Decoding CEO Faces
The basic concept of facial coding is that a trained observer can detect fleeting facial movements that indicate the true emotions that the subject is experiencing, even if the subject is trying to conceal those emotions. I’ve written about the process of interpreting these facial expressions in the past, notably in Facial Coding and Emotionomics. Yesterday, USA Today has picked up on the topic with a major story about the possible applications of different facial coding techniques in choosing investments:
So far on Wall Street such strategies have barely moved ahead of palm reading as an investment strategy. That might be changing. A paper called “The Face of Success” published in February’s issue of the journal Psychological Science found that students who looked at photographs of Fortune(TWX) 1,000 CEOs were able to identify the most successful. They knew nothing about the executives before looking at the photos, but used naive judgments to rate them on traits such as competence, dominance, likability and trustworthiness, says co-author Nicholas Rule, a psychology professor at Tufts University. [From USA Today – It’s written all over their faces.]
The news release from the Association for Psychological Science elaborates:
Despite the ambiguity of the images, which were cropped to the face, put into grayscale and standardized in size, ratings of power- and leadership-related traits from CEOs’ faces were significantly related to company profits.
“These findings suggest that naive judgments may provide more accurate assessments of individuals than well-informed judgments can,” wrote the authors. “Our results are particularly striking given the uniformity of the CEOs’ appearances.” The majority of CEOs, who were selected according to their Fortune 1000 ranking, were Caucasian males of similar age. [From Lasting Impression: Does the face of a CEO determine a successful company?]
The CEO study is quite different from the facial coding analysis practiced by firms like Sensory Logic. Still images apparently communicate something significant to the viewers, but contain a lot less information than a video of the same individual reacting to different questions and comments. The USA Today article has some interesting commentary from Sensory Logic CEO Dan Hill on several business leaders. You might be surprised by the positive/negative ratings (which indicate the percentage of positive expressions like social and true smiles vs. negative expressions indicating disgust, frustration, anger, etc.
- Warren Buffet (Berkshire Hathaway) – 69% Positive, 39% Negative
- Jeff Bezos (Amazon) – 51% Positive, 49% Negative
- Michael Dell (Dell Computer) – 47% Positive, 53% Negative
- Bill Gates (Microsoft) – 73% Positive, 27% Negative
- Steve Jobs (Apple) – 48% Positive, 52% Negative
- Rupert Murdoch (News Corp.) – 17% Positive, 83% Negative
- Donald Trump (Trump Organization) – 16% Positive, 84% Negative
- Larry Ellison (Oracle) – 0% Positive, 100% Negative
If you buy into facial coding, then Larry Ellison is proof positive that money can’t buy happiness. Despite a net worth of $23 billion, he couldn’t muster a single positive emotion in a taped interview.
“Ellison is really grim, nervous. He’s driven and determined. He shows anger and arrogance on his face,” Hill says. “I would not consider this an open person, the opposite of an Oprah. See how his eyebrows are pulled together. This is someone who is not very comfortable in his own skin.
Nike chairman Phil Knight, meanwhile, manages to keep emotionally positive even while being sandbagged by rogue filmmaker Michael Moore. According to Hill’s analysis, Knight actually seems to be enjoying the altercation with Moore, as evidenced by many true smiles which involve not only the mouth but the upper face as well.
Will facial coding become a key element of investment analysis? That’s hardly likely, but if you find that an executive is giving off a squirrely vibe, you might want to go with your gut and avoid the company. As with many areas of neuromarketing analysis, we look forward to more controlled research that validates the conclusions of techniques like facial coding.