Sensory test gives fast results in natural setting

A new test developed by the University of Copenhagen allows sensory researchers to conduct rapid product tests under natural circumstances using larger groups.

The classic product-oriented approach employs a trained panel of experts to analyse sensory data, which runs the risk that experts and consumers could evaluate the same products very differently. It also requires a strictly controlled test environment that is artificial when compared to real-life circumstances.

Market researchers have long-held doubts that results obtained under laboratory circumstances are capable of evaluating consumers’ behaviour in real-life conditions. They are increasingly calling for testing to be conducted under the most natural conditions possible.

Led by Davide Giacalone, a working group from the Department of Food Science/Sensory Science at the University of Copenhagen - a member of the European Sensory Network - developed the All-In-One (AI1) test that combines a variety of new approaches and collects a large amount of consumer data quickly. The test gathers:

• demographic information concerning the consumer and his or her personal experiences in relation to the type of test product;
• consumer expectations of an ideal product;
• hedonic data on the degree of liking the test product;
• how the test person describes the products using a list of different descriptors.

The researchers invited 160 visitors to a Danish culture festival to sample six styles of beer. The participants answered a questionnaire including personal details (age, gender, education, etc) and their interest and knowledge in beer. They were also asked to complete a CATA (check all that apply) list to define the characteristics of their ‘ideal’ beer.

The participants were served six beer samples in neutrally coded glasses and asked to rate each sample on a seven-point scale. Using the same CATA list, participants then ticked off the taste nuances and sensory attributes that best described their opinion of how each sample tasted.

The study showed that participants found a marked difference between all six beers. Psychographic aspects - particularly participants’ knowledge of and interest in beer - indicated a stronger connection to taste preferences than did demographic characteristics.

“Collecting this information seems to be a useful addition to consumer tests, since it affords us a relevant consumer segmentation,” Giacalone said. “Concerning consumer product expectations, we can derive important information from the descriptive profiles of the ‘ideal’ beer, as well as compare these results directly with the consumers’ experiences during the product tasting.”

Unlike traditional approaches, the AI1 test produces a product sensory profile that is directly based on the consumers’ perception. Answering concrete CATA questions presents few cognitive challenges to consumers, while allowing them to describe their sensory perceptions.

“It offers the possibility of a quick screening in order to integrate the ‘voice of the consumer’ into the early stages of product development. It is flexible and can be adapted to special questions,” said Giacalone. “This instrument is especially suited for early stages of research in which new possibilities need to be identified. It increases the marketing orientation of sensory research.”