ABSTRACT
Most studies have investigated brain activation changes after the course of arithmetic learning, and the question remains whether these changes are detectable during the course of learning, i.e., before memory consolidation. Twenty-four fifth graders solved multiplication problems while ongoing electroencephalography (EEG) was recorded. The arithmetic training revealed reduced errors together with a power increase in theta (4–7 Hz) but not in lower alpha (8–10 Hz) or upper alpha (10–13 Hz) bands. We conclude that increases in theta power subserved a shift from slow, procedural strategies to more efficient, automated procedural and retrieval strategies, which led to more efficient performance.
Notes
1. Although there was a gender imbalance in our sample (15 boys and 9 girls), the comparison between boys and girls did not reach significance in either RT, t(22) = 0.39, p = .704; or error rate, t(22) = 0.54, p = .603. Henceforth, no gender difference was observed in arithmetic learning in our study. Similar between-gender comparisons were conducted for the EEG data. No significant difference was observed between boys and girls in the alpha bands (lower alpha: t(22) < 1.88, p > .077; upper alpha: t(22) < 1.34, p > .199). In theta, the only significant differences were found at Pz and Oz (Pz: t(22) = 2.58, p = .017; Oz: t(22) = 2.10, p = .048), which did not survive FDR correction for multiple comparisons. This correction has been applied for all comparisons in the current study. Therefore, we conduct all the analyses for both groups together and will not consider gender as a factor.