How sensory sensitivity to graininess could be measured?

Considering the lack of the literature and the need for developing a valid method to measure the texture sensitivity, in this study, we investigated the individual sensitivity to discriminate among different levels of graininess. With this purpose, five samples of cocoa-based creams were prepared, by changing the refining time. Samples were first characterized in terms of particles size distribution, by means of laser diffraction. Then, 59 subjects evaluated the cream sample graininess intensity, by using gLM scales, and their liking, by using LAM scales. The graininess scores of each subject were fitted with a power model, already observed with the instrumental results, estimating both the power law exponent and the R2 coefficient, and using them as clustering parameters. Subjects were then clustered into three groups: high sensitivity; moderate sensitivity; low sensitivity. First, as it was hypothesized, results showed a significant difference between the three groups in terms of perceived graininess. Second, even though results showed a significant difference between the three groups in terms of perceived graininess, only little differences were found in terms of liking scores. Indeed, all the samples were equally liked for both the moderate and low sensitivity groups, whereas a significant trend was observed for the highly sensitive subjects who liked more the most refined samples. No significant relationships were found with age, and only a little trend was observed with gender: females seemed to be more sensitive than males. Texture attributes discrimination ability, as for example sensitivity to graininess, could affect food rejection or preferences and choice, but it is poorly investigated. This exploratory study, proposes a method to cluster consumers, based on their sensitivity. A statistical methodology has been developed to discriminate among consumer sensitivity levels. The results provide useful information about graininess sensitivity suggesting that the used methodologies could be applied to other texture properties resulting in a valid tool for the industry in the development and optimization of tailored new products.