Integrated Sensory Response (ISR) Modeling: A New Methodology to Understand and Predict Sensory Attributes in Terms of Physical Properties

The goal of this study was to develop and evaluate a tool that can assist in physical understanding and predictability of sensory from physical properties. Intriguing results have been obtained of graphical and mathematical ways of relating physical and perceived texture properties of several categories of baked foods with and without crispy layers. This led to a new methodology called integrated sensory response (ISR) modeling that integrates the whole set of receptor and brain responses on physical stimuli, all essential for a given sensory attribute, by delivering a quantitative, multivariate, multiplicative Stevens model of the sensory attribute in physical terms: Sensory attribute = a × Phys1 b × Phys2 c ×…× Physn n . The multivariate approach allows description of the more complex sensory attributes in terms of instrumentally measured physical properties where one‐to‐one relationships fail. It also allows description of how rather different physical properties may lead to the same sensory score. ISR modeling does not derive models from an exact physical analysis and mathematical description of oral processes and physical stimuli, nor from a theoretically derived psychological sensation scale. Instead, it identifies a few physical parameters that appear to dominate sensory perception, in this case quantified in terms of texture attributes. The decision by (a segment of) consumers about their sensory perception of a given product set is the integrated result of many interactive, parallel, and consequential psychological, physiological, and physical processes. The physical parameters identified and validated as dominating a given sensory attribute are likely to characterize product effects in those physical processes that have dominated the quantified perception. This implies that other product properties and other physical processes are less relevant to study and optimize for the product category and consumer group investigated. Thus, a better focus in product development is obtained. ISR models provide understanding of a sensory attribute valid for a given set of products and panelists, not in terms of a complete mechanistic explanation, but in terms of knowledge about the relative importance of different physical properties and, through this, about the relative importance of different physical processes and mechanistic elements.