About Ψ Lab

In the first generation of computing devices, mainframes and personal computers were used primarily for data management and processing, following a set of predefined rules. Input devices designed specifically for interacting with these systems provided a deterministic mechanism for identifying the user’s intent. For example, when a mouse is moved or a key on a keyboard is pressed, the system receives this information with high certainty. The next generation realized—and perhaps continues to realize—the vision of ubiquitous computing. Portable devices of various sizes, such as smartphones and tablets, used at different times and locations, gave rise to applications that utilized available information about the context of the interaction to present relevant information to users on 2D displays. How we interact with these systems, including pen and touch input, has also moved away from stationary input devices toward interactions that more closely resemble how we manipulate physical objects.

Today, in the era of spatial computing, computation is increasingly distributed and situated in our surrounding environment. Wearable devices, such as augmented reality (AR) headsets and lightweight AI glasses, take the notion of “display” from a 2D screen to digital information that can be presented through multisensory channels (e.g., visual, auditory, tactile) at any time and within its real-world context. We are also increasingly surrounded by physical AI systems, such as autonomous mobile robots, that need to make inferences about people’s intent while operating in unseen environments. This generation of computing introduces greater complexity and uncertainty, posing new challenges beyond previous paradigms.

At Princeton’s Situated Interactions Lab (Ψ Lab), we tackle these challenges by studying and leveraging existing theories of human perception and cognition. Our goal is to build interactive technologies that are closely integrated with the world around us and create new opportunities to extend our abilities and augment our intelligence in situ.

Our research falls under three broad categories:

  1. On the user input side, we facilitate embodied interactions and interpret users’ low-effort multimodal communication, providing mechanisms to resolve ambiguity. 

  2. On the system output side, we aim to infer users’ implicit needs as they act in the world and intelligently present timely, minimal information in an appropriate representation and sensory modality. 

  3. Finally, we explore new interaction paradigms in 3D physical spaces to achieve symbiosis between humans and physically situated AI (e.g., robots, IoT devices), facilitating shared mental models and increasing communication bandwidth.

Our research focuses on lightweight wearables (e.g., AI glasses with minimal or no visual displays), but we also pursue a narrower line of work using VR and passthrough AR headsets. Grounded in sensorimotor control theory, we explore illusory interactions, which use subtle, unnoticeable sensory mismatches to enhance user experience (e.g., improving the perceived performance of haptic devices), and beyond-real interactions, which use noticeable but adaptable sensory mismatches to extend human abilities (e.g., walking at faster perceived speeds).

 

Applying to Princeton HCI

If these research directions interest you and you’d like to apply to join Ψ Lab and the broader Princeton HCI community, please read the information below.