Revisiting Human Information Foraging: Adaptations for LLM-based Chatbots

TL;DR

This paper reevaluates Information Foraging Theory (IFT) in the context of LLM-based chatbots, asking whether the cost-value optimization, patch models, and diet model translate to interactive, non-patchy chatbot environments. It contrasts traditional web foraging with chatbot interactions, using thought-experiment narratives to derive a set of initial hypotheses about prey specification, patch structure, patch transience, information availability, and cost-value adaptations. A central claim is that trust acts as an ecological adaptation shaping cost-value judgments in chat-based foraging, potentially modulating how users frame prompts and assess responses. If validated, these ideas could guide the design of more effective, trust-aware chatbot interfaces and inform theory-building for information seeking in AI-assisted environments.

Abstract

Information Foraging Theory's (IFT) framing of human information seeking choices as decision-theoretic cost-value judgments has successfully explained how people seek information among linked patches of information (e.g., linked webpages). However, the theory has to be adopted and validated in non-patchy LLM-based chatbot environments, before its postulates can be reliably applied to the design of such chat-based information seeking environments. This paper is a thought experiment that applies the IFT cost-value proposition to LLM-based chatbots and presents a set of preliminary hypotheses to guide future theory-building efforts for how people seek information in such environments.