The School of Psychology at UQ is pleased to present a special neuropsychology seminar
on WEDNESDAY, 26th April, from 2-4pm, in the Psychology Building (24A), Room 202.
Guest speaker Associate Professor Stephen Bowden, School of Behavioural Science,
Department of Psychology, University of Melbourne, will speak on:
How can measurement models illuminate approaches to neuropsychological assessment?
Throughout the history of neuropsychology there has been much debate about appropriate models of brain function and hence, the most appropriate techniques to assess the effects of cognitive disability. This debate has led to a plurality of assessment practices and development of diverse batteries. A common view has been the assumption that a "normal" or unified model of cognition is not adequate to capture the effects of neuropsychological disorders. In extreme form this view has led to a repudiation of "psychometrics" or approaches based on psychological test theory.
However rejection of a common or unified model of cognition produces problems for valid assessment. To understand the practical impact of these issues, it is helpful to consider the representation of test score meaning known as a measurement model.
Recent developments in measurement modelling permit a level of precision in the specification of test score meaning that was not possible a few years ago. An additional benefit of measurement modelling is the opportunity to examine two related hypotheses (i) the relevance or validity of any particular psychological construct or latent variable model in a specific diagnostic group, and (ii) the equivalence of test score meaning across different groups.
These techniques have direct application to evaluation of the neuropsychological controversy regarding the relevance of models of cognition or intelligence. Recent research with a bearing on these issues is reviewed.