Phil’s Findings #2: More Tasting Notes

31 May 2011

Phil’s Findings #2: More Tasting Notes

Phil Howes, the Institute of Making’s post-doc researcher, shares his thoughts and findings as he delves into the sensoaesthetic world of materials. Check this space every Monday for Phil's latest posting.

In my last post, I left you with the question ‘is the basic tastes model too simple?’ Well, some people would argue that yes, the four (or five) basic tastes do not seem enough to be able to encompass all possible tastes. The concept of basic tastes relies on what is called the ‘labelled lines’ model. The taste buds in your mouth are composed of up to 100 taste receptor cells, and in the labelled lines model it is supposed that each one of these cells has a specific taste that it can detect, and that when stimulated it sends a nerve signal directly to the brain along a specific line (or neuron). For example, if a sweet receptor is stimulated by sugar, the cell sends an electrical signal along the ‘sweet pathway’ to your brain, and you experience the sweet taste. The same goes for bitter, sour and salty, the point being that each sensation has a dedicated ‘labelled’ line to your brain. However, within science research there is an on going conflict between people who subscribe to labelled lines, and people who support an alternative called the ‘cross-fiber’ model. This model suggests that taste cells are broadly tuned to a range of different tastants much like the cells in your eye are sensitive to a range of colours of light. This would mean that taste sensations are created by patterns of stimulation over populations of cells, producing a ‘taste code’ which is then deciphered in your brain. There would therefore exist a continuum of tastes rather than the distinctly separate basic tastes, and the electrical signals sent to the brain would not have a dedicated pathway but share a common path, with the coding allowing us separate to separate one taste sensation from another.

When I started my research into taste, I didn’t realize there was even a debate here, and I just thought it to be fact that there is a distinct set of basic tastes. However, it does seem that the jury is out on the issue of taste, and it is not as well understood as one might imagine. Considering the complexity of cross-fiber model, one can appreciate the attraction to the basic tastes and labeled line models, however it might just be that things are more complicated than we all thought.

An important part of my work at the Institute of Making will involve using a statistical modelling method called ‘multidimensional scaling’. I will spare you an exhaustive explanation (for now, I may revisit it later!), but briefly, MDS allows researchers to collect lots of information (size, weight, etc) on a set of objects and plot it all on a map which shows you the overall similarities between the objects. This is useful for uncovering patterns in similarity which might not be immediately obvious from simple observations. So, as part of my research into the senses, I diligently searched for articles on taste and multidimensional scaling. You can imagine my delight when the search returned articles with titles like “Impact of flavor attributes on consumer liking of Swiss cheese”, “Preference by sheep and goats among hay of eight tall fescue cultivars” and “From asparagus to zucchini: Mapping cognitive space for vegetable names”. It seems that nothing is safe from scientific scrutiny...