Research on the construction method of kansei image prediction model based on cognition of EEG and ET

• Published in: International journal on interactive design and manufacturing Vol. 14; no. 2; pp. 565 - 585
• Main Authors: Yang, Mingqing, Lin, Li, Chen, Zhiang, Wu, Ling, Guo, Zhuen
• Format: Journal Article
• Language: English
• Published: Paris Springer Paris 01.06.2020; Springer Nature B.V
• Subjects: Algorithms; Artificial intelligence; Bioelectricity; Brain; Brain research; CAE) and Design; Cognition; Cognition & reasoning; Computer-Aided Engineering (CAD; Consumers; Decision making; Electroencephalography; Electronics and Microelectronics; Emotions; Engineering; Engineering Design; Feedback; Industrial Design; Instrumentation; Measurement techniques; Mechanical Engineering; Morphology; Original Paper; Performance evaluation; Physiology; Prediction models; Product design; Qualitative research; Questionnaires; Research methodology; Semantics; Smartphones; User experience; Kansei image; Electroencephalogram (EEG); Product form evaluation; Eye tracking (ET); Kansei engineering
• ISSN: 1955-2513; 1955-2505
• DOI: 10.1007/s12008-020-00651-2

In order to more accurately predict kansei image of user when they interact with a product, and help designer to design a product that meets user’s emotional appeal from a user-centric perspective, a method for constructing kansei image prediction model from a more objective perspective of eye–brain physiological cognition was presented in this paper. Firstly, deconstructing product shape to obtain form design elements, and then recombining those elements in different ways to get experimental samples. Next, EEG and ET techniques were introduced to obtain bioelectric data of eye–brain physiological cognition when subjects view the experimental samples, and a space containing user kansei needs was established. Finally, the kansei image value of the product, the form design elements, and eye–brain physiological data were associated together to establish a kansei image prediction model. To verify the effectiveness of this model, smartphones were used as validation cases in this research. User’s kansei prediction model for smartphones was established based on subjects’ evaluation of 30 experimental samples, and the model’s performance was evaluated by a comparison verification method. The research results showed that the kansei image prediction model based on eye–brain physiological cognition can more accurately predict kansei image in the process of user interaction with product, thus helping the designer to find out the key product from that trigger kansei image, and design a product that was more in line with the user’ kansei experience.