Emotional Contagion Engineering in Human-AI Collectives

Building on Kadiyala et al.'s (2025) fascinating work on emotional integration in LLMs, this research takes it a step further by examining how emotions spread between humans and AI systems in decision-making contexts. While they showed emotional states affect AI responses, nobody has really studied how those AI-generated emotions might influence human group members in return. This would create a feedback loop where AI systems don't just respond to emotions but actively shape the emotional landscape of the group to optimize decision outcomes. Think of it as emotional architecture for collectives - designing systems that might introduce calculated optimism when groups are overly risk-averse, or foster appropriate concern when groups are becoming reckless. This diverges from the purely technical approaches in papers like He & Wang (2025) and Bai et al. (2022) by recognizing that emotions aren't noise to be filtered out but signals to be orchestrated. The radical innovation here is treating emotional dynamics as an engineering problem rather than a psychological phenomenon, potentially opening up entirely new ways to enhance collective intelligence.

References:

1. Analyzing the Service Quality Evaluation of Railway Cold Chain Logistics Based on Probabilistic Linguistic Group Decision-Making. Tianhong He, Qingmao Wang (2025). International Journal of Decision Support System Technology.
2. Incomplete intuitionistic fuzzy behavioral group decision-making based on multigranulation probabilistic rough sets and MULTIMOORA for water quality inspection. Wenhui Bai, Chao Zhang, Yanhui Zhai, A. K. Sangaiah (2022). Journal of Intelligent & Fuzzy Systems.
3. Enhancing Collective Intelligence in Large Language Models Through Emotional Integration. L. Kadiyala, Ramteja Sajja, Y. Sermet, Ibrahim Demir (2025). arXiv.org.