from overseas, it had to pass through the hands of our University Customs Officer, a wily, knowledgeable and accommodating gentleman with a finger in every pie, and believed by some to be able to supply anyone with anything they might at least half reasonably ask for. At the time a flasher was upsetting female students, and it being known that we had just taken delivery of an IdentiKit, we were asked to provide professional identificatory help. However the victims were uniformly unable to say anything useful about the perpetrator's facial features, one even suggesting we should develop a system based on quite different principles.

Strangely, most of the faces we constructed from the system did possess a rather sinister aspect, maybe somehow reflecting the original source material. However, we showed that a right-hemisphere (right visual field) superiority for faces - and for certain displays constructed from elements constituting other meaningful objects, like bugs and cottage fronts - only occurred with easy, rapid comparisons where all features or elements different between any two items. With confusingly similar configurations which differed minimally from each other, like identical twins faces, the opposite, or language side of the brain, proved pre-eminent. Was this why language had apparently evolved in the left hemisphere, because complex, feature-discriminatory processes common to many operations resided there, with simpler, global comparisons operating in the right? Indeed, we later showed that certain aspects of language itself, such as humour and context, were in fact also handled in the 'non-language' hemisphere.

Still unresolved is the question as to whether there is a specialist face processor on the right, perhaps as recent imaging studies suggest, in the fusiform gyrus; alternatively, does that region also handle all complex patterns which are of a generic and significant nature in the individual, like makes of cars and so on, of which faces are just one, if a major, example? It was an exciting time, with split-brain studies now going on in the United States, under the Nobel laureate Roger Sperry, involving patients whose commissures had had to be surgically divided. Also in Italy, a group which we fondly came to refer to as the 'spaghetti mob', was simultaneously publishing almost identical studies and findings to our own. We soon entered into a happy gastronomic collaboration with them.

Brain injury can cause a very specific loss of the ability to recognise people by the face. Some such prosopagnosics, whose work or hobbies have previously made them very skilled at discriminating between particular species of bird, breeds of dog or makes of car, may also lose that capacity, suggesting that the affected region handles more than just face processing. Other researchers, unconvinced, claim there is still something special about faces, maybe involving their function in communicating emotions, as well as personal identity.

We have long known that left-side brain damage can seriously affect our capacity to speak, understand speech, read or write, and that some people are as if born with a particular deficit in reading. Might such developmental dyslexia, as it is known, possibly have a counterpart in face recognition? I have always myself had problems recognising people, and often have had the feeling that I am again seeing the same person walking past me when the spatial and temporal aspects of the situation make it very unlikely. So I long ago wondered about the possibility of a syndrome such as developmental prosopagnosia, by analogy with development dyslexia. Some years ago, I was gratified to read the very first account of a family of just such individuals; it is in fact now known that maybe some 5% of the population may meet those criteria.

Some 150 years ago the cult arose of phrenology, ascribing such attributes as Veneration, Wonder, Wit, Vanity, Cautiousness, Amativeness, Philoprogentiveness, to certain brain regions identifiable on an individual by the presence of well-developed 'bumps', which could be felt on the skull's surface. Language, Memory and other such localisable capacities with which we are still very comfortable, are of course yet with us. So what should we make of modern imaging and lesion studies apparently indicating local regions of specialisation for the elaboration of face recognition, grammar, route-finding, processing tool-related (as opposed to animal-name) words. Is this just a new phrenology? Is the brain a 'Swiss Army Knife' of discretely-functioning modules? Or should we instead invoke the operation of whole circuits or systems encompassing many brain regions, to be co-opted, in different ways and combinations, for quite different functions?

I don't know in truth if the marsupial brain is indeed 'commissurally challenged', but it does illustrate how, just like my first faulty experiment to study the commissures, a possibly unsound premise can still lead to useful and enlightening consequences. And teaching can inform research, not just the other way round!

Robyn Williams: Good to know. Also that we don't have two brains each, but many. John Bradshaw is Professor of Neuropsychology at Monash. Still.

And next week, Stuart Taylor goes to the bottom of the harbour, and I don't mean for a business or a tax dodge.

I'm Robyn Williams.

Guests

Emeritus Professor John Bradshaw
Professor of Neuropsychology
Monash University
Melbourne

Presenter

Robyn Williams

Producer

Brigitte Seega