Figure 5.
Activation of KF-localized voxels. A 2 × 2 mixed ANOVA (Training Group × Font) was computed (untrained controls and pattern stimuli were not included in the ANOVA). There was a Training Group × Font interaction ( p < .001) and a main effect of Font ( p < .001), but no significant main effect of Training Group ( p = .932). Significant independent sample t tests are shown. The trained groups showed a significantly greater response in the KF-localized voxels when viewing their trained font compared with when the untrained groups viewed the font. When viewing the FaceFont stimuli: FF group response vs. KF group response, p = .002; FF group response vs. control response, p = .018; no significant differences between the two untrained groups, KF group vs. controls, p = .373. When viewing KoreanFont stimuli: KF group response vs. FF group response, p < .001; KF group response vs. control response, p < .001; no significant differences between the two untrained groups, FF group response vs. controls, p = .628. The magnitude of the trained font response was significantly greater for the KF group compared with the FF group, p = .032.