They continue to use both languages in daily life in a mixture of contexts. Therefore, all of the 8 participants in this group were considered to be proficient early bilinguals. A total of 20 concrete nouns were used in the present study. All of the words were chosen from a set of stimuli previously used for predicting fMRI activation patterns (Akama, Murphy, Li, Shimizu, & Poesio 2012) but without using pictures. They were classified into two categories and two languages: 10 tool words in Korean and their corresponding ones in Chinese,

10 mammal words in Korean and their corresponding ones in Chinese. Using the E-Prime 2.0-Standard software package, which synchronised during the experiments with the trigger Everolimus pulses transmitted by the fMRI control PC, the 40 words were randomly Gemcitabine shown on the screen. A slow event-related design was used in the present study. The participants participated in two separate scanning

sessions carried out over two different days whose order was counter-balanced across participants over the two days. Each session lasted 50 min. Each session had 6 repeated runs for a total of 240 trials. In each trial, each word was presented for 3000 ms, followed by a fixation cross for 7000 ms. There were six additional presentations of a fixation cross, 40 s each, distributed immediately after each run to establish a BOLD baseline. During the 3000 ms stimulus period, the participants were asked to perform a silent property generation task (Mitchell et al., 2008) with these word stimuli by thinking of the appropriate features of the corresponding concept and caption in a required language. This step was followed by a fixation cross presentation time of

for 7000 ms, during which the participants were asked to silently fix their eyes on the cross and no response was required. In one session, the participants were asked to perform the task covertly using the same language as the orthographic stimuli on the screen. We refer to this session as the ‘situational non-translation language-switching condition’, abbreviated here as SnT. In the other session, the participants were asked to perform the task using the other language, which is not visually presented in each trial. We refer to the second session as ‘focused simultaneous translation language-switching condition’, abbreviated here as FST (Fig. 2). To ensure that each participant had a consistent set of properties to think about during the on-line tasks, the participants were asked to acquaint themselves with these stimuli and to perform a property rehearsal task before the scanning session (Mitchell et al., 2008). Functional MRI scans were performed with a 3.0-T General Electric Signa scanner at the Tokyo Institute of Technology, Japan, with an 8-channel high-resolution head coil. The scanning parameters were based on those of Mitchell et al. (2008).