A Hierarchical Watershed Model of Fluid Intelligence in Childhood and Adolescence

Abstract

Fluid intelligence is the capacity to solve novel problems in the absence of task-specific knowledge and is highly predictive of outcomes like educational attainment and psychopathology. Here, we modeled the neurocognitive architecture of fluid intelligence in two cohorts: the Centre for Attention, Leaning and Memory sample (CALM) (N = 551, aged 5-17 years) and the Enhanced Nathan Kline Institute-Rockland Sample (NKI-RS) (N = 335, aged 6-17 years). We used multivariate structural equation modeling to test a preregistered watershed model of fluid intelligence. This model predicts that white matter contributes to intermediate cognitive phenotypes, like working memory and processing speed, which, in turn, contribute to fluid intelligence. We found that this model performed well for both samples and explained large amounts of variance in fluid intelligence (R2CALM = 51.2%, R2NKI-RS = 78.3%). The relationship between cognitive abilities and white matter differed with age, showing a dip in strength around ages 7-12 years. This age effect may reflect a reorganization of the neurocognitive architecture around pre- and early puberty. Overall, these findings highlight that intelligence is part of a complex hierarchical system of partially independent effects.

Keywords: fractional anisotropy; processing speed; structural equation modeling; watershed model; working memory.

Figure 1

The watershed model. Schematic representation of the watershed model developed by Cannon and Keller (2006) and adapted for the present study. Fluid ability is hypothesized to be the downstream product of working memory and processing speed, which are, in turn, the product of white matter contributions. Figure adapted from Kievit et al. (2016).

Figure 2

White matter tracts modeled in the analyses.

Figure 3

Different measurement models of cognition. Abbreviations: WM: working memory, PS: processing speed

Figure 4

Watershed model in CALM. See Supplementary Table 2 for regression estimates. Residual covariances between white matter tracts were allowed but are not shown for simplicity. Abbreviations: uncinate fasciculus: UF, superior longitudinal fasciculus: SLF, inferior fronto-occipital fasciculus: IFOF, anterior thalamic radiations: ATR, cerebrospinal tract: CST, forceps major: FMaj, forceps minor: FMin, dorsal cingulate gyrus: CG, ventral cingulate gyrus: CH, inferior longitudinal fasciculus: ILF.

Figure 5

The watershed model in NKI-RS. See Supplementary Table 3 for regression estimates. Residual covariances between white matter tracts were allowed but are not shown for simplicity.

Figure 6

Configuration of alternative models. Alternatives A and B are watershed compatible, while C and D are watershed incompatible. The best-fitting model for CALM is the original watershed model; the best-fitting model for NKI-RS is Alternative A. Regression paths only are shown for simplicity. Square shapes denote manifest variables and oval shapes latent variables.

Figure 7

Cognitive factor scores by age.