Learning and attention: brain and behavioural approaches: the 1995 Banff Annual Seminar in Cognitive Science

Canadian Psychology, Feb 1996 by Shapiro, Kimron, Dixon, Peter

The Banff Annual Seminar in Cognitive Science (BASICS) was founded in 1982, and thus the meeting this past May marks the 14th anniversary of this conference. BASICS was originally intended to be and still is a broad forum for the discussion of all matters pertaining to cognitive science. In the 14 years since BASICS began, over 80 one - hour seminar - type talks have been presented on topics encompassing a wide range of issues regarding perceptual and cognitive processes. Many recent talks have stressed the importance of studying these processes in conjunction with each other. In keeping with this theme, the focus of BASICS 1995 was on recent neuropsychological, neurophysiological, and behavioural findings with respect to attention and learning. Further, many of the talks provided evidence in support of attentional organization centered around objects, rather than locations.

The first talk, from world - renowned scientist Michael Posner, centered on his work involving neuroscientific approaches to the study of high - level skills such as reading. Posner and his colleagues correlate findings from both the measurement of cerebral blood flow and scalp electrical activity to understand the temporal and anatomical basis of visual word reading. Their method involves the subtraction of the activation found during reading single words from activation arising when a subject is requested to generate a use for the same word. Posner's group has also found that brief practice with these same words alters the pattern of activation, from which he concludes that automatization plays an important role, even in simple mental tasks. Michael Posner's talk was highly thought - provoking and provided those in attendance with a glimpse into the ability to understand the acquisition and change associated with high - level cognitive activities.

Steven Luck's talk also emphasized the use of multiple neurophysiological techniques to study high - level cognitive function, in this case, the binding of features. The so - called "binding problem" has occupied the interests of scientists employing more traditional behavioural techniques for a considerable time. The problem, as was described by Luck, is that, when observing multi - element displays, humans must associate the various features of the elements with the particular element with which they actually belong. From a neurophysiological perspective, this becomes potentially problematic, as the neurons whose activity is associated with these objects have large receptive fields encompassing multiple elements. Luck pointed to various studies in both monkeys and humans, suggesting an important role of attention in solving the binding problem. He argued that attention acts to restrict processing to a single element, and at the same time, suppresses information from competing elements. Luck and his colleagues study this phenomenon by employing both single - unit recording techniques in macaque monkeys and event - related potential techniques in humans. Luck's talk was one of a number of talks of this year's BASICS which provided evidence for object - based theories of attention.

The talk by Lynn Robertson about her work in collaboration with Anne Treisman, switched to a focus on neuropsychological, rather than neurophysiological, findings regarding attention in humans. Robertson has examined patients with various forms of visual neglect, and has found evidence which she takes as support for Treisman's feature integration theory (FIT). FIT suggests that attention acts to join together features for the purpose of object identification. Robertson finds that, whereas patients with lesions producing unilateral visual neglect exhibit normal feature search, this same population appears to be deficient when it comes to performing conjunction search tasks. Other patient data reveals individuals with different lesion sites showing normal conjunction search ability. Robertson interprets data from such patients to argue for the existence of multiple spatial maps in humans with access to certain spatial information available to conscious perception, but access to other aspects of spatial information unavailable to conscious perception.

The last speaker of the first afternoon, Patrick Cavanagh, gave a talk replete with impressive visual demonstrations regarding his work in the behavioural analysis of spatial visual attention. For a considerable time now, investigators of spatial visual attention have embraced the metaphor that attention acts as a "spotlight" to illuminate spatial areas of perceptual interest. Cavanagh began his talk with the argument that such attentional focusing yields a cost in distorting judgements of objects surrounding the attended area. He then went beyond findings concerning static attention to discuss dynamic, or attentive, tracking of objects by comparing it to well - studied eye - movement mechanisms. In comparison to smooth pursuit eye movements, attentive tracking reveals the properties of 1) creating the impression of motion, 2) being limited in capacity, and 3) showing independence from low - level motion mechanisms. Finally, Cavanagh spent some time comparing attentive tracking to saccadic eye movements.