This research explores the impact of general trust networks on the willingness to communicate (WTC) in English among Japanese individuals. Previous research has established a positive correlation between general trust and WTC in English for Japanese learners; however, the underlying network structure and its influence remain unexplored. This study addresses this gap by employing a network analysis approach. The study aims to unveil the intricate relationships between various facets of general trust and their impact on English communication, providing a more nuanced understanding than traditional regression analysis. Understanding these dynamics is crucial for developing effective interventions in language education and improving the communicative competence of Japanese English learners. The research is particularly significant given the cultural context of Japan, where interpersonal trust and network structures are known to differ from those in Western societies.

Prior research has extensively explored factors affecting attitudes toward second language (L2) communication, focusing on psychological and contextual influences. Willingness to communicate (WTC) is widely considered an indicator of positive attitudes towards L2 communication. Studies have examined correlations between WTC and L2 communication frequency, incorporating psychological aspects such as language confidence, personality traits, attitudes, motivation, and contextual factors such as classroom environment and teacher support. Previous studies have investigated the effects of classroom social climate (teacher support, mutual respect), classroom environment (teacher support, student cohesiveness, task orientation), and situational antecedents of WTC (situation cues, situation characteristics) on L2 WTC among learners from diverse backgrounds (Canadian, Chinese, Iranian, Japanese, Korean). However, these studies mainly focused on interpersonal relationships within specific settings (e.g., teacher-student relationships within a classroom). This study differentiates itself by focusing on general interpersonal relationships and the broader concept of general trust as a key factor influencing WTC in English for Japanese speakers.

The study employed a network analysis approach using data from 1362 Japanese participants: 761 university students (372 men, 384 women, 5 others; mean age = 19.37 years) and 601 social survey participants (284 men, 317 women; mean age = 39.65 years). University students were surveyed in their classes using paper-based or online questionnaires, while the social survey was conducted online. The questionnaires included the General Trust Scale (Yamagishi & Yamagishi, Japanese version) with six items assessing general trust (Cronbach's alpha = 0.86 for all participants), and the WTC in English scale (McCroskey, Japanese version) with twelve items measuring WTC across various communication contexts and receiver types (Cronbach's alpha = 0.96 for all participants). Data analysis involved regression analysis to assess the overall effect of general trust on WTC, followed by network analysis. LASSO (least absolute shrinkage and selection operator) regularization with EBIC (extended Bayesian information criterion) model selection was employed to estimate the network structure and partial correlations between variables. The R packages qgraph, glasso, and dplyr were used for network estimation. Centrality indices (node strength, closeness, betweenness, expected influence, clustering coefficient) were calculated to determine the importance of each node in the network. Bootstrapping methods (bootnet package) were used to assess the stability and reliability of the estimated network structure and centrality indices. Case-dropping bootstrapping was performed to evaluate the stability of the centrality indices.

Regression analysis confirmed a significant positive effect of general trust on WTC in English for all participants (β = 0.24, p < 0.01). Network analysis revealed that four variables from the general trust scale directly and positively influenced WTC: "Most people are trustworthy," "Most people are trustful of others," "Most people are basically honest," and "I am trustful." Variables 2 ("Most people will respond in kind when they are trusted by others") and 6 ("Most people are basically good and kind") had indirect effects. Analysis of the network structure for university students and social survey participants separately showed some differences. Variable 2 showed a positive effect on WTC for university students but a negative effect for social survey respondents, suggesting a context-dependent effect. Centrality analysis revealed that variables 1 ("Most people are trustworthy"), 3 ("Most people are trustful of others"), and 4 ("Most people are basically honest") held central positions in the network across all participant groups, indicating their significant role in influencing WTC. Bootstrapping analysis confirmed the reliability of node strength, closeness, and expected influence, but not betweenness. The edge between variables 3 and 4 exhibited the highest edge weight, while the edge between variable 1 and WTC had the strongest effect on WTC.

The findings demonstrate the value of network analysis in complementing traditional regression analysis. While regression analysis revealed an overall positive relationship between general trust and WTC, network analysis provided a more detailed understanding of the interactions between specific variables and their relative importance in influencing WTC. The identification of central variables within the general trust network highlights potential targets for intervention strategies in language education. The differing effects of variable 2 across participant groups underscore the importance of considering contextual factors and potential negative experiences in shaping the relationship between general trust and WTC. The study’s focus on general trust as a reflection of Japan's social structure offers valuable insights into cultural influences on L2 communication. The high centrality of specific general trust variables suggests that targeting these variables in language education programs might effectively improve WTC among Japanese learners.

This study successfully mapped the network structure of general trust and its effects on WTC in English among Japanese individuals. Network analysis provided a richer understanding of the interplay between variables compared to regression analysis alone. The identified central variables represent potential targets for intervention strategies to enhance WTC in language learning contexts. Future research should explore general trust networks and their influence on WTC in other cultural contexts, particularly in Western societies to compare and contrast the findings.

The study is limited by its focus on a Japanese sample. The findings may not generalize to other cultural contexts. Further research on diverse populations is needed. The study did not explicitly measure opportunities for English communication outside of the classroom settings, which might influence the relationships observed. Future research should explore this aspect more thoroughly.