Prediction of acceptable colour of cooking bananas (matooke) using instrumental and sensory measurements

Abstract

The colour of cooking bananas (Matooke) is a critical attribute for consumer acceptance and the uptake of newly developed hybrids. The objective of this study was to evaluate the potential of instrumentation to predict consumer-acceptable matooke colour. Instrumental and sensory measurements were used to assess the colour of twenty-three matooke cultivars. A Konica Minolta chroma metre, CR-400 was used to measure colour intensity of both the raw and cooked matooke pulp in terms of lightness (L*), greenness (a*), and yellowness (b*). Sensory evaluation was performed by 12 trained panellists at the Food Biosciences Laboratories, NARL-Kawanda to establish the sensory quality of the matooke. Correlations between instrumental and sensory parameters were performed using XLSTAT. Partial Least Square regression analysis (PLS-R) was used to assess the strength of the relationships between the instrumental and sensory attributes. Pearson’s correlation revealed that the instrumental yellowness (b*) of the raw pulp showed a positive moderate correlation with the instrumental yellowness (b*) of the cooked matooke (r = 0.56). Similarly, positive correlations were observed between instrumental yellowness (b*) and homogeneity of the yellowness (r = 0.50). Partial Least Square regression analyses showed moderate predictive models for preferred colour, R2=0.50, and overall liking, R2=0.53. PCA results showed that instrumental measurements can be used to select matooke with acceptable colour. Four hybrids; NAROBAN 4, M30, NARITA14, and NARITA 4 had similar colour attributes as the landraces in terms of both instrumental and sensory yellowness. Overall, the results showed possible predictability of consumer acceptable colour of matooke which suggests that instrumental measurements can be used to predict the consumer-acceptable matooke colour thus a tool for selecting hybrids with acceptable colour.