Forging engaging, actionable paths to understanding human behavior with AR research

Session Host/Speaker(s)

We make decisions every day. Nevertheless, our knowledge about the processes associated with decision-making are still poorly understood. Basic research evaluating our decision-making processes usually struggles with noisy and inconsistent data.

Augmented Reality has experienced an explosion of mind-share in the last few years, with a high and growing percentage of modern mobile devices supporting AR. Research using AR has concentrated in two main topics: marketing and clinical applications. The former involves decision-making processes. However, this existing body of research carries a burden of poorly understood assumptions related to our basic knowledge on the processes and mechanisms behind choice and decision-making in humans.

In a typical experiment that explores these core mechanisms, researchers will use a computer task in which participants choose between two options (for instance, two circles). The consequence of that choice is usually a high score with no meaning, and the behavior is a click on the screen. These tasks are usually extremely tedious, resulting in participants' engagement and attention being limited at best. They are also a weak emulation of how we make decisions in the real world: we are usually confronted with sequential decisions (e.g. do you accept this job offer or keep looking?*). In this presentation we outline an applied, data-driven method to create a more naturalistic and more engaging task to study the foundations of decision-making via AR.

*In evolutionary terms, we were not evolved to make simultaneous but sequential choices.