Generalize HGR effects across contexts and systems

Determine whether different historical-grounding source configurations produce similar differentiated experiential effects across places with different historical-information densities, spatial forms, user populations, generative models, and narrative forms.

Background

The field study evaluated three source-allocation conditions on one route in one historical district, using a single model configuration and 18 participants. Consequently, the observed relationship between the Historical Grounding Ratio (HGR) and user experience may depend on the specific urban, technical, and participant context. Replication across substantially different settings is needed to determine whether the differentiated effects—especially the stronger place relevance at higher HGR and the apparent advantage of balanced configurations for other outcomes—are robust.

References

The present results therefore apply first to the urban historical walking context studied here. Future research should replicate the manipulation in places with different densities of historical information, spatial forms, and user populations, and compare whether source configurations produce similar differentiated effects across generative models and narrative forms.

Balancing Evidence and Interpretation: Historical Grounding Ratio as a Design Parameter for AI-Generated Urban Storytelling  (2608.24157 - Zhang et al., 25 Aug 2026) in Research Boundaries and Future Directions, Section 6.1

The present study cannot completely separate the effects of source composition and expression length. Future experiments should constrain the numbers of information units and sentences together with text length during generation, or prespecify length as a covariate in the analysis plan, to more strictly distinguish the experiential effect of source configuration itself.

Balancing Evidence and Interpretation: Historical Grounding Ratio as a Design Parameter for AI-Generated Urban Storytelling  (2608.24157 - Zhang et al., 25 Aug 2026) in Research Boundaries and Future Directions, Section 6.3

Future research should separate place relevance, referential specificity, historical credibility, and situatedness into distinct constructs and use multi-item measures to determine which place-related experience HGR changes first.

Balancing Evidence and Interpretation: Historical Grounding Ratio as a Design Parameter for AI-Generated Urban Storytelling  (2608.24157 - Zhang et al., 25 Aug 2026) in Research Boundaries and Future Directions, Section 6.4

This issue did not eliminate the overall manipulation difference among the three HGR conditions, but makes it difficult to determine why the low-HGR condition did not show the expected advantage in scene integration. Future studies should ensure consistent-quality situated input for every experimental segment and distinguish whether situated information is retrieved, whether it enters the generation context, and whether it ultimately appears in the narrative received by the user.

Balancing Evidence and Interpretation: Historical Grounding Ratio as a Design Parameter for AI-Generated Urban Storytelling  (2608.24157 - Zhang et al., 25 Aug 2026) in Research Boundaries and Future Directions, Section 6.5