Multiple areas in the cerebellum have been reported to be engaged in reading. However, how these regions cooperate with the reading-related areas in the cerebrum remains unclear. Here, brain images of fifty-two adults were acquired via functional magnetic resonance imaging. By comparing the cerebellar activation across three localization tasks targeting orthographic, phonological, and semantic processing, we first identified three different reading-related areas in the cerebellum, biased toward orthography, phonology, and semantics, respectively. Then, functional connectivity (FC) analyses demonstrated that the mean FC between functionally corresponding areas across the cerebrum and cerebellum was greater than that between noncorresponding areas during silent word reading. FC patterns of functionally corresponding areas could significantly predict reading speed, with the FC driven from orthographic and semantic areas contributing the most. Effective FC analyses further showed that orthographic and semantic areas in the cerebellum had selective and direct connectivity to areas in the cerebrum with similar functional specificity. These results suggest that reading-related areas vary in their functions to reading, and cooperation between areas with corresponding functions was greater than that between noncorresponding areas. These findings emphasize the importance of functional cooperation between the cerebrum and cerebellum during reading from a new perspective.
This study examined the effect of short-term language-switching training on the cognitive control mechanism in bilingual word production. In two experiments, two groups of relatively proficient but unbalanced Chinese–English bilinguals performed a cued picture-naming task, in which they switched between their two languages. On two consecutive days, the participants took part in four sessions. The same procedure was employed on 2 days in Experiment 1, whereas the cue-language mapping was reversed on Day 2 in Experiment 2. In both experiments, picture naming in the dominant language (L1, Chinese) was slower than that in the weaker second language (L2, English) in all sessions. In addition, the reversed language dominance effect was enhanced with training, suggesting that training proactively increases the amount of inhibition of the dominant L1 at the global level. Furthermore, switching costs in the L1 were reduced with training in Experiment 1, but not in Experiment 2. These results indicate that language-switching training improves the efficiency of reactively exerting inhibitory control over the dominant L1 at the local level. However, when a cue matches with different target languages, the effect of training is absent at the local level. These findings reveal the plasticity and complexity of the cognitive control mechanism as a function of bilingual experience, particularly in language switching.