Figure 2.

Naming accuracy difference scores. Shows differences scores of T4 (final time point) minus T1 (baseline) in each condition. Trained words achieved a difference of less than 2%, while control words declined over 20%. Errors bars represent standard error of the mean.