Functional brain organization of working memory in adolescents varies in relation to family income and academic achievement

Abstract

Working memory (WM) capacity reflects executive functions associated with performance on a wide range of cognitive tasks and education outcomes, including mathematics achievement, and is associated with dorsolateral prefrontal and parietal cortices. Here we asked if family income is associated with variation in the functional brain organization of WM capacity among adolescents, and whether that variation is associated with performance on a statewide test of academic achievement in mathematics. Participants were classified into higher-income and lower-income groups based on family income, and performed a WM task with a parametric manipulation of WM load (N-back task) during functional magnetic resonance imaging (fMRI). Behaviorally, the higher-income group had greater WM capacity and higher mathematics achievement scores. Neurally, the higher-income group showed greater activation as a function of WM load in bilateral prefrontal, parietal, and other regions, although the lower-income group exhibited greater activation at the lowest load. Both groups exhibited positive correlations between parietal activations and mathematics achievement scores, but only the higher-income group exhibited a positive correlation between prefrontal activations and mathematics scores. Most of these findings were maintained when higher- and lower-income groups were matched on WM task performance or nonverbal IQ. Findings indicate that the functional neural architecture of WM varies with family income and is associated with education measures of mathematics achievement.

Socioeconomic Inequality and the Developing Brain: Spotlight on Language and Executive Function

“…In another study, lower family income tended to be associated with reduced PFC activation as a function of higher working memory load, as well as with reduced accuracy (though lower family income was associated with greater PFC activation at lower working memory loads; Finn et al., ). These differences in brain function explained differences in mathematics achievement (Finn et al., ). Taken together, this work could suggest that children from higher and lower SES backgrounds rely on different patterns of neural activation to perform EF tasks (Luna, Padmanabhan, & O'Hearn, ).…”

“…For example, youth from disadvantaged families performed less successfully on EF tasks but showed greater recruitment of PFC regions than youth from more advantaged families (Sheridan, Peverill, Finn, & McLaughlin, 2017;Sheridan, Sarsour, Jutte, D'Esposito, & Boyce, 2012;Spielberg et al, 2015). In another study, lower family income tended to be associated with reduced PFC activation as a function of higher working memory load, as well as with reduced accuracy (though lower family income was associated with greater PFC activation at lower working memory loads; Finn et al, 2017). These differences in brain function explained differences in mathematics achievement .…”

The Neuroscience of Socioeconomic Status: Correlates, Causes, and Consequences

“…This combination of both positive and negative relations between SES and brain activity suggests the possibility of different ways of performing the task across levels of SES. A working memory study by Finn et al (2016) showed statistically significant moderation of task-activity relations by SES. As in the study by Sheridan et al 2012, the higher-SES children performed better than their lower-SES counterparts.…”

The relationship between socioeconomic status and white matter microstructure in pre‐reading children: A longitudinal investigation

“…Indeed, the association between SES and brain structure and function has been demonstrated in multiple ways (e.g., Betancourt et al, ; D'Angiulli, Herdman, Stapells, & Hertzman, ; Finn et al, ; Hackman & Farah, ; Hanson et al, ; Jednoróg et al, ; Leonard, Mackey, Finn, & Gabrieli, ; Luby et al, ; Mackey et al, ; Noble et al, ; Noble, Houston, Kan, & Sowell, ; but see Brain Development Cooperative Group, ; Eckert, Lombardino, & Leonard, ; Lange, Froimowitz, Bigler, Lainhart, & Brain Development Cooperative Group, ; Raizada, Richards, Meltzoff, & Kuhl, ). Specifically, lower SES has been linked to reduced gray matter volume (Hanson et al, , ; Jednoróg et al, ; Luby et al, ), reduced cortical thickness (Mackey et al, ), reduced degree of cortical gyrification (Jednoróg et al, ), and reduced surface area (Natalie & Noble, ) in bilateral occipito‐temporal, temporo‐parietal, and inferior frontal regions that support reading development (Booth et al, ; Martin, Schurz, Kronbichler, & Richlan, ; for a review see Ozernov‐Palchik & Gaab, ).…”