Individual variation in taste perception : investigating sensory perception in thermal tasters and its impact on food choice behaviour : thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Food Technology at Massey University, Palmerston North Campus, New Zealand

Abstract

Thermal taste is a phenomenon whereby some people experience taste from thermal stimulation in the absence of a chemosensory stimulus. Thermal tasters (TT) differ from thermal non-tasters (TnT) in orosensory sensitivity but are also thought to differ from TT in behavioural aspects, including food choice. In addition, TT are hypothesised to differ as a group in their thermal taste experiences. However, traditional phenotyping methods have failed to establish differences in food choice due to their inefficiency (a large proportion of participants are unclassified). In addition, traditional methods are incapable of collecting detailed data on the thermal taste response, as responses are recorded after thermal exposure.

This study developed a new phenotyping method, RapCoTT, which comprises a singular thermal exposure protocol, combined with TCATA data collection. When validated compared to existing methods, this approach was found to be superior in effectiveness and efficiency, and to agree with existing approaches. From data collected through RapCoTT, TT clusters were established, each reporting different thermal tastes and reporting tastes at different temperatures. When modelling thermal taste response data, further evidence was found for the proposed link between TRPM5 and sweet thermal taste. In addition, several novel mechanistic links were discovered, including several potential pathways involving TRPA1.

A group of participants with known thermal taste response groupings were profiled for a range of demographic, food behavioural, personality and orosensory responsiveness variables. Initially, links were explored between TT Status and TT Cluster and these variables, and a machine-learning approach was used to establish the relative importance of predictors. Overall, links between TT Status and TT Cluster and dimensions of the Big Five Inventory (personality) were established, and TT and TnT were found to differ in their salt and spice addition behaviours. Evidence was also found for several other factors impacting TT Status, including food neophobia, Private Body Consciousness, oral processing behaviour, and aspects of spicy food liking. TT Clusters differed in their hedonic response patterns to sucrose (Sweet liker status clusters). The Set & Disposal subscale of the Food Involvement Scale discriminated TT Clusters.

Finally, food choice behaviour in the form of food frequency data was modelled from thermal taste, demographic, food behavioural, personality and orosensory responsiveness variables. Overall, TT Cluster was the most important variable set related to thermal taste in determining food choice, indicating that not only TT Status, but also the temperature and nature of thermal taste perceived affects food consumption. TT Cluster also affected the specific consumption frequencies of the most items. When clustering food items based on the impact of TT Cluster, items did not group according to predetermined categories, but some textural patterns were apparent.

Overall, this work sets the scene for a paradigm shift around thermal taste, its interaction with other factors and its influence on food behaviour as a whole.