Development of CATATIS Feature with Givaudan

This document is a technical report detailing the collaboration between Givaudan, Oniris, and XLSTAT to enhance the CATATIS feature for analyzing performance in CATA experiments. The report outlines the challenges faced when traditional CATA analysis methods were applied to trained panel data, particularly regarding the repeatability and consistency of the panel's evaluations. It describes how Laure Bonnet, a sensory project manager at Givaudan, conducted research on fish-flavored products using CATA tasks and identified issues with data reliability. The collaboration led to the development of innovative methods integrated into the CATATIS feature, allowing users to perform analyses that account for different sessions and assess inter-subject repeatability. The report presents the results of the study, which demonstrated improved data reliability and consistency across sessions, and highlights the significance of the findings at the Eurosense conference in September 2022. The document emphasizes the importance of understanding panel homogeneity and the potential misunderstanding of specific attributes during evaluations.