All the samples were transferred to plastic dishes. Samples were coded with three-digit ran- dom numbers and judges ranked the samples from most preferred to least preferred according to Boverall mango flavor^ preference. Samples to be evaluated for taste preference were presented randomly under red light to reduce the effect on difference of appearance. Judges also indicated one of five different pre-defined options of frequency of consumption of any type of fresh or processed mango products. Subsequently, judges evalu- ated visually a different set of whole starch-molded mango snack formulations presented randomly under natural lighting to define preference according to Boverall appearance^ by ranking coded mango snack formulations from most to least preferred.