Understanding the sensory-instrumental nexus of walnut kernels: A dissertation submitted in partial fulfilment of the requirements for the Degree of Master of Science in Food Innovation

Walnuts (Juglans regia L.) are valued for their nutritional profile, sensory attributes, and economic significance. In New Zealand, smaller walnut growers face challenges in accessing professional sensory evaluation services due to high costs. There is a need for low-cost, easy-to-use sensory methods that can provide meaningful insights into walnut sensory attributes and physicochemical properties, both of which are key drivers of consumer acceptance. Rapid profiling techniques, such as Napping®, offer a practical alternative for characterizing sensory differences efficiently. This study examines the effectiveness of Napping® as a rapid profiling method for characterizing the sensory profile of nine commercially available walnut brands in New Zealand. The sensory data was integrated with physicochemical analyses of peroxide value (PV), acid value (AV), colour (L, a, b*), moisture content, and texture properties to establish the sensory-instrumental nexus of walnuts. Results show that Napping® is an efficient and applicable method for walnut kernel products, effectively differentiating sensory attributes associated with overall liking and consumer clustering. High liking scores were associated with fresh, nutty, creamy, and sweet attributes, while low liking scores were linked to astringent, rancid, sheen, and bitter characteristics. Consumer clustering was influenced by age, with preference patterns varying across different demographic groups. Instrumental analysis revealed that higher fat (>66%) and protein (≥15.3g/100g) content, moderate moisture levels (3.1-3.4%), lighter colour (L ≥ 41, b ≥ 7.5), and softer textures (hardness ≥ 7000g; chewiness ≥ 150g; gumminess ≥ 350g) were associated with higher liking scores. Walnuts with elevated PV (>14 meq O₂/kg fat), AV (>2.85 meq O₂/kg fat), carbohydrate content (>13.5g/100g) and firmer textures exhibited lower consumer acceptance. PLS-R modelling confirmed statistically significant associations between sensory attributes and physicochemical properties. These relationships confirm the presence of a sensory-instrumental nexus in walnut kernels where physicochemical properties help explain consumer sensory perception. These insights offer a cost-effective sensory evaluation approach, supporting small walnut growers in optimizing quality and targeting consumer preferences, with applications in post-harvest handling and marketing.