OUR GENERAL APPROACH
CARISMA uses advanced laboratory technologies of today to simulate and study the transformtive impact of the communication technologies of tomorrow. We ask:
What are the technical requirements to meet our perceptual criteria of reality and create credible virtual experiences?
How can we use virtual environments and avatars to increase individual wellbeing and improve communication and social outcomes?
How can experiences in virtual worlds help solve problems in the real world?
CARISMA conducts research at the intersection
of human behavior and technology
CARISMA uses VR technology as a methodology to study the basics of human communication and nonverbal interaction. Motion capture, neurophysiological measures and avatar animations provide new insights into the bio-behavioral mechanisms underlying person perception, emotion inferences, conflict management, and social bonding.
Nonverbal behavior is a subtle means of social influence and relational communication. Beyond universal mechanisms underlying our inferences about others emotions and intentions there is considerable inter-individual variance in nonverbal skills and habits. This diversity enriches our social world but also can cause misunderstandings and conflict. In CARISMA we use avatars and character animations to study the diversity of communication and social cognition, including gender and cultural differences as well as communication impairments, such as Autism Spectrum Disorder (ASD).
Diversity in Nonverbal Communication
Whenever two or more humans show some kind of repetitive motion, they are prone to synchronize. Movement synchrony is a crucial element in social coordination and collaboration. It occurs when people are walking together, sitting in rocking chairs, rowing boats, cheering their teams, and, more subtly, in our everyday conversations. Synchrony in behavior apparently resonates in neurophysiological synchrony and shared intentionality. Using our interactive avatar platform we study the influence of bio-behavioral synchrony on pro-social behavior and interpersonal trust.
Bio-behavioral Synchrony and Collaboration
Nonverbal behavior a subtle yet powerful means to bond with others but often also drives people apart. Yet, the mechanisms which guide our perception of others and foster mutual attunement and rapport are still elusive. Using our unique movement and neurophysiological capture facilities we record large database consisting of hundreds of dyadic interactions of formal and informal character. In collaboration with Computer Science we will apply machine learning algorithms to identify the major constituent of rapport on the behavioral as well as on the brain level.
Machine Learning Approaches to Nonverbal Rapport
Public speaking skills are crucial in many professional domains, including business presentations, persuasive speeches in politics or teaching in schools or universities. In close collaboration with the developer, CARISMA applies the public speaking platform Virtual Orator to train our undergraduates and to investigate possibilities to support to enhance self-regulation and behavioral performance.
VR Platform for Public Speaking Training
In close collaboration with media industry CARISMA helps to develop and evaluate practical VR and AR applications for entertainment, training and therapy. We study user responses and outcomes of VR/AR usage along a variety of cognitive, emotional and behavioral measures, including enjoyment, stress, learning and others more.
In collaboration with InnateVR, a Michigan-based software company specialized in therapeutic VR applications, CARISMA contributes to the development and evaluation of a Virtual Exposure Therapy app for opioid use disorders. The application aims to support the management of opioid cravings. It will include simulations of typical use situations including social interactions with realistic avatars and provide biofeedback game elements to support stress management and self-regulation.
Immersive Social VR for Opioid Use Disorder treatment
The CARISMA Lab is partnered with researchers in the College of Education to produce a program called “Global Citizens VR.” The project amounts to a virtual class shared by children across the world. There are virtual reality classes which are augmented by a variety of VR activities. For example, in one session students in Cairo will show students in Lansing around their home towns, pointing out places of interest to them like their school, homes, or favorite places to eat. The anticipated outcomes are reductions in outgroup bias and increased empathy.
VR Global Citizens: VR in cross-cultural education
TECHNOLOGY-ORIENTED REALIZATION RESEARCH
CARISMA conducts technology-oriented research in close collaboration with Computer Science and Engineering as well as media industry. We implement prototypes of anticipated future communication and information technologies and explore its potential benefits and downsides before market entry thus providing feedback for optimization for technology developers and providers.
Hybrid-Avatar-Agent Technology (HAAT) combines the naturalness of human behavior displayed through avatars with an assistive agent technology that monitors communication flow and flaws in VR-based interactions. HAAT is supposed to be an assistive component in future VR-based communication media that helps to optimize social outcomes and overcome impediments through feedback or behavior modification.
Hybrid Avatar Agent Technologies
Emergency situations, such as earthquakes, fires, explosions that implying structural damagers to buildings are extremely threatening to occupants and impose high levels of stress to the emergency experts. Novel sensors technologies can provide both groups with relevant information regarding escape plans or rescue measures. Augmented Reality (AR) technology can be used to superimpose the physical environment with visualizations of relevant data. CARISMA helps to develop and evaluate appropriate visualization strategies using VR simulations of disaster scenarios as a testbed.
Augmented Reality Technology for Emergency Teams (Nazir)
The most challenging task for artificial intelligence components in autonomous cars is the prediction of human behavior, particularly of people outside cars, such as pedestrians and bicycle riders. CARISMA runs VR simulations of traffic situations and collects motion capture data to identify trajectories and dynamics in the human movement that can help to predict action continuation at an early stage of movement initiation (e.g. walking toward the curb). In collaboration with Computer Science machine learning algorithms will be applied extract the relevant signatures.
Human Behavior Prediction for Autonomous Cars
OUR RESEARCH FACILITIES
The CARISMA lab hosts a dual maker-based motion capture system (Optitrack, 24 camera Prime 16) which allows to capture the movement of up to four people in each of two separate rooms. Movement data can be streamed in real time from the capture software Motiv 2 to the animation software Motion Builder and attached to a diversity of different avatars carrying out the actors’ behavior in a shared virtual environment. Deeper insights into the psychological processes underlying immersive VR experiences are gained through mobile psychophysiological and neural measurements. These measures include Heart Rate, Galvanic Skin Response, Respiration (Neulog, IOM2) as well as multiple mobile EEG systems.
The Lab employs a variety of virtual reality systems, including consumer level products such as the Oculus Rift as well as FOVE and wireless HTC Vive, which both have integrated eye gaze tracking. Separate, dedicated motion capture systems track the participants’ headset and controllers through the Lab, and allow for free movement throughout the virtual environment. Additionally, the speed of the Lab’s graphical processing eliminates lag and prevents the nausea that has traditionally been associated with VR activity. Custom-made virtual spaces can be created in the Lab to suit a variety of purposes. We employ software tools, such as Unity, Autodesk and Vizard to create believable virtual worlds and lively avatar inhabitants. Our own software developments are focused on behavior analysis and simulation.