Triangulating on Possible Futures: Conducting User Studies on Several Futures Instead of Only One

2024-09-21

arXiv.org

AI Summary

HCI future research is often constrained by single-scenario prototyping, limiting the generalizability of predictions. To address this, we propose the “Triangulated Futures” method: concurrently designing and empirically evaluating multiple AI-augmented knowledge work prototypes across distinct future scenarios, and triangulating qualitative user data from cross-context studies to systematically distinguish co-occurring phenomena—yielding generalizable insights—from context-specific phenomena—representing localized findings. This approach transcends the conventional single-future-prototype paradigm by integrating future-oriented prototyping, multi-context comparative analysis, and rigorous HCI user research. Applied in two divergent AI-knowledge-work futures—one centered on collaborative AI writing and another on AI-mediated expert consultation—we identified robust, cross-context interaction patterns and technology adoption mechanisms. The method significantly enhances the generalizability, critical foresight capacity, and scholarly reproducibility of HCI future research.

Technology Category

• Humans and AI: Understanding People, Theories, Concepts and Methods
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience

Abstract

Plausible findings about futures are inherently difficult to obtain as they require critical, well-informed speculations backed with data. HCI addresses this challenge with user studies where futuristic prototypes and other props concretise possible futures for participants. By observing participants' actions, researchers can"time-travel"to the future and see it alive, in action. However, a single study may yield particularised findings, inherent to study's intricacies, and lack wider plausibility. We suggest that triangulation of possible futures helps researchers disentangle particularities from findings that have wider plausibility. We explored this approach by arranging two studies on different futures of AI-augmented knowledge work. Some findings emerged in both studies while others were particular to only one or the other. This enabled us both to cross-validate their plausibility and gain deeper insights. We discuss how triangulation of possible futures makes HCI studies more future-proof and provides means to more critically anticipate possible futures.

Problem

• Triangulating multiple futures for broader plausibility
• Conducting user studies on AI-augmented knowledge work
• Enhancing HCI studies with reflective future anticipation

Innovation

• Triangulating multiple futures
• User studies with prototypes
• Cross-checking plausibility insights