2.9.1. Activation Analysis

For individual-level analysis, the following two contrasts were created for the HbO data of the younger group, the expert older group, and the non-expert older group: “rest vs. calculation” and “rest vs. planning”. The β-values for each experimental condition and each channel were calculated. The statistical significance of the task-related brain activity was then verified by testing the null hypothesis that the estimated β coefficient was not significantly different from zero (t-test, p < 0.05). Note that multiple comparisons were corrected using the Benjamini–Hochberg method [42]. In addition, to test the hypothesis that there is no difference in the level of IPL activation between older adults who continually practiced Wellness Darts and younger adults who were not experienced in Wellness Darts, β-values of the IPL were compared between the three groups (t-test, p < 0.016, Bonferroni corrected).

Next, for the group-level analysis, we constructed the most consistent contrast images of group-level brain activity using simultaneously the β-values obtained from the individual-level analysis of all participants, based on [28]. To estimate significant group-level activity, we used participants as random effects and conditions (planning, calculation, and rest blocks) as fixed effects. Finally, all of the participants were considered as a sample drawn from the population, and a one-sample t-test was used to test if the population mean was greater than zero with respect to the t-value. The significance level was set at p < 0.05 (false discovery rate corrected at peak level), and the brain regions activated during the calculation and planning blocks were estimated.