Renaissance Men Target Consumer Research

A Netherlands-based firm whose developers hit the headlines recently for a scientific analysis of the Mona Lisa's smile, is hoping to add to the toolbox of consumer researchers. Visual Recognition's software 'recognises' facial expressions to gauge the real feelings behind respondent comments.

As part of a pilot commissioned by food and consumer goods giant Unilever, 300 women (chosen because of their tendency to have more expressive faces than men) in six European countries were filmed as they ate ice cream, chocolate, cereal bars, yogurt and apples, and not surprisingly ice cream and chocolate produced the most 'happy' expressions. Research was conducted at universities, shopping centres, and city centers and cameras first recorded volunteers eating and then each one provided a 'posed' version of their expression for computer comparison.

Unilever commissioned Visual Recognition's software developers Theo Gevers and Nicu Sebe to run these tests after hearing about their experimental work to interpret the Mona Lisa's smile, which they had found to be 83% 'happy'.

Gevers explained that when we smile or frown, thousands of tiny facial muscles are used. The system's Emotion-Recognition Software (ERS creates a 3-D face map that identifies 12 key trigger areas such as the area around the eyes and corner of the mouth. A face-tracking algorithm then matches the movements to six basic expression patterns which correspond to anger, sadness, fear, surprise, disgust and happiness.

'I was happy when the testing was over,' said Gevers. 'Using the software on people eating was a challenge, something we would not have done in an academic lab. We didn't know precisely how well it would work, but it did.'

The software registered fewer smiling faces for healthy foods. Apples produced 87% neutral expressions, with Italians and Swedes showing disappointment when eating them, and yogurt prompted 'sad' expressions for 28% of Europeans.

Mandy Mistlin, Consumer Scientist at Unilever UK added: 'The software may eventually be used to test reduced-fat and -calorie ice creams to see if they maintain the pleasure principle.'

Visual Recognition was established by Gevers in 2006 to commercially exploit research from the ISLA Laboratory at the Universiteit van Amsterdamn. The organisation is online at www.visual-recognition.nl.

All articles 2006-23 written and edited by Mel Crowther and/or Nick Thomas, 2024- by Nick Thomas, unless otherwise stated.