Double dissociations in developmental disorders? Theoretically misconceived, empirically dubious.

Single dissociations are the bread and butter of neuropsychologists, and double dissociations their chocolate cake! So it is unsurprising that developmental psychologists working on genetic disorders find the concept of double dissociations an attractive one. The following quotations bear eloquent witness to this: “... the study of mental retardation would profit from the application of the framework of cognitive neuropsychology” ... In cognitive neuropsychology, one key question running through the investigator’s mind is «Is this process or mechanism intact or impaired in this person?”. (Baron-Cohen, 1998, p. 335) “...overall, the genetic double dissociation is striking ... The genes of one group of children [SLI] impair their grammar while sparing their intelligence; the genes of another group of children [WS] impair their intelligence while sparing their grammar” (Pinker, 1999, p. 262). We argue that the use of the double dissociation method in developmental disorders is not only inappropriate theoretically, but also erroneous empirically, often based on a dubious choice of control groups. It rests on a false assumption: that the brain of an infant with a genetic disorder comprises a pattern of neatly segregated, “intact/impaired cognitive modules”. In doing so, it neglects one vital factor: the actual process of ontogenetic development (Karmiloff-Smith, 1998). The effects of a genetic mutation during embryogenesis and postnatal brain growth are likely to be widespread across the developing system. Some domains may be more affected than others due to the features of their particular problem space, but in-depth studies reveal subtle impairments in domains that originally seemed “intact” (Karmiloff-Smith, 1998; Karmiloff-Smith et al., 2002). As Dunn and Kirsner (2003, this issue) highlight, the notion of “intact” performance depends on the sensitivity of the measurement scale. In sum, an uneven pattern in the cognitive profile of a genetic disorder cannot be taken for granted to imply impaired versus totally preserved abilities. All domains may be impaired, some more subtly than others. The case is different for adult neuropsychology patients. In their previously normal brains, specialisation and localisation of function had already stabilised, so selective impairments might emerge if pure cases exist. But what about the double dissociation claimed above for Specific Language Impairment (SLI) and Williams syndrome (WS)? Empirical data show that absolute statements about “sparing” of intelligence or grammar should actually