Camps, J. 2 , Vandekerckhove, W. 1 & Mélan, C. 1
1 Laboratoire CLLE-Travail et cognition, Université Toulouse
2 Laboratoire Octogone Lordat, Université de Toulouse
Dyslexic individuals exhibit poor performance on memory verbal span tasks which have been suggested to result either from impaired processing in the Phonological loop, from poor processing of the central executive system (Sela et al., 2012), a reduction in perceptual processing speed (Stenekken et al., 2011) or working memory storage capacity. Recent evidence indicates that the deficit also involves visual-object and visual-spatial information (Menghini et al., 2011). Hence, as has been shown for normal readers, a number of taskrelated factors modulate working-memory performance, including memory load, phonological similarity, and word-length. The aim of the present study was to explore the effects of 1/word concreteness and 2/ active imagery strategy on working-memory performance in 20 dyslexic children aged between 11 and 13 years, compared to matched normal readers. Participants were asked to successively recall four 6-item lists presented orally, 2 lists were comprised of concrete words and 2 lists of abstract words. Participants first recalled one list of each type (random order) and were thereafter instructed to use a mental imagery strategy during presentation of the two remaining lists (random order). Presentation of each list was followed by an interferent task during 20s to limit mental rehearsal. Dyslexic children had comparable performance than normal readers in all conditions, except for recall of abstract words not supplemented by an imagery strategy. The overall effect of word concreteness was significant and comparable in both groups (recall of two additional words), while the mental imagery strategy significantly improved recall only in dyslexic children, and only for abstract words. Thus, active mental imagery may improve working memory performance in dyslexic children, as has been previously shown for other learning strategies, including cued performance, and strategy instruction (Swanson et al., 2009), and thus develop specific working memory components.