Writ Large on Your Face: Observing Emotions Using Automatic Facial Analysis

Open access

Abstract

Emotions affect all of our daily decisions and, of course, they also influence our evaluations of brands, products and advertisements. But what exactly do consumers feel when they watch a TV commercial, visit a website or when they interact with a brand in different ways? Measuring such emotions is not an easy task. In the past, the effectiveness of marketing material was evaluated mostly by subsequent surveys. Now, with the emergence of neuroscientific approaches like EEG, the measurement of real-time reactions is possible, for instance, when watching a commercial. However, most neuroscientific procedures are fairly invasive and irritating. For an EEG, for instance, numerous electrodes need to be placed on the participant's scalp. Furthermore, data analysis is highly complex. Scientific expertise is necessary for interpretation, so the procedure remains a black box to most practitioners and the results are still rather controversial. By contrast, automatic facial analysis provides similar information without having to wire study participants. In addition, the results of such analyses are intuitive and easy to interpret even for laypeople. These convincing advantages led GfK Company to decide on facial analysis and to develop a tool suitable for measuring emotional responses to marketing stimuli, making it easily applicable in marketing research practice.

References

  • Ekman, P. and Friesen, W. (1978): Facial Action Coding System: A Technique for the Measurement of Facial Movement. Palo Alto: Consulting Psychologists Press.

  • Scherer, K. (2005): “What Are Emotions? And How Can They Be Measured?” Social Science Information, Vol. 44, No. 4, pp. 695 - 729.

  • Garbas, J.-U.; Ruf, T.; Unfried, M. and Dieckmann, A. (2013): Towards Robust Real-time Valence Recognition from Facial Expressions for Market Research Applications. Proceedings of the 5th International Conference on Affective Computing and Intelligent Interaction, pp. 570 - 575.