The term developmental dyslexia (DD) is used to characterize a disorder in which the core deficit involves reading and spelling. Several accounts have been proposed for DD, with some postulating an impairment in the phonological system and others in the visual system. In this paper, intelligence data from the WISC-IV were collected from more than 300 Italian children with a diagnosis of DD (Mage = 11.72 years, SD = 2.61 years; age range 7-16 years; 55% males).
To examine the underlying structure of VCI, PRI, WMI, and PSI we used a model-based clustering analysis approach, where clusters are modeled as a finite mixture of Gaussian distributions (Fraley & Raftery, 1998). This analysis was performed using the mclust package in R environment (Scrucca, Fop, Murphy, & Raftery, 2016), which allowed to evaluate different clustering solutions in terms of model parameters.
The EII model with diagonal, equal volume and shape covariance matrix was selected as suggested by the BIC values. This model confirmed the existence of two different clusters of children with DD.
Between-group comparisons at the univariate level revealed that the two groups differed significantly in the IQ, t(314) = 22.31, p < .001, Cohen's d = 2.62, with cluster 1 (M = 113.56, SD = 7.07) having an higher IQ compared to cluster 2 (M = 94.77, SD = 7.23). As expected, a statistically significant difference emerged between the two groups at the multivariate level, Pillai’s trace = .66, F(4,311) = 153.12, p < .001. Between-group comparisons at the univariate level revealed that the two groups differed significantly and with a large effect size in VCI, t(314) = 15.32, p < .001, Cohen’s d = 1.80, PRI, t(314) = 15.96, p < .001, Cohen’s d = 1.87, and WMI, t(314) = 9.97, p < .001, Cohen’s d = 1.17, but not in PSI, t(314) = 1.35, p = .18, Cohen’s d = .16. In particular, in one cluster the deficit was more pronounced in the phonological component, while both clusters were generally impaired in visual processing.
These data were used to support the idea that DD is an umbrella term that encompasses different profiles. From a theoretical perspective, this paper demonstrates that dyslexia cannot be explained by an isolated phonological deficit; visual impairment plays a crucial role. Moreover, general rather than specific accounts of DD are discussed. These results also have practical and clinical implications. In fact, the evaluation of the cognitive profile of children with DD can be beneficial for both the assessment and intervention of these children.
This paper demonstrate that intelligence tests might be useful distinguishing between children with different profiles of dyslexia. The existence of two clusters has important theoretical relevance. Although the two clusters present with polar patterns of impairment severity (cluster 1: vision
