Modeling conversational flows for in-store mobile decision aids

Abstract

Based on the Human-Elaboration-Object-Construal (HEOC) Contingency Model, we propose design principles for modeling conversational flows between consumers and an in-store mobile decision aid (MoDA) with artificial intelligence, functioning as a virtual sales associate. Through an on-going assessment of the quantity, type, and specificity of the decision preferences from the user’s spoken input, MoDA is modeled to identify the user’s levels of decision elaboration and construal, which leads to its recognition of the user’s use of and shifts across four decision strategies commonly applied in consumer decision-making contexts. Upon identification of the user’s decision-making strategy, MoDA is modeled to (1) identify strategy-relevant assistive tasks, (2) generate or access strategy- and task-relevant intelligence, and (3) utter strategy-, task-, and intelligence-relevant speech to naturally support the user’s decision making strategy. The proposed design principles further map the types and examples of the agent tasks, intelligence, and speech required across the four consumer decision making strategies.