The Continuous Architecture of Working Memory: Evidence from Learning Disabilities and Genetic Syndromes.

Since the theoretical account of a multi-componential short term memory responsible for the temporary storage and/or processing of information (i.e., Working Memory, WM) has been proposed by Baddeley and Hitch (1974), a wide range of experimental studies has been investigating its validity. Overall, research on the effect of individual differences in WM functions supports the hypothesis that temporary mnestic system is composed of functionally independent elements, which can be distinguished in terms of the nature of the stimuli (e.g., verbal, visual, spatial) to be manipulated and of the function accomplished within the system (i.e., maintenance or control function). However, a series of studies has investigated the features of the functions underlying the input processing to elucidate the architecture and functioning of WM. Thus, it has been proposed a distinction between low and high control WM tasks referred to different operations using an increasing amount of attentional resources respectively to temporarily maintain or transform information. In this perspective, the Continuity Model proposed by Cornoldi and Vecchi (2003) is conceived to better specify the multi-componential model by Baddeley and Hitch, emphasising the role played by the central control processes in carrying out memory tasks. In Cornoldi and Vecchi’s view, control may range from a simple rehearsal activity to more complex order change, selection, inhibition, and transformation tasks up to tasks requiring dual WM. Cornoldi and Vecchi argue that low control tasks, traditionally defined as short-term memory tasks, simply require the subject to remember the material as it was presented, thus involving very low control; whereas high control tasks require very high control and do not specifically involve either verbal or visuospatial tasks. The main advantage of using Cornoldi and Vecchi’s model is the possibility to make specific hypotheses about the WM difficulties of particular categories of individuals such as learning-disabled children. According to Cornoldi, Carretti, and De Beni (2001), highly intelligent children with specific learning disabilities (e.g., dyslexic) should fail mainly in low control tasks, whereas children with reading comprehension and/or problem-solving difficulties may have problems at higher levels of control (e.g., at an intermediate level of control, when selection and inhibition of specific verbal information is required). In contrast, children with mental retardation may fail at an even higher level of control. In the current chapter the most recent experimental evidence coming from research on learning disabilities and genetic syndromes which supports the continuous architecture of WM will be reviewed.