Sensory test results were analyzed by a statistical significance template for triangle tests of similarity (Meilgaard,
Civille, & Carr, 1999; Quadt, Schwarz, & Schönberger, 2009) according on a binomial distribution. Parameters
for triangle method of similarity were defined with α-risk-level of 0.01 (probability of concluding that a
perceptible difference exists when 1 does not) and β-risk-level of 0.01 (probability of concluding that no
perceptible difference exists when 1 does) and pd-level of 0.3 (true proportion of population able to detect a
difference between samples). A pd-level of 0.3 was chosen because it equates a medium range of population,
which are able to detect difference between samples (Busch-Stockfisch 2003).