The goal of this article is to investigate factors that influence trust in Human-Robot Interaction (HRI), focusing on communication variables and social behaviors that might shape the users’ perception of robots. The research was conducted within the I-CATER project, which explores social robots in work environments. A questionnaire combining the Godspeed Questionnaire Series (GQS) and the Big Five Inventory-10 (BFI-10) was created to assess trust perceptions in different scenarios featuring different communication settings in error situations, initiatives to start a task, and varying facial expressions. Data analysis with the Friedman and Wilcoxon tests revealed that justifications and apologies improve perceptions of likeability and intelligence, while dynamic facial expressions increase perceptions of anthropomorphism, likeability, and animacy. Although age, occupation in the technology field, and ownership of robotic equipment at home did not significantly correlate with trust dimensions, weak gender correlations indicated that male participants tended to rate trust-related dimensions lower. The study contributes to the understanding of how robot communication and social behaviors influence trust in HRI.