Temporal Drivers of Liking Based on Functional Data Analysis and Non-Additive Models for Multi-Attribute...

Area of Science:

• Sensory Science
• Consumer Behavior
• Data Analysis

Background:

• Traditional liking analysis often overlooks the time-dependent nature of sensory perception.
• Understanding how attribute perception evolves over time is crucial for product development.

Purpose of the Study:

• To extend liking analysis by incorporating a time dimension, developing temporal drivers of liking (TDOL).
• To explore the dynamics and relative importance of attributes over time in influencing consumer liking.

Main Methods:

• Utilized functional data analysis (FDA) methodology.
• Applied non-additive models, including Choquet integral and fuzzy measures, to multiple-attribute time-intensity (MATI) data.
• Explored TDOL dynamics using derivatives of relative importance functional curves with R packages ('fda', 'kappalab', 'relaimpo').

Main Results:

• Developed and applied TDOL to analyze MATI data.
• Demonstrated that the relative importance of MATI curves reveals temporal aspects of consumer liking.
• Identified direct and interaction effects of attributes on overall liking over time.

Conclusions:

• TDOL offers a novel framework for analyzing the temporal dimension of consumer liking.
• The methodology provides valuable insights into how sensory attributes influence preference dynamically.
• This approach enhances understanding of consumer responses to time-varying product characteristics.