EXPERIMENTAL EVALUATION OF PERCEIVED MEMORABILITY IN KNOWLEDGE BASED AUTHENTICATION FOR VIRTUAL AND AUGMENTED REALITY

With the recent advances in Augmented and Virtual Reality technologies (AR/VR), these devices have reached a point of being available for mainstream use. This can be seen in current generation AR/VR devices such as the Meta Quest series, the Apple Vision Pro and Microsoft Hololens devices. These wearable devices change the assumptions about I/O that exist in the traditional model of computing, namely the use of input peripherals such as the keyboard and mouse. Despite this change, traditional artifacts that have remained prevalent in computing for decades have been ported into the AR/VR space, specifically the traditional text-based password as a form of Knowledge Based Authentication (KBA). This paper proposes a new measure in the context of KBA, perceived memorability and differentiates it from empirical memorability. This new measure is evaluated through a randomized controlled experiment that evaluates a novel KBA scheme designed for AR/VR, and assessed for reliability and validity using Cronbach’s alpha and principle component analysis respectively. We find that this measure shows ideal characteristics, and has potential for use in better understanding why users choose to engage in insecure authentication related behavior.