Development of a sensory protocol for testing palatability of sheep meats

A total of 108 grill and 108 roast samples were prepared from lamb (n = 10) and mutton (n = 8) carcasses for sensory testing using a consumer taste panel. Grill and roast samples were prepared from the left and right sides of the carcass, respectively, using 'longissimus', 'biceps femoris', 'gluteus medius', 'serratus ventralis' and 'semimembranosus'. Due to size constraints, muscle from both sides was used to form grill samples for the 'vastus lateralis', and roast samples from the 'triceps brachii'. Grill and roast samples were sensory tested using 360 untrained consumers. Each consumer was given a total of 6 experimental samples and each sample was tested by 10 different consumers. Sensory scores for tenderness, juiciness, like flavour and overall liking from both the grilled and roasted samples were highly correlated (P<0.05). For grilling, the different muscles were correlated for tenderness and overall liking scores (P<0.05), with the exception of the 'semimembranosus' (P>0.05). In contrast, juiciness and like flavour scores were poorly correlated between grilled muscles. For the roasted samples, sensory scores were generally uncorrelated between muscles. The statistical significance of the age category and muscle effects was greater in grill samples, but stimulation effects were of similar significance using either cooking method. It was proposed that roasting reduced treatment effects that affected sensory via differences in connective tissue toughness, due to gelatinisation of connective tissue during cooking. For testing production and processing effects on palatability, grilling was more sensitive for detecting treatment effects, than roasting.