New tools and techniques, often made possible by technology, are revolutionising the way in which we conduct research. We are keen to embrace innovative approaches which enhance insights and understanding for our clients. One area in which we have seen increasing interest is facial analysis. A wealth of emotional information can be drawn from our facial expressions, which can be very valuable for uncovering emotional insights

Does your expression give you away?

Emotional information can be assessed from our facial expressions. Brief involuntary flashes of emotion or ‘micro-expressions’ can reveal a lot about a person’s true feelings. For market researchers interested in the emotional responses driving people’s attitudes and behaviour, this information is invaluable but can often be difficult to obtain. One of the latest innovations in market research is facial analysis, an artificial intelligence tool that detects a range of emotions in people’s faces.

Behavioural scientists have concluded that stimulus materials that are designed to have an emotional impact are most likely to have a behavioural resonance but when asked
directly, respondents can often find it difficult to articulate how they really feel. Furthermore, the post-rationalisation of responses can lead researchers further from the truth. Facial analytics help to overcome this by capturing instinctive reactions through which responses can be better contextualised.

How it works

Artificial emotional intelligence or Emotion AI uses optical sensors to measure facial
expressions of emotion and cognitive states the same way people do. The technology works by identifying a human face, either in real time or in an image / video and, by using computer algorithms, identifies key facial landmarks e.g. the corners of the eyebrow, the tip of the nose or the corners of the mouth. Machine learning algorithms then analyse pixels in those regions to classifying
facial expressions. Combinations of these facial expressions are subsequently mapped
to a range of emotions.

When used in combination with traditional market research methods, facial analysis can deliver an additional layer of insight by uncovering the emotional reaction to stimulus, whether static or dynamic.

It is particularly useful for testing communications designed to evoke emotional responses such as product messages, creative concepts, disease awareness campaigns and digital resources.