Difference in Language Switching Cost in Language Switching between Dominant and Weak Languages in Early Adult Javanese-Indonesian Bilinguals

This replication experiment study examines the asymmetric cost of language code switching in bilingual Javanese-Indonesian and tests the suitability of the cost pattern with the inertial task set hypothesis. Participants consisted of 72 students (49 women, 23 men; aged 20–23 years) who were native Indonesian speakers with Javanese as a second language (L2) or vice versa. The experiment was carried out using a Microsoft Excel-based program to pronounce numbers in both languages. The response records were analyzed to measure the reaction time (in seconds) of the correct responses. Reaction times for incorrect answers were excluded from the analysis. The average reaction time was analyzed using ANOVA and paired sample t-tests with SPSS version 22. The results showed that there was no significant difference in code transfer costs in the transition from L1 to L2 and vice versa, either in the group with the dominant Indonesian language (L1?L2: M = 0.62 s, SD = 0.12; L2?L1: M = 0.62 s, SD = 0.11; t = 0.24, p = 0.98) or Javanese (L1?L2: M = 0.59 s, SD = 0.11; L2?L1: M = 0.59 s, SD = 0.11; t = 1.05, p = 0.3). However, a significant difference was found between mentions with code transfer (M = 0.61 s, SD = 0.11) and without code switching (M = 0.57 s, SD = 0.1; t = 3.88, p = 0.00). These findings do not support the task set inertia hypothesis and indicate balanced bilingual characteristics in the study subjects.