Figure 2.

Schematic representation of language training with a deblocking method. The tablet-based training comprised three tasks: spoken word–picture matching for the intact modality, and naming and sentence production for the impaired modalities. In the spoken word-picture matching task, six photos were displayed on each page which contained one object describing one of 20 trained words and five filler objects, and then the trained word was presented by a female narrator’s voice. We instructed the participants to touch one photo corresponding to the spoken word. On each training day, the participants repeated the spoken word-picture matching task for the 20 trained words four times while they were given tDCS or sham stimulation to Broca’s area. After performing the spoken word-picture matching task for 20 min, the participants performed the naming task, which was designed to display one photo corresponding to one of the 20 trained words on each page. We instructed the participants to name the names of the 20 objects orally. After the naming task, they performed the sentence production task which was designed to display one photo corresponding with one of the 20 trained words on each page. We instructed the participants = to produce a simple sentence containing the word corresponding to the presented object. The time limit for each page was 15 s for all three tasks.