Fit Matters: Format-Distance Alignment Improves Conversational Search

TL;DR

This paper addresses how conversational-search responses can be tailored to users' cognitive states by aligning information presentation formats with psychological distance, grounded in Construal-Level Theory. It tests a $2 × 2 × 4$ factorial design across distance (temporal/spatial near vs far), granularity (abstract vs concrete), and media (text vs image-and-text) using a travel-planning task, demonstrating that format–distance alignment improves decision confidence, perceived usefulness, ease of use, enjoyment, credibility, and adoption intent while reducing perceived risk. Concrete information increases cognitive load, but gains are productive when matched to near-distance tasks; multimedia benefits depend on content granularity, with image-enhanced concrete content driving the strongest user experience improvements. The findings establish format–distance alignment as a distinct usability dimension and offer actionable guidance for adaptive information presentation in conversational AI, including cue-based detection of distance, granularity calibration, and nuanced media selection. Together, these contributions advance cognitive-fit design in AI systems and have practical implications for delivering more effective, engaging, and trustworthy conversational search experiences.

Abstract

Existing conversational search systems can synthesize information into responses, but they lack principled ways to adapt response formats to users' cognitive states. This paper investigates whether aligning format and distance, which involves matching information granularity and media to users' psychological distance, improves user experience. In a between-subjects experiment (N=464) on travel planning, we crossed two distance dimensions (temporal/spatial x near/far) with four formats varying in granularity (abstract/concrete) and media (text/image-and-text). The experiment established that format--distance alignment reduced users' risk perceptions while increasing decision confidence, perceptions of information usefulness, ease of use, enjoyment, and credibility, and adoption intentions. Concrete formats imposed higher cognitive load, but yielded productive effort when matched to near-distance tasks. Images enhanced concrete but not abstract text, suggesting multimedia benefits depend on complementarity. These findings establish format--distance alignment as a distinctive and important design dimension, enabling systems to tailor response formats to users' psychological distance.

Figures (7)

• Figure 1: Overview of the experimental design and participant flow. The left panel (a) summarizes the 2 $\times$ 2 $\times$ 4 between-subjects structure that crosses psychological distance (temporal vs. spatial $\times$ near vs. far) with four information presentation formats (abstract text, detailed text, abstract image-and-text, detailed image-and-text), yielding sixteen experimental conditions. The right panel (b) depicts the common sequence of steps that each participant completes, from consent and scenario presentation through manipulation checks, the single-turn chat interaction with the assigned format, the travel-plan writing task, and the post-task outcome measures. Note. PD = psychological distance; IPF = information presentation format.
• Figure 2: Cognitive load (NASA-TLX) as a function of psychological distance and information granularity. Lower scores indicate reduced cognitive burden. Box plots show median, quartiles, and data range. The lines connecting the Near/Far group medians are provided as visual aids and were not used in statistical analyses. Significance levels: *p < .05, **p < .01.
• Figure 3: Symmetric matching effects of information granularity and psychological distance alignment. Matched conditions consistently outperformed mismatched ones across all four user-experience dimensions. Box plots show median, quartiles, and data range. The lines connecting the Near/Far group medians are provided as visual aids and were not used in statistical analyses. Significance levels: *p < .05, **p < .01, ***p < .001.
• Figure 4: Asymmetric matching effects revealing differential sensitivity to format--distance misalignment. Panel (a) shows that decision confidence benefits exclusively from concrete--near matching, and is unaffected by abstract information. Panel (b) shows that risk perception increases only when abstract information mismatches with near distance, and that concrete information maintains stable risk across distances. These unidirectional patterns contrast with the bidirectional benefits observed in the other dimensions. Box plots show median, quartiles, and data range. The lines connecting the Near/Far group medians are provided as visual aids and were not used in statistical analyses. Significance levels: **p < .01, ***p < .001; ns = not significant.
• Figure 5: Main effects of media type on user-experience outcomes. Error bars represent 95% confidence intervals. Significance levels: *p < .05, **p < .01.