Presented at the 2014 SenseAsia Symposium in SingEx, Singapore.
Difference testing is one of the most commonly available tools for sensory scientists who are studying process and ingredient modifications for cost-saving, supplier replacement, or health-related initiatives. Despite its name, difference testing usually focuses on studying the similarity of products, i.e., their equivalence, rather than an actual difference. Because reformulated products will always be sensorily different, if even by just a small amount, the true question of interest is: Are the two products ‘close enough’ so that a change can be implemented without offending the consumer?
To answer this question, it is essential to establish the size above which a sensory difference becomes relevant. While this is a fundamental aspect of discrimination testing, it is usually not taken into account in sensory testing programs. This in turn often results in experiments with low statistical power and thus the failure of identifying critical differences before releasing a modified product in the market.
In this presentation, we will review several methods that can be used to establish the threshold above which a difference becomes too large. These methods include linking consumer and trained panel sensitivities, relating consumer preference tests with internal discrimination results, and using the same-different test to set the criterion of sensory equivalence. We will show how this information can then be used in tandem with a suitable testing protocol such as the tetrad test to estimate the sample size that will provide sufficient statistical power, or measurement accuracy, for future investigations. By taking these concepts into account, scientists will be better equipped to build reliable sensory discrimination testing programs.
Colleagues can download the full presentation here:
Not a Colleague? Click here to join for free!