An Adaptive User Interface Based on Psychological Test and Task-Relevance

Abstract

The current advances into Human-Computer Interaction (HCI) are accompanied by the quick growth of Machine Learning (ML) and Artificial Intelligence (AI). It opens countless possibilities to enhance the interaction and communication between the machines and humans. Nevertheless, the human does not process the information at same way of computers (step-by-step); thus, the interfaces must be adequate to human’s capabilities and limitations. The Graphical User Interfaces (GUIs) are the most used way to provide information to the user because the visual information is a straightforward and natural method to interact with the human. A new step forward is integrated some intelligent behavior for adapting the GUIs to the context, user’s necessities and personalization; hence, the Adaptive User Interfaces (AUIs) are a new proposal which integrates intelligent and adaptive capabilities to achieve an enhanced human-computer interface. In this paper, an AUI is proposed based on psychological tests, which are used to assess the cognitive load that interface produces to the user so to know how much information must be shown and thus the user does not undergo overloading. These cognitive data is joined with a relevance weight, which represents what information is necessary to show in a given situation (Task-Relevance), in order to give to the user the most important data with the less cognitive load. The paper’s aim is to show how psychological tests can be used as input information for an adaptive interface; in that sense, each them are explained and a possible user case is shown.