Predicting thickness perception of liquid food products from their non-Newtonian rheology

The sensory experience of "mouthfeel" significantly impacts food acceptance and consumer preference. Predicting mouthfeel attributes like thickness and creaminess is a major challenge for the food industry, often hampered by the complexities of oral processing. Previous attempts to link rheological parameters (shear viscosity, modulus) and tribological properties to sensory perceptions have relied on simplified models that fail to accurately capture the dynamic interactions of liquid foods within the oral cavity. These complexities include the non-Newtonian nature of many food products (viscosity dependent on shear rate), the complex flow patterns (elongational flow, lubrication) in the mouth, interactions with saliva, wetting properties, and the presence of particles. Further complicating matters is the potential for sensory confusion among panelists, for instance, between stickiness and cohesiveness. This study aims to overcome these limitations by developing a physically based model that accounts for the non-Newtonian rheology and the dynamic squeeze flow of liquid foods between the tongue and palate during consumption. The model will be used to predict perceived "thickness", a key mouthfeel attribute, and explore its relation to the Weber-Fechner law, which postulates a logarithmic relationship between stimulus intensity and perceived sensory strength. This law has been observed for other senses, but its applicability to oral thickness perception remains largely unexplored, particularly for shear-thinning fluids where viscosity changes with shear rate. The study focuses on low-viscosity, shear-thinning liquid bouillons to establish a foundational understanding, expanding the findings to include higher viscosity xanthan gum solutions to test the robustness of the proposed logarithmic relationship.

Existing research has explored the relationship between rheological properties and mouthfeel, but often with limitations. Simple models have been used to link viscosity at a single shear rate to perceived thickness, neglecting the non-Newtonian behavior and complex flow patterns in the oral cavity. Studies have considered aspects like shear thinning, but often with rudimentary incorporations or assumptions of a single characteristic shear rate. While the Weber-Fechner law, suggesting a logarithmic relationship between stimulus and perceived intensity, has been observed in various sensory modalities, its applicability to oral thickness perception, particularly for low-viscosity, shear-thinning liquids, has been largely unexplored. Previous attempts at modeling oral fluid dynamics have often simplified the process, neglecting the dynamic aspects of the tongue's movement and the complex rheological response of shear-thinning materials. This study addresses these limitations by incorporating a more comprehensive model, bridging the gap between rheological properties, oral fluid mechanics, and subjective sensory perception.

The study used 14 liquid bouillons with viscosities ranging from 1 mPa s to 1 Pa s, created both custom-made (Set 1) and using commercial bouillon soups (Set 2). The viscosity of the bouillons was carefully controlled through varying concentrations of xanthan gum, potato starch, corn starch, and salt (NaCl). Rheological properties were characterized using an Anton Paar MCR302 rheometer with a cone-plate geometry at 40 °C (mouth temperature). Flow curves were fitted to a power law model (σ = κγ̇ⁿ) to determine the consistency parameter (κ) and power law index (n). Extensional rheology measurements were also performed to assess viscoelasticity contributions. A sensory panel of 11-14 trained individuals assessed the perceived "thickness" of the bouillons using a 16-point category scale (0-15) in a sequential monadic blind test. Wetting properties were examined by measuring the contact angle of a drop on paper. To model oral processing, a dynamic squeeze flow model was developed, representing the oral cavity as two parallel plates (tongue and palate). The model considers the tongue's movement speed (V) and force (F_N), the gap between plates (h), and the non-Newtonian rheology of the fluid to calculate the shear stress (σ) on the tongue during squeezing. The initial gap (h₀) and contact radius (R) were determined experimentally. Finally, the model's predictions of shear stress were compared to the sensory panel's thickness scores, assessing whether logarithmic or power-law relationships best fit the data. Data from a previous study on xanthan gum solutions was also incorporated to extend the range of shear stress and confirm the findings.

The rheological analysis showed that the liquid bouillons exhibited shear-thinning behavior, well-described by the power-law model. The sensory panel provided thickness scores that showed significant differences between the samples. The dynamic squeeze flow model successfully predicted the perceived thickness of the bouillons, showing a strong correlation between the calculated shear stress on the tongue and the subjective thickness scores. The data revealed a logarithmic relationship between perceived thickness and shear stress, supporting the Weber-Fechner law. This logarithmic relationship held true even when the dataset was expanded to include higher viscosity xanthan gum solutions from a previous study. The range of calculated stresses spanned approximately one to two orders of magnitude. The good agreement between the model and the experimental data, without using adjustable parameters, strongly indicated that the rheology of the liquid food can accurately predict its perceived thickness. The shear-thinning behavior of the bouillons significantly influenced the perceived thickness, because of the importance of the shear rate in the determination of the viscosity.

This study provides a significant advancement in predicting the mouthfeel attribute of thickness in liquid food products. The findings establish a strong link between the non-Newtonian rheology of shear-thinning fluids and their perceived thickness, demonstrating the validity of the Weber-Fechner law in this context. The dynamic squeeze flow model overcomes limitations of previous models by accounting for the complex flow patterns and dynamic interactions during oral processing. The robust logarithmic relationship observed across a wide range of viscosities underscores the importance of considering non-Newtonian rheology in sensory science. The results have implications for the food industry, potentially enabling the design of food products with precisely tailored mouthfeel characteristics by controlling the rheological properties of their ingredients. Future research could focus on extending these findings to more complex food products (e.g., yield stress fluids) and incorporating additional mouthfeel attributes, refining the model to handle a wider range of food textures.

This research successfully linked the perceived thickness of liquid food products to their non-Newtonian rheology via a dynamic squeeze flow model. The observed logarithmic relationship between perceived thickness and shear stress confirmed the Weber-Fechner law for this sensory modality. This model offers a valuable tool for predicting and controlling the mouthfeel of liquid foods. Future work could focus on extending the model to encompass more complex fluids and other mouthfeel attributes.

The model assumes a simplified geometry for the oral cavity, neglecting the complex anatomy of the tongue and palate. While the study focused on low-viscosity shear-thinning fluids, further research is needed to validate the model's applicability to other fluid types and food products. The sensory panel involved a relatively small number of participants, and cultural differences in taste preferences might influence the results. Future studies should explore a wider range of viscosities and include other mouthfeel aspects such as creaminess and stickiness.