Twelve panelists out of the twenty recruited were preselected
using a difference-from-control test relating to crispness. The three
cylindrical snacks, among which one of them was standardized as a
control sample based on the instrumental analysis, were subjected
to the difference-from-control test. The panelists were preselected
according to their discriminative capacity (Fsample 6 0.50) and
reproducibility capacity (FrepetitionP0.05) (ASTM, 1981).