Understanding How Psychological Distance Influences User Preferences in Conversational versus Web Search

Research Background and Issues

• What problems or challenges did the authors identify?
  The study explores how users' perceived psychological distance in tasks influences their preference between conversational search systems (e.g., ChatGPT) and traditional web search engines (e.g., Google Search). While previous research has compared the performance of these two types of information retrieval systems, few studies have examined how changes in the nature of search tasks affect user preferences.

• Why is this issue important?
  Psychological distance impacts how people process and evaluate information, which can directly influence user acceptance and satisfaction with technology. Understanding these factors is crucial for designing better search engines and improving user experience.

• Research Motivation and Related Work
  This study is grounded in Construal Level Theory (CLT), which posits that psychological distance affects users' levels of thinking (concrete vs. abstract) and their information needs. Additionally, previous research has shown that while conversational search systems are more humanized, they face challenges such as unverifiable sources and concerns about information reliability, which may influence users' system choices.

Solutions

• What methods or solutions did the authors propose?
  The authors designed an experimental study to manipulate the psychological distance of search tasks (near vs. far) as a variable, investigating how psychological distance affects user preferences for traditional web search versus conversational search.

• What is innovative about this solution?
  The study is the first to systematically explore how the psychological concept of distance interacts with information retrieval (IR) technologies. It integrates an extended Technology Acceptance Model (TAM) to analyze changes in users' perceptions of "ease of use," "information credibility," "usefulness," and "enjoyment." Furthermore, the study offers specific recommendations for improving information retrieval design based on psychological distance.

• What are the implementation steps and key techniques used?

  • Experimental Design: A 4×2 between-subjects design was employed, categorizing psychological distance into four dimensions (spatial, temporal, social, hypothetical) and two levels (near and far).
  • Experimental Process: 128 participants completed tasks with manipulated psychological distances, using both web search and conversational search. They then filled out questionnaires evaluating ease of use, information credibility, usefulness, and enjoyment, while also reporting reasons for their preferences.
  • Data Analysis: The effects of psychological distance were analyzed using paired-sample t-tests, Wilcoxon signed-rank tests, and chi-square tests, complemented by semantic and thematic analyses to uncover deeper reasons behind user preferences.

Research Findings

• What specific findings were achieved?

  • Relationship Between Psychological Distance and User Preference: Greater psychological distance led to a stronger preference for conversational search, which was perceived as more credible, easier to use, enjoyable, and useful.
  • Ease of Use (H1): For tasks with greater psychological distance, conversational search was perceived as easier to use due to its ability to save effort and integrate information.
  • Information Credibility (H2): Users perceived conversational search as more credible (especially in terms of "accuracy" and "reliability") as psychological distance increased, though it still lagged behind web search in "authenticity."
  • Usefulness (H3) and Enjoyment (H4): With greater psychological distance, conversational search was seen as more useful and enjoyable due to its efficiency, natural interaction, and strong integration capabilities.
  • System Acceptance (H5): Overall, users were more inclined to accept conversational search when psychological distance was greater.

• What advantages does it have compared to existing solutions?
  This study fills a research gap by examining the impact of psychological distance on IR tool selection, offering a theory-driven approach to guide the optimization of conversational search systems. It also provides comprehensive design recommendations based on user tasks and psychological states.

• What were the experimental or evaluation results?
  Experimental data showed that conversational search outperformed web search in terms of "ease of use," "credibility," "usefulness," and "enjoyment" when psychological distance was greater. When psychological distance was smaller, the user experience differences between the two systems were minimal.

• Limitations and Future Directions

  • The experimental task scenarios were limited to travel planning, which may not generalize to more specialized or highly specific applications.
  • The experiment did not account for counterbalancing task order (e.g., web search first, then conversational search), requiring future studies to verify whether task order affects the findings.
  • Future work could extend to different task scenarios (e.g., healthcare, education) and analyze other potential behavioral factors (e.g., cultural background or individual skills).

Conclusion and Design Implications

1. Optimization of Conversational Search: Systems should dynamically adjust the abstraction level of output information (concrete vs. abstract) based on task psychological distance to better meet diverse user information needs. For example, provide detailed plans for near-distance tasks and high-level overviews for far-distance tasks.
2. Integration of Web and Conversational Search: Embedding web search functionality within conversational interfaces could combine the strengths of both modes to meet diverse information needs. Additionally, conversational systems could leverage natural language processing to proactively assess the psychological distance of user tasks and dynamically recommend appropriate information retrieval methods.
3. User Trust and Transparency: Future conversational search systems should improve the visibility of source citations and reliability while explicitly presenting biases in model training data and their potential impacts to enhance transparency.

This study lays a theoretical and empirical foundation for optimizing user experience in information retrieval technologies, demonstrating the significant value of applying the psychological theory of psychological distance to the field of technology applications.