An Interdisciplinary Approach to Unmasking Emotional Behaviors for Better Understanding Emotional Intelligence
Region
Finland
Researcher
Guoying Zhao
Institution University of Oulu

Goal

Emotions are a central part of human communication, play an important role in everyday social life. We cannot communicate well without understanding the other's feelings. Knowing what other people feel is an important element of everyday social interactions and for people's emotional well-being. Emotions are complicated. Sometimes people intentionally express their emotions to help deliver messages and sometimes people would suppress and hide their emotions for different reasons. So, emotion should have a key role in new human-centered HCI and computer mediated human-human interaction, that enables computers to understand human emotions alongside with other tasks, e.g., speech recognition, object detection, face recognition, to lead the interactions into another level with high intelligence.
The goal is to unmask emotional behaviors (e.g., facial expression, body gestures, voice, physiological reactions), understand and better computationally model the mechanism of humans when express or suppress emotions, including e.g., how our brain works, what are authentic behaviors and what are fake ones, how authentic/fake behavior works with each other, how people co-regulate, across neuroscience, psychology, social behaviors and computer sciences, for high level interactions with taking into emotion intelligence into account. This will fit well to the human's affective and social aspects of human flourishing.

Opportunity

Most previous works and tools about emotion recognition from behaviors focused only on basic problems such as expressed facial expression and body gesture analysis, while very few works ever concerned about the various behaviors including both authentic and fake, of emotions. Expressions of emotions in human behaviour has remained poorly understood mainly because research targeting emotions has been fragmented into different scientific disciplines. We now have the opportunity to combines expertise on affective computing, machine learning, psychology, cognitive and affective neurosciences and learning sciences for completing research and theory on emotions as part of human mental capacity.

Roadblocks

The main challenge would be various understanding of "emotion", as there is no unifying concept across the fields. People coming from different disciplines have different understanding. Without confirmed consistent understanding, any joint efforts over different fields would easily break down.
Another challenge is limited data sharing. It is common that one group in one field just collect data for their own use, and it causes repetitive work and make the experiments and conclusion based on totally different data hardly compare and integrate.
Another challenge is imbalanced development of different disciplines, thus causing challenges in effective collaboration among different areas.

Breakthroughs Needed

Summer schools, conferences and hackathons to effective exchange knowledge for reaching agreements in key concept understanding and data analysis.
Methods and data should be shared openly in a "data & methods home" to allow for a fair analysis. Challenges based on shared data will be organized to advance not only data & methods sharing, but also disciplines development.
Proof-of-concept grants for collaborators from at least two different fields can be supported.
More investigation to the less developed disciplines supported by other disciplines in the beginning and later large grants for collaborative projects.

Key Indicators of Success

3 years: Are summer schools, conferences, challenges and hackathons well attended? Did the call for proof of concept lead to competitive projects? (Any 'no' would indicate a failure).
5 years: Were the less developed disciplines recognized and supported? Have the data and methods are shared frequently? ('Yes' to both would indicate success but a 'no' in the latter one may not necessarily be a failure.)
10 years: Were the consortium projects participated by strong teams with world leaders in different fields? Did any grants lead to new useful discoveries in at least two disciplines? (Any 'no' would indicate a failure).

Disclaimer

These research ideas were submitted in response to Templeton World Charity Foundation’s global call for Grand Challenges in Human Flourishing, which ran from September through November 2020.

Opinions expressed on this page, or any media linked to it, do not necessarily reflect the views of Templeton World Charity Foundation, Inc. Templeton World Charity Foundation, Inc. does not control the content of external links.