Have you ever not understood your doctor but were either too shy, too confused, or just didn’t know what to ask to get better clarification?

This study seeks to leverage advances in natural language processsing to analyze a collection of 383 conversations between oncologists and their late state cancer patients discussing prognosis and treatment options in order to develop ways to improve doctor-patient communication.

Through a series of experimental studies in paired communication, several novel models of facial expression and multimodal communication are being developed to help identify characteristics which help discern honest and deceptive commuinication.

The ubiquity of web cameras, cloud computing, along with recent advances in computer vision and crowd-sourcing are enabling are enabling profound improvements in communications research. We have developed an online framework for investigating the possibility of using online web camera screening for a variety of tasks from identifying undiagnosed Parkinson's Disease, to evaluating the trustworthiness of an interviewee.

Most people have 43 muscles in their face. If we consider just two muscle contraction levels (i.e. contracted or relaxed) there are over 2⁴³ (>8 million million) possible faces one can make! In order to represent facial expressions in a more meaningful and numerically tractable way, several encoding methods are being investigated.

Several studies have established that nonverbal communication, including facial expressions, pose, and speech audio characteristics, often account for a substantial portion of the meaning conveyed.Yet, a fundamental question remains: How are the different modalities of communication, including textual, auditory, and visual, interdependent in affecting a communication's effectiveness?

This study seeks to answer this question as well as many others. Click to find out more!

Imagine an app on your phone or computer that can tell if you are being dishonest, just by processing affective features of your facial expressions, body movements, and voice. People could ask about your political preferences, your sexual orientation, and immediately determine which of your responses are honest and which are not. In this study, we investigate the likelihood that artificial intelligence-based, non-invasive lie detection technologies are likely to experience a rapid advancement in the coming years, and examine the implications such an advance may bring.