Implementação do FaceReader

: estudo piloto

Abstract

Sensory science considers the impact of emotional reactions on sensory attributes. FaceReader is an emerging technique that consists of facial analysis software and uses algorithms to identify and analyze facial expressions. This technique can identify facial expressions such as joy, sadness, disgust, anger, fear and surprise, as well as the neutral state. The objectives of this internship report are: to describe the methodology and select the pilot study matrix for implementing FaceReader technology; to learn about facial expressions measured with Noldus technology; and to describe the activities carried out during the internship period at Sonae MC. A bibliographic survey was carried out in the main databases with the descriptors ―face reader‖ AND ―food‖; a study was carried out on the application of the FaceReader method proposed by Noldus Information Technology and through consultation of technical materials made available by the company, in this way the methodology to be used in the Sensory Analysis laboratory was constructed. To implement the method, the target population, inclusion and exclusion criteria, limitations, ethical aspects, statistical analysis, limitations and data collection instrument were considered. The matrix selected was yogurt, because it has different flavors and provoked facial expressions. Discrimination tests were carried out with 4 trained tasters. The tasters selected 4 samples that presented the greatest sensory difference between them, characterizing the sweet, bitter and acid taste. A total of 161 people took part in tests using the hedonic scale with different types: cheeses, yogurts, desserts, fruit, bakery, sauces, among others. In addition, participation in physicochemical tests complemented the sensory tests. In conclusion, the pilot study to implement FaceReader is an important preliminary stage and a promising tool for analyzing emotional reactions accurately in consumer research. The activities carried out made it possible to acquire technical and scientific skills in the field of sensory analysis.

Date of Award	16 Jul 2025
Original language	Portuguese
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Sara Margarida Bernardes Queda da Silva (Supervisor)