Challenges in computational modelling of food breakdown and flavour release

Abstract

A dynamic, three dimensional (3D) computational model that predicts the breakdown of food and the release of tastants and aromas could enhance the understanding of how food is perceived during consumption. This model could also shorten the development process of new foods because many virtual foods could be assessed, and discarded if unsuitable, before any physical prototyping is required. The construction and testing of a complete 3D model of mastication presents many challenges including an accurate representation of: the anatomical movements of the oral cavity (including the teeth, tongue, cheeks and palates), the breakdown behaviour of the food, the interactions between comminuted food and saliva as the bolus is formed, the release and transport of taste and aromas and how these physical and chemical processes are perceived by a person. These challenges are discussed in reference to previous experimental and simulation work and using results of new applications of a coupled biomechanical-smoothed particle hydrodynamics (B-SPH) model. The B-SPH model is demonstrated to simulate several complicated aspects of mastication including: (1) the sensitivity of particle size to changes in the movements of the jaw and tongue; (2) large strain behaviour of food due to softening by heating; (3) interactions between solid and liquid food components; (3) the release of tastants into the saliva; and (4) the transport of tastants to the taste buds. These applications show the possibilities of a model to viably simulate mastication, but highlight the many modelling and experimental challenges that remain.

Physics of food structure breakdown and bolus formation during oral processing of hard and soft solids

“…Harrison et al suggest that extending this work beyond comminution to bolus formation is M a n u s c r i p t restricted due to the lack of information on the effect of saliva on the mechanical softening and lubrication of food particles, which is complicated by the continual addition of saliva to the oral process [36].…”

“…Attempts are underway to model oral processing of foods using smoothed particle hydrodynamics in order to simulate the action of the major oral surfaces upon manipulating fluids, brittle foods, elastoplastic foods and elastobrittle foods [35][36][37]. The models have so far been able to provide a qualitative fit with the comminution of ideal agar gel samples [37], with the insight that the breakdown path of the food is altered by the starting position of the food, which is indicative of a change in the selection function due to food position [28].…”

Modelling food digestion

“…This approach was first used by Potapov et al (2001) for simple shear flow in two dimensions. This approach is ideally suited for human digestion modelling since significant detail around and within each piece of food typically needs to be resolved (Harrison et al, 2014b). This approach is ideally suited for human digestion modelling since significant detail around and within each piece of food typically needs to be resolved (Harrison et al, 2014b).…”

“…Predicted size distributions of food fragments for an agar model food also have been satisfactorily evaluated against measurements from chew-and-spit experiments (Harrison et al, 2014a). Mastication of inhomogeneous foods, the release of tastants into saliva and temperature-based softening also have been modelled (Harrison et al, 2014b). Although much more sophisticated than earlier models, this model addresses only a subset of what is ultimately required for use in the design and optimisation of foods with superior health performance and customer acceptance.…”

“…Mastication of inhomogeneous foods, the release of tastants into saliva and temperature-based softening also have been modelled (Harrison et al, 2014b). Future developments required for this model are outlined in detail by Harrison and Cleary (2014) and Harrison et al (2014b). Future developments required for this model are outlined in detail by Harrison and Cleary (2014) and Harrison et al (2014b).…”