DFG-Projekt Harmful Help from Information Systems
On the threatening effects of the help of agentic information systems and their implications for user-system collaboration
Aim of the research project:
The aim of the research project is to investigate the nature and effects of help invocation types (HATs) of agentic information systems (IS) and their impact on user-IS collaboration. Some studies already distinguish between user- and IS-invoked help (e.g., user- or IS-initiated provision of advice or execution of tasks), but these studies are mainly conceptual and there is still little understanding of how users respond to different HATs and in particular to anticipatory help (i.e., the IS invokes the offer of help) as opposed to reactive help (i.e., the user invokes help from the IS). Without knowing how, why and when users respond to anticipatory (vs. reactive) help, IS designers might use anticipatory help with the expectation that they will achieve the same or even better user responses than with reactive help, but in reality this could be detrimental to user-IS collaborations. Even though anticipatory help may be objectively useful for many user-IS collaborations, users may perceive their identity (e.g., competence, status) threatened by intervening IS. As a consequence of this perception, users may not only ignore the anticipatory help offered, but may even stop interacting with the IS altogether, ultimately turning a well-intentioned design of agentic IS into an unintended sabotage of user-IS collaboration.
Against this background, with this research project we aim to develop validated conceptual definitions and taxonomy for HATs based on a solid theoretical foundation and comprehensive empirical testing. In addition, we plan to validate reactive and anticipatory helping as the predominant HATs in a theoretical model (i.e., Self-Affirmation Theory) through a series of randomized online experiments. In particular, we want to investigate the short-term effects of HATs on (subjective) user perceptions, i.e., especially the perceived self-threat of a user, as well as the boundary conditions that enhance or mitigate these threatening effects. Finally, we aim to complement and corroborate the findings by examining the (long-term) outcomes of user-IS collaboration from an instrumental (e.g., acceptance of help, performance) and a humanistic perspective (e.g., satisfaction, perceived utilization).
To this end, we intend to cooperate with two companies that use different HATs in large-scale field experiments. With the research project, we aim to contribute to the literature on IS assistance and user-IS collaboration by providing a more theoretically sophisticated understanding of HATs and their implications and designs, which is becoming increasingly important with the proliferation of agentic IS.
Contact person:
Sponsor: