Prediction of residual film perception of cosmetic products using an instrumental method and non-biological surfaces: The example of stickiness after skin application.

Prediction of sensory texture attributes using instrumental measurements is a very important challenge for cosmetic industry because in vivo sensory studies are expensive, time consuming and limited by the safety issue of applied products. The aim of this work is to investigate how residual sensory properties of cosmetic products can be predicted without using a panel of assessors, focusing on the residual film attribute "Stickiness". 10 cosmetic products with different galenics have been selected and evaluated in vivo using a classical sensory protocol, developed according to the Spectrum™ Descriptive Analysis method. In addition to this study, products were evaluated after their application on non-biological skin models in order to compare perceptions onto in vivo skin and artificial surfaces. Results obtained show that in vivo perceptions can be compared with the ones on artificial surfaces meaning that residual film stickiness is similar between in vivo skin and non-biological skin models. An instrumental protocol using a texture analyzer has been set up to evaluate residual film adhesiveness. This protocol has been tested and validated in vivo (r²adjusted = 0.90; RPD = 3.07) before being optimized on a selected non-biological skin model Bioskin® (Beaulax, Co. Ltd. Tokyo, Japan) owning good correlation with in vivo perceptions. Established model shows excellent predictive ability with a r²adjusted of 0.94 and a RPD of 3.38, as highlighted by the 4 steps cross-validation performed. It proves that physical stimulus responsible for cosmetic film stickiness can be instrumentally measured on both in vivo skin and artificial skin.