We heard via the BBC earlier this week that Thailand’s National Innovation Agency has developed an electronic tongue and nose which is said to be able to judge the authenticity of a Thai dish. The instrument is calibrated to rate the appearance and flavour of three classic Thai dishes compared to selected standard recipes. Nine sensors in the instrument measure the balance of six flavours important for Thai food – sweet, sour, bitter, salty, savoury and spicy. Although named the ‘e-Delicious’, the machine cannot actually tell you if a dish is delicious or not. What it can do is provide a ‘taste-reading’ to see how close a recipe fares against the ‘ideal’ and, when put to the test, the machine could tell the difference between low cost versions of a recipe and one made by the most celebrated chef in Bangkok.
Seriously though, is the ‘ideal’ recipe not at least a little subjective? People have such diverse taste preferences that it seems infeasible that a machine can tell us how a dish fares vs. the ideal. And how about mouthfeel, or aftertaste? We know these are often key indicators of product performance, so it feels as though the e-Delicious falls short if it doesn’t provide any indication of how a dish performs in this respect. Not that there isn’t a place for the electronic tongue and nose; these instruments can be extremely useful for detecting taint compounds and off-flavours in an automated production environment which would be too fatiguing for human senses, and so do have a role to play. But these machines are only as good as the level of detection and accuracy of the sensors to identify specific compounds and the sample set used to calibrate them (rather like the accuracy and precision of a trained Sensory panellist).
I for one would still prefer and trust the output of a sensory panel than a machine when it comes to evaluating the smell and taste of food!