In the first set of analyses, the demographic characteristics
of participants were compared. In a second set of
analyses, the data on primary outcome measures for
treated participants were analyzed separately for each
task using mixed ANOVAs that included Group (MCI,
NI) as a between-subject factor, as well as Intervention
(Pre-intervention; Post-intervention) and experimental
condition as repeated factors. A post-hoc HSD Tukey test
for unequal sample size was used to locate differences
in case of significant main effects or interactions. The
Bonferonni correction (0.007) was used to compare prepost-
effects on primary outcome measures. In a third
section, data analyses for the untreated participants are
presented. These were analyzed separately because this
was conducted as an independent experiment. As a result,
assignment in the two conditions was not random,
but consecutive, and participants were not matched on
demographic and cognitive variables. In such conditions,
running a single analysis is undesirable.