Does a Brief Executive Function Training Intervention Help to Narrow the Attainment Gap?

• In Event: 2-215 - Training of executive functions in childhood: Moderators of training gains and individual differences in outcomes

Integrative Statement

There is a well-established connection between socio-economic status (SES) and children’s academic attainment, particularly in mathematics (Starkey & Klein, 2008). The gap in mathematical skills between children from higher SES backgrounds and children from lower SES backgrounds starts early and increases over time (Rathbun, West & Hausken, 2004). However, we have only a limited understanding of why this gap exists. The aims of the present study were two-fold: first, to examine whether any effect of SES on mathematical skills was mediated by differences in executive functions (the high-level skills involved in controlling and regulating behaviour); and second, whether a brief executive function training intervention could help narrow the attainment gap in mathematics.
One hundred and seventy-five preschoolers (Mage = 48 months) were recruited from socio-economically diverse neighbourhoods (measured using a neighbourhood deprivation index). Around half the children were from areas among the 20% most deprived areas in the U.K, and around a third were from areas among the 20% least deprived areas of the UK. Children completed baseline measures of inhibitory control, short-term memory, working memory, and mathematics, and were then randomly allocated to an intervention group (N =87) or an active control group (N=88). The intervention was a randomised control trial that followed CONSORT guidelines. Children in both groups completed computerised tasks lasting 20 minutes, in four sessions over one month. The training group completed tasks involving working memory and inhibitory control, and the control group completed superficially similar tasks that did not involve executive functions. The baseline assessment was re-administered four times: at one week, three months, six months and one-year post-training, by an experimenter blind to the child’s condition.
SES explained significant variance in both preschoolers’ mathematical skills (R2 = .05, p = .003) and their executive functions (R2 = .06, p = .002); and executive functions explained significant variance in preschoolers’ mathematical skills (R2 = .29, p < .001). Furthermore, executive functions significantly mediated the relation between SES and mathematical skills at baseline when children were aged four. However, the training intervention did not help to narrow SES attainment gaps. The intervention did not lead to any improvements in inhibitory control, short-term memory or mathematical skills post-training, or at any follow-up assessments. There was a small but significant SES-by-condition interaction on working memory one year later (F(1, 142) = 3.81, p = .05, η2p= .03), such that higher SES children in the training group improved their working memory more than lower SES children in the training group, and more than children in the control group.
The results suggest that differences in executive functions may explain the gap we see in early mathematical skills between children from advantaged backgrounds and children from disadvantaged backgrounds. The next step is to identify ways to reduce this gap early in development. We propose that cognitive training directly targeting executive functions is not an effective approach to narrow the attainment gap.

Authors

• Emma Blakey, The University of Sheffield
  Presenting Author

• Danielle Matthews, University of Sheffield
  Non-Presenting Author

• Lucy Cragg, University of Nottingham
  Non-Presenting Author

• Daniel Carroll, The University of Sheffield
  Non-Presenting Author