Research and Methodology
Professor James T. Townsend has worked for many years within the broad information processing approach. He has been particularly involved in the development of theory-driven methodologies which are capable of solving deep issues of elementary cognition, perception and action, by testing very large classes of models against one another in a non-parametric and distribution free manner.
One train of that research is known as Systems Factorial Technology and another is General Recognition Theory. Both of these will appear again below.
Within the domain of Systems Factorial Technology, he has carried out theoretical and experimental research on the parallel vs serial processing issue and on letter, elementary pattern perception, face processing, decision making, and more recently problems involving speech and hearing.
The pattern recognition work established a firm empirical foundation for certain models that now form a basis for work in identification and categorization. It also has aided understanding of how feature perception occurs. Later investigations include development of the General Recognition Theory in collaboration with Dr. F. Gregory Ashby. Recent efforts have centered on developing a rigorous framework for the investigation of perceptual dependencies in multi-dimensional stimuli such as letters, words, and faces.
His earlier work on parallel and serial processing indicated broad areas where model mimicking occurred. Later efforts have discovered a number of powerful experimental strategies, backed up by mathematical results that can be used to experimentally distinguish the two modes of processing. Similar projects have involved related topics such as identifying decisional stopping rules, measuring workload capacity in attentional and other cognitive settings, and assessing dependencies among psychological subsystems.
This branch of work led into collaborative efforts with Dr. Richard Schweickert of Purdue University on his theory of more complex mental networks as well as providing axioms for the use of the principle of selective influence. Another important collaborator has been Dr. Hans Colonius on topics such as theory and methodology of redundant signals experiments and variance of parallel processing systems as workload is varied.
Another line of research has involved collaboration with Dr. Jerome Busemeyer on a theory of dynamic decision making known as decision field theory. This theory attempts to provide a quantitative setting for explanation and prediction that, in contrast to traditional approaches based on utility theory, is inherently psychological and motivational, dynamic, and stochastic.
There have also been valuable interactions with clinical scientists such as Dr. Richard McFall on philosophy of science issues in clinical science. He values a long and highly fruitful collaboration with clinical and mathematical psychologist Dr. Richard W. J. Neufeld.
