There may be nothing special about the association between working memory capacity and fluid intelligence

a r t i c l e i n f o The association between working memory capacity (WMC) and fluid intelligence (g f) has been described as substantial and important. In a recent investigation, Gignac (2014a) contended that WMC and g f share closer to 60% of their variance, rather than the commonly cited 50%, based on an analysis of the Wechsler Adult Intelligence Scale—IV (Wechsler, 2008) normative sample (N = 2200). However, Gignac's (2014a) investigation was limited in that it included only completely homogeneous g f (spatial) and WMC (verbal) subtests, as well as only adults in the sample. Consequently, the purpose of this investigation was to replicate and extend Gignac (2014a) by estimating the association between WMC and g f , based on the Wechsler Intelligence Scale for Children—Fifth Edition (Wechsler, 2014) normative sample (N = 2200), which includes a mix of verbal and spatial WMC subtests. Based on a correlated two-factor model, the correlation between WMC and g f was estimated at .77 (r 2 = .59) which is a perfect replication of Gignac (2014a). However, based on a higher-order model which included all 18 of the WISC-V's subtests, the association between WMC and g f was found to be non-significant (−.10, p = .152) after controlling for the effects of general intelligence. Consequently, the commonly suggested notion that WMC and g f share unique cognitive and or neural processes was not considered supported, based on the results of this investigation. The nature and magnitude of the association between working memory capacity (WMC) and fluid intelligence (g f) has been the subject of a substantial amount of empirical and theoretical research (Conway & Kovacks, 2013). The nature of the association between WMC and g f has been described as fundamental, as WMC is considered a rate limiting factor in the performance of g an empirical perspective, it is commonly stated that approximately 50% of the true score variance between WMC and g f is shared, as Kane, Hambrick, and Conway (2005) reported a meta-analytically derived latent variable correlation of .72 (r 2 = .52), based on 14 samples (total N = 3168). In a recent investigation, Gignac (2014a) suggested that the commonly cited 50% shared variance reported by Kane et al. may be smaller than what would be expected at the adult population level, as the vast majority of the samples included in the Kane et al. …