I Memory representations: From (visual) perception to stored information
THE ORGANIZATION OF VISUOSPATIAL WORKING MEMORY
The organization of visuospatial working memory: Evidence from the study of developmental disorders*
Cesare Cornoldi and Irene C. Mammarella
Visuospatial abilities and their relation to visuospatial working memory
Visuospatial ability is not a unitary process, but instead can be broken down into various distinct types. The differentiation and classification of these types has been influenced by the findings from various instruments chosen to examine visuospatial ability.
Factor-analytic studies of visuospatial ability tasks point to the existence of distinct spatial abilities. For example, some authors (Hegarty & Waller, 2004; McGee, 1979) have distinguished between main aspects, i.e., visualization and orientation. Visualization refers to the ability to mentally rotate and manipulate objects, while orientation refers to the ability to retain spatial orientation with respect to oneself. Linn and Peterson (1985) and Voyer,Voyer, and Bryden (1995) distinguished three categories of spatial ability based on the various different processes required to solve problems representing each ability.These categories were:
- 1 spatial perception (ability to determine spatial relationships with respect to one’s own orientation);
- 2 mental rotation (ability to mentally rotate a two- or three-dimensional figure rapidly and accurately); and
- 3 spatial visualization (ability to manipulate spatially presented information in complex ways).
Examples of tests are: for (1) the water level test (Inhelder & Piaget, 1958); for (2) the Mental Rotation Test (Vandenberg & Kruse, 1978); and for (3) the Differential
Aptitude Test spatial relations subtest. Carpenter and Just (1986) distinguished only two categories of spatial ability:
- 1 spatial orientation (ability to identify spatial configurations from a different perspective);
- 2 spatial manipulation (ability to mentally restructure a two- or three-dimensional object).
Cornoldi and Vecchi (2003, p. 16) instead presented a broader classification, distinguishing between ten different groups of visuospatial abilities, which included visuospatial working memory. Finally, Bunton and Fogarty (2003) examined the relationship between visual imagery and spatial abilities using a confirmatory factor analysis. Their findings supported the notion that the abilities targeted by the tasks referred to above can be classified along a continuum. The self-report imagery questionnaires are located on the left-hand side of the continuum, while experimental tasks examining spatial-imagery and visuospatial memory can be located at the centre. On the right of the continuum they placed the creative imagery tasks of Finke, Pinker, and Farah (1989), and—at the far end—spatial intelligence tests (primary mental abilities,Thurstone &Thurstone, 1965, and Raven’s Advanced Progressive Matrices, Raven, 1965).The main thrust of Bunton and Fogarty (2003) has therefore been to offer a description of the relationships between visual imagery, visuospatial memory, and spatial abilities, also showing their proximity.
In summary, psychometrical research has clearly shown that visuospatial ability is not a homogeneous concept, but consists of subcomponents that are quite distinct, albeit closely related. Nevertheless, despite all the attempts at a suitable classification of the spatial subfactors, the correct operationalization of these factors and their relationships remain unclear. For example, there is evidence to suggest that there is indeed a relationship between working memory capacity and visuospatial ability (fust & Carpenter, 1985). Solving a mental rotation or spatial visualization task requires the ability to maintain an active representation of all the parts and their interrelations, while simultaneously rotating and manipulating the image mentally. This elaboration, involving both storage (holding the constituent parts in memory) and the concurrent processing of spatial representations (the rotation component), fits closely with current conceptions of working memory (Miyake & Shah, 1999). In studies of visuospatial abilities and visuospatial working memory individual differences consequently overlap to some extent.