Past Research

Flow in Artificial Intelligence

People experiencing the condition of flow appear to be fully engaged in their task, and their cognitive faculties and attention are fully focused on the task. In order for this to happen, there has to be a match between the task complexity and the individual’s abilities. Thus, it is of interest to model flow mathematically as well as to consider whether giving artificial cognitive agents an ability to sense flow and the desire to maximize flow can optimize their performance too.



Interactive Storytelling

In interactive storytelling, the audience is an active participant of a story, able to perform actions that shape the story’s progression (e.g., in a video game). The key research problem of interactive storytelling is how to balance the need for a coherent story with audience agency. In response, over the last twenty years the field has developed a number of experience managers. An experience manager is an AI agent that observes an audience’s actions and decides how to respond to them by modifying the narrative world. Accordingly, most existing work on experience management in interactive storytelling draws heavily from AI decision-making technology such as automated planning, search and machine learning.



Trainee-Specific AI-managed Training

Main problem-solving domains such as real-time damage control aboard naval vessels or emergency medical care require a human to make critical decisions in real-time with respect to several simultaneous phenomena. Such domains can be challenging due to the time and resource constraints and stressful due to high costs of making a mistake. Moreover, opportunities for real-life training are often limited as such training can be expensive and dangerous.


Computational Models of Human and Animal Cognition

Examines problems of the traveling salesperson problem using pigeons with increasing set sizes of up to six goals, with each set size presented in three distinct configurations, until consistency in route selection emerged.

The study of human hide and seek behavior in real-world environments as well as their virtual reality counter-parts. In addition to statistical analysis of the collected data, a predictive model of hide-and-seek behavior was built.

Emotions and culture have a profound effect on most human behavior and, therefore, they should be modeled in any high-fidelity virtual human simulation. The combination of an existing computational model for culturally affected behavior (CAB) with a subset of an appraisal-based emotion model (EMA) resulted in Culture, EMotion and Adaptation (CEMA), a light-weight computational engine capable of adding culturally and emotionally affected behaviors to non-playable characters in immersive training systems.