A series of products were evaluated independently by an experienced trained sensory panel and by consumers.
The two groups were intentionally very different. The trained panel was highly experienced and tested under controlled conditions, while the consumer sample represented normal consumer sensitivity.
Importantly, comparability did not depend on using the same test method. Each group could be assessed using the method most appropriate for its role and level of sensitivity.
To make the results directly comparable, sensory differences were expressed using d′, a standardized measure of perceptual difference derived from Signal Detection Theory.
This created a common measurement framework in which panel and consumer sensitivity could be compared, even when different test methods were used, and the relationship could be modelled.