Kim Uittenhove
Laboratoire de Psychologie Cognitive
CNRS & AMU
The dynamics of active information maintenance
One of the cornerstones of human cognition is our ability to actively maintain information over time (i.e. working memory). Severely limited, this capacity depends on the interplay between the degradation and maintenance of representations held in working memory over time. Formal models have been proposed that characterize this interplay (e.g., Time-based Resource Sharing model, Barrouillet, Bernardin, & Camos, 2004; Serial-Order-in-a-Box-Complex-Span, Oberauer et al., 2012). However, these modeling attempts have been met with some surprising results. For example, in our experiments, the dynamic between loss and maintenance seems to depend on the type of information that is maintained (i.e., verbal or visuospatial), something that is not currently taken into account by any model. This surprising finding may be explained by revising the implicit assumptions underlying working-memory research for the past 40 years, assumptions that were derived from the multi-component model of working memory (Baddeley & Hitch, 1974). This model evokes specific yet equivalent storage systems for verbal and visual information. However, an increasing body of results are difficult to accommodate by this view (see for example Postle, 2006), calling for a fundamental revision. I propose an alternative framework that takes into account different stimulus dimensions (i.e., scale, type and richness of information), which may be important in determining to what extent maintenance is afforded. Several of our experiments exemplify how this perspective may be useful for better understanding the dynamics between loss and maintenance of working-memory representations. Moreover, this alternative framework may advance the understanding of other topics as well. In this vein, I will discuss how it may be used to examine the development of working memory as well as its evolution through aging.
