Promoting diversity in ranking while maintaining the relevance of ranked results is critical for enhancing human-centered search systems. While existing ranking algorithm and diversity IR metrics provide a solid basis for evaluating and improving search result diversification in offline experiments, it misses out possible divergences and temporal changes of users' levels of Diversity Acceptance, which in this work refers to the extent to which users actually prefer to interact with topically diversified search results. To address this gap between offline evaluations and users' expectations, we proposed an intuitive diversity acceptance measure and ran experiments for diversity acceptance prediction and diversity-aware re-ranking based on datasets from both controlled lab and naturalistic settings. Our results demonstrate that: 1) user diversity acceptance change across different query segments and session contexts, and can be predicted from search interaction signals; 2) our diversity-aware re-ranking algorithm utilizing predicted diversity acceptance and estimated relevance labels can effectively minimize the gap between diversity acceptance and result diversity, while maintaining SERP relevance levels. Our research presents an initial attempt on balancing user needs, result diversity, and SERP relevance in sessions and highlights the importance of studying diversity acceptance in promoting effective result diversification.

How users think, behave, and make decisions when interacting with information retrieval (IR) systems is a fundamental research problem in the area of Interactive IR. There is substantial evidence from behavioral economics and decision sciences demonstrating that in the context of decision-making under uncertainty, the carriers of value behind actions are gains and losses defined relative to a reference point, rather than the absolute final outcomes. This Reference Dependence Effect as a systematic cognitive bias was largely ignored by most formal interaction models built upon a series of unrealistic assumptions of user rationality. To address this gap, our work seeks to 1) understand the effects of reference points on search behavior and satisfaction at both query and session levels; 2) apply the knowledge of reference dependence in predicting users' search decisions and variations in level of satisfaction. Based on our experiments on three datasets collected from 1840 task-based search sessions (5225 query segments), we found that: 1) users' search satisfaction and many aspects of search behaviors and decisions are significantly associated with relative gains, losses and the associated reference points; 2) users' judgments of session-level satisfaction are significantly affected by peak and end reference moments; 3) compared to final-outcome-based baselines, models employing gain- and loss-based features often achieve significantly better performances in predicting search decisions and user satisfaction. The adaptation of behavioral economics perspective enables us to keep taking advantage of the collision of interdisciplinary insights in advancing IR research and also increase the explanatory power of formal search models by providing them with a more realistic behavioral and psychological foundation.