Numerous studies have explored the relationship between cooperation and inter-individual trust. However, the understanding of the dynamic interaction process of trust among networked agents with bounded rationality remains limited. To bridge this gap, this paper develops an evolutionary game model that incorporates a trust perspective considering emotional scaling. Trust is quantified based on individuals’ limited memory of interaction experiences, acting as a perturbation factor that influences cooperative strategies. The role of emotional factors, particularly guilt and anger, in trust formation is emphasized. Extensive numerical simulations reveal that introducing trust greatly enhances cooperation in the system, especially in networks with lower randomness. Trust-based cooperative decision-making demonstrates evident historical path dependence characteristics. Notably, the effects of emotions differ. Guilt effectively motivates defectors to switch strategies, thereby enhancing cooperation. In contrast, the influence of anger on cooperation diminishes as it becomes intertwined with other factors during the trust-building process. The empirical results of higher returns for low-trust groups have also been validated in this paper. Overall, this work contributes to a deeper understanding of the mechanisms underlying cooperation formation.