Last week was a good week for our workload-capacity coefficient. The capacity coefficient is a measure of processing efficiency developed by Jim Townsend and his colleagues at Indiana University (e.g., Townsend & Nozawa, 1995). In the Newcastle Cognition Lab we use it to assess how people are affected by an increase in the amount of information they need to process. We developed a model-based parametric version (Eidels, Donkin, Brown, & Heathcote, 2010) and extended the use of the capacity coefficient to novel tasks.
Two articles from our lab that make use of workload capacity assessment were accepted for publication last week. Interestingly, they tackle the issue of processing efficiency at rather different levels. A paper by Hawkins, Houpt, Eidels, and Townsend in the journal Vision Research tested whether Gestalt advantage in processing can be found with stimuli as simple as a pair of dots. Typically, Gestalt (aka ‘configural’) processing is measured with complex stimuli, such as faces. Ours was a quest for the simplest Gestalt, using simple stimuli in a very simple perceptual task.
Hawkins, R. X. D., Houpt, J. W., Eidels, A., Townsend, J. T. (in press). Can two dots form a Gestalt? Measuring emergent features with the capacity coefficient. Vision Research.
On the other end of the spectrum is another study, by Heathcote et al. (Memory & Cognition). Here, the task was quite complicated, requiring information from both auditory and visual streams as well as use of working memory, and accompanied by an elaborate cover story; in this computer-based game, called ‘Gatekeeper’, the participant is said to be the door person of an exclusive nightclub. Patrons are trying to sneak in and the Gatekeeper’s job is to decide who is allowed in based on auditory and visual cues. Capacity in this case is gauged at a very different level, compared with simple dots.
Heathcote, A., Coleman, J., Eidels, A., Watson, J., Houpt, J. W., & Strayer, D. (in press). Working memory’s workload capacity. Memory & Cognition.
Together, the two papers show how modelling tools can be quite general, used to understand human behaviour at different and rather distinct levels. To read more about these studies and their results, please access copies via Researchgate or via our lab’s website: newcl.org.
You can also request a copy via email from the authors:
Andrew Heathcote: Andrew.Heathcote@utas.edu.au
Ami Eidels: Ami.Eidels@newcastle.edu.au