Baseline EF Skills are Associated With Children’s Responsiveness to Different Types of Numeracy Training

• In Event: Novel Approaches to Strengthening Young Children's Math and Executive Functioning Skills

Abstract

Executive function (EF) skills (i.e., inhibition, working memory, and cognitive flexibility) support mathematical learning (Cragg & Gilmore, 2014). Relational language (RL) provides a tool for describing comparisons between items (e.g., more, after), and is important for learning about number relations (e.g., 5 is more than 4 and 5 comes after 4). Both sets of skills predict numeracy skills (e.g., McClelland et al., 2014; Purpura & Reid, 2016). We conducted a training study to compare the effects of number training vs. number+EF+RL training on children’s numeracy skills.
We measured 104 kindergartners’ (46 boys; mean age=5.9 years) performance on eight numeracy tasks and two EF tasks at pretest and posttest. Children received four 15-minute training sessions, over 2 weeks, in one of three conditions. In both number and number+EF+RL trainings, children practiced counting forward and ordering numbers. In the number+EF+RL training, children also reflected on their responses, used counting backwards as an alternative problem-solving strategy, and received relational language instruction on cardinal (e.g., 5 is more than 4) and ordinal (e.g., 5 comes after 4) number relations. In the alphabet training (baseline condition), children practiced reciting and ordering alphabetic letters. We excluded 24 children who were at ceiling on five or more numeracy tasks at pretest, and thus the final sample included those 80 children who had not yet attained these basic numeracy skills. Because EF was measured with two tasks, the two task scores were standardized and averaged to form an EF composite for the analyses.
A 2(pretest vs. posttest)×3(Training Condition) MANOVA on the eight numeracy tasks revealed that (a) children improved on counting after, numeral identification, number ordering (i.e., arranging three numerals from smallest to largest), and number line estimation across all conditions, (b) only children in the number+EF+RL training improved on counting backwards and only children in the number training improved on counting aloud. On the four numeracy tasks on which we observed overall improvement, we classified children from either number training condition as having improved more, or less, than the average task-specific gains across all 80 children (Figure 1). We found that in the number training, children who benefitted more than average on number identification and number ordering had lower initial EF skills compared to children who benefitted less than average on these two tasks (Figure 2).
Among children who have not mastered the basic numeracy skills, how much they learn from a particular type of instruction may depend on their initial EF skills. Focusing on counting and ordering numbers without the added demands associated with reflection, alternative strategies, and relational language instruction may help children with lower EF skills to improve their numeracy skills. Although EF skills should be supported in children with low EF, the added demands may interfere with improvements in numeracy skills seen in the number only training.
Thus, targeting numeracy skills themselves, rather than integrating EF and relational language in the instruction, may be more efficient at improving children’s early numeracy skills, especially among young children with low numeracy and EF skills.

Authors

• Jenny Yun-Chen Chan, TheEducation University of Hong Kong
  Presenting Author

• Maria D Sera, University of Minnesota
  Non-Presenting Author

• Michele Mazzocco, University of Minnesota - Twin Cities
  Non-Presenting Author