ABSTRACT
The main purpose of this chapter, in line with the other papers presented in this book, is to address the question of the development and the role of working memory capacity by relying on neo-Piagetian theories, and more particularly on Pascual-Leone’s (1969, 1970, 1987) model, on the one hand, and on experimental cognitive psychology (e.g., Cowan’s model, 1995), on the other hand. Neo-Piagetian and experimental approaches seem to be very different at fi rst sight. The theoretical conception of working memory is a priori not the same, and the tasks developed were specifi c to each approach. Nevertheless, the fi ndings obtained were highly convergent. Moreover, as we will develop below (see also de Ribaupierre, 2000; de Ribaupierre & Bailleux, 1994, 2000), in the last two decades strikingly similar hypotheses were advanced to explain age and individual differences in working memory capacity, not only in developmental psychology, but also across different fi elds, in particular in cognitive psychology and in neuropsychology. Four studies will be presented as an empirical illustration of the convergence between a neo-Piagetian and an experimental approach as concerns the processes underlying cognitive development. In the fi rst and second experiments, Working Memory span tasks, Processing Speed tasks, and Inhibition tasks were administered to children, young adults, and older adults. The main objectives were to assess the role of underlying processes of age differences in working memory capacity tasks, by using structural equation modeling, and to determine whether similar processes are at work across the lifespan by comparing results obtained with children, young adults, and older adults. In the third experiment, two neo-Piagetian tasks of attentional capacity, and three Piagetian tasks were administered to children. In the fourth experiment, the Raven’s Progressive matrices (Raven, 1938) and Working Memory span tasks, together with other tasks, were administered. The goal of these two latter studies was to determine to what extent agerelated variance in the Piagetian tasks or in the Raven’s Progressive matrices can be accounted for by age differences in the neo-Piagetian or in working memory span tasks. It was also to verify that similar results could be obtained when administering tasks originally developed within a neo-Piagetian
perspective or an experimental one. Of course, we believe that these tasks are very similar and should yield comparable results, as long as they are geared towards assessing attentional resources. However, it is often believed, in experimental cognitive psychology, that different tasks necessarily tap different processes or reciprocally that a same task invariably assesses the same processes, regardless of age or of familiarity, refl ecting a confounding between task and processes (e.g., de Ribaupierre & Pascual-Leone, 1984). Several defi nitions of working memory are available in the literature. There is, however, a consensus that working memory refers to a system or a set of processes responsible for the simultaneous storage and manipulation of information during cognitive tasks (Baddeley, 1986; Richardson, 1996), for temporary purposes. Most importantly, it is assumed that working memory capacity is limited. In other words, only a limited amount of information can be attentionally, or effortfully, simultaneously activated by attentional resources and this amount of information increases with age or varies across individuals (Cowan, 1995; Engle, Kane, & Tuholski, 1999; Engle, Tuholski, Laughlin, & Conway, 1999; Pascual-Leone, 1970, 1987). In the present framework, and following Pascual-Leone’s or Cowan’s propositions, we assume that the content of working memory consists of both activated long-term memory traces and representation of current information that requires attention, and serves to hold and attentionally process information relevant for the task being presently processed (e.g., de Ribaupierre, 2000; Engle et al., 1999). Moreover, we prefer to think in terms of working memory tasks rather than working memory per se, meaning thereby that there is no dedicated system to working memory; performance in such tasks depends on more general resources (such as attentional activation and inhibition) that are also at work in other cognitive tasks. Neo-Piagetians have claimed for a long time (e.g., Case, 1985; PascualLeone, 1970, 1987) that the limits in attentional capacity and their gradual lifting with age under the effect of maturation and experience play an important role in cognitive development, by setting a ceiling on performance beyond which children of a certain level cannot perform, unless the task is facilitated (see also Chapman, 1990; Dasen & de Ribaupierre, 1987; de Ribaupierre, 2007; de Ribaupierre & Lecerf, 2006). By conducting task analyses, they were able to predict which cognitive tasks could be solved, given a certain level in attentional capacity. During the last years, adult developmentalists or cognitive aging theorists, and experimentalists have also stressed the central role played by working memory capacity in individual and developmental differences in higher cognitive abilities, such as language comprehension (Daneman & Carpenter, 1980), reasoning (Süß, Oberauer, Wittman, Wilhelm, & Schulze, 2002), or general ability (Engle et al., 1999). While developmental psychologists were more concerned with changes in working memory capacity with age, experimental psychologists were more interested in analyzing the processes that determine performance in the working memory span tasks themselves. Similar hypotheses were nevertheless formulated by both perspectives.
