How to measure emotional responses in political debates

Wednesday, 17 October, 2012

Updated! Now including the analysis of the third and final presidential debate!

Interested in the presidential elections in the US? And would you like to know more about emotion analysis? Read this blog and learn how facial expressions can affect voters, according to Chris Kowal, Purdue University, who is interested in communication of emotions.

Three presidential debates

Presidential debate

Basic emotions, represented by facial expressions, are not conscious and can thus deliver great information about subconscious processes. Prof. Chris Kowal analyzed the first presidential debate using FaceReader software. What did he measure on Romney's face? Which emotions did he measure on Obama’s face? And, will the show of negative emotions influence the undecided and maybe angry voters? 

Second presidential debate

Obama expresed a wider rage of emotions, Chris Kowal explains. The second presidential debate is now also analyzed. 

UPDATE 24-10-2012
Third and final presidential debate

Romney expressed anger and disgust when he mentioned, or led up to mentioning, George Bush and the auto industry, Chris Kowal explains. FaceReader reads facial expressions from the face and classifies them as anger, disgust, scared, happy, sad, neutral, and surprised. Kowal then explains that when Obama listened to Romney talk about this subject, he expressed contempt, anger, disgust, and sadness. "Obama was much more emotionally intense this time", said Kowal. Read an interesting article on ABC News of the presidential debate decoder.

Other debates

Vice presidential debate

FaceReader software was also used to analyze the vice presidential debate between Joe Biden and Paul Ryan. According to Chris Kowal, Paul Ryan expressed disgust and sadness, particularly when Biden talked about abortion. Read more on ABC News.

Mustache hinders ability to connect to voters 

In the state of Indiana, a governor election debate was followed in great detail. One candidate, John Gregg, wears a mustache which has become a trademark in his campaign. Although recognizable, Chris Kowal indicates that it is withholding the candidate to connect on an emotional level with his voters. 

The technology

FaceReader software 

Researchers worldwide measure emotions by using FaceReader, software that automatically classifies facial expressions. Facial recognition software is used to automatically detect and analyze facial expressions (six basic emotions). By using this software, facial expressions are classified directly in one of the following categories: happy, sad, angry, surprised, scared, disgusted, and neutral. These emotional categories have been described by Ekman (1970) as the basic or universal emotions. Facial expressions can vary in intensity and are often a mixture of emotions.


FaceReader methodology white paper

Request the FREE FaceReader methodology note to learn more about facial expression analysis theory.

  • Learn what FaceReader is and how it works
  • Learn how the calibration works
  • Get insight in quality of analysis and output