Abstract:

 

Historically, when using discrimination testing methodologies, decisions have been made using statistics and the hypothesis of no difference. Due to the high costs and time constraints of investigations involving consumers, companies typically conduct discrimination research internally using a trained or employee-based panel with the objective of predicting consumers’ behavior. The underlying principle behind such approach is that if the internal panel does not detect a difference, then consumers will not. Consequently, they will not prefer one sample over the other. In this webinar, we will outline the limits of this reasoning and illustrate situations where a non-significant outcome in a discrimination test can still result in a consumer preference.

​

How is this possible? What are the underlying principles of sensory science that can help us explain this phenomenon? What steps can be taken by the sensory scientist to avoid this position and improve the predictive power of a company’s internal testing?

​

Using examples we will illustrate how this situation arises and we will introduce theoretical tools to describe the underlying principles without which no satisfactory explanation seems possible. We will also show how these ideas can be used to link and better predict results conducted in different conditions, such as an ingredient supplier sending prototypes to a client who in turn conducts testing internally to decide whether to accept an ingredient replacement

​

This webinar is intended for a general audience of sensory professionals and graduate students. No detailed technical knowledge is assumed.

​

A digital recording of this webinar is available for purchase ($289).