<p>One of the primary dilemmas currently faced in environmental pollution conflict is the complex psychological game and multi-party collaboration among decision-makers (DMs), which significantly increases the difficulty of resolving environmental conflicts. To address this challenge, this study introduces a new coalition analysis method that considers DMs’ psychological behavior characteristics (PBCs) and full participation of sanctioning opponents within the framework of the graph model for conflict resolution. Due to differences in experience, targets, and cultural background, DMs take action not only based on the criterion of whether they can benefit from it, but also based on the changes in their opponents’ interests, thus leading DMs to exhibit aggressive, conservative, or balanced behaviors when decision-making. Firstly, satisfaction thresholds and agreement support functions are defined to introduce the concept of soft consensus. This integration can more effectively capture the coalition members’ consensus behaviors and psychological satisfaction during the process of preference aggregation. Secondly, reachable lists of the coalition based on full participation of sanctioning opponents and the influence of PBCs are constructed. Subsequently, the logical and matrix representations for coalitional stabilities considering the PBCs of DMs and full participation of sanctioning opponents are developed. Finally, an environmental pollution conflict in China is analyzed to illustrate the feasibility and effectiveness of the proposed method.</p>