Abstract

A computer-based simulation of a multi-agent networked, virtual organization is experimented with to explore the impact of human-agents misrepresenting information to others when performing an organization-level decision task. It is a common-place phenomenon that humans make misrepresentations (or errors) in judgment (accidental or otherwise) when evaluating information and then passing their resulting conclusion on to others, which can affect the accuracy of organization-level decisions. This can have great relevance in organizations under stress or undergoing large-scale change, e.g., as in the case of a merger or acquisition, as humans can make even more misrepresentations. For this study, repeated simulations are executed with different levels of actor knowledge-misrepresentation to assess the impact on the dynamics of organization performance. Organization performance is quantified by the accuracy of a knowledge-based task, repeated over multiple time periods. Each task is a single binarydecision, which is derived by actor majority vote. Actors base their vote on the partial, factual knowledge that each holds, but their best-effort vote can be reversed according to the actor’s tendency to misrepresent the totality of the accurate information they truly hold. The underlying

Citations