During the past decades, working memory (WM) training has attracted considerable research attention, but its transfer to untrained tasks is still controversial. In a randomized controlled trial, we investigated the possible transfer effects of a novel sentence-level WM training regime. Sixty-eight healthy Finnish adults were randomized into either a WM training group or an active control group. The WM training group practiced for 4 weeks with two adaptive sentence-level WM training tasks, namely, a novel sentence-level updating task and a Reading span task. The active control group practiced on a quiz task that called for long-term memory but did not load on WM. There were no statistically significant training effects on the pre–post measures of near and far transfer. We suggest that the lack of training effects may reflect the specificity and automaticity of the sentence-processing system.
- working memory