Colour me .... minty.

A recent article in the UK’s The Guardian newspaper (Friday, 25 October 2013) was a potent reminder of the power of expectations in determining food likes and dislikes. The paper sent a camera crew armed with samples of a new flavor of Pringles potato crisps to observe the reactions of the public in a “taste test” ( The facial expressions shown in response to the new flavor are unambiguous, and almost all highly negative. Why? Well, its wasn’t that the flavor itself was vile – no earwax or stale fish – but merely an unexpected …. mint chocolate! One participant, close to disgust, even commented that this was his favourite flavour, but just not in a potato crisp.

This demonstration reflects very well how expectations are set up by past experiences, in particular, experiences with combinations of sensory properties. This includes not just tastes, odours and textures, but especially those qualities – visual cues such as shape and colour – that set up a
template of what something is about to tastes like. In the case of potato crisps, this might include flavours such as salty potato (ie., plain), “chicken”, vinegar or even more exotic combinations such as lime and pepper or chilli - savoury flavours all. But what induces the extreme reactions to what is, after all, a crispy, choc-mint treat? One possible explanation lies in considering that the predictability of a food’s flavour, based on external, visual cues (at least for visually-dominated species like us and the birds) is an important key to survival in the wild. In this sense, violations of expectations are a warning signal and like many warning signals involving food, they are underpinned by an unacceptable taste experience that is likely to inhibit eating.

This is a cognitive component to our taste experience. And it essentially means that any combination of colours, odours, textures, tastes can be acceptable if together they are consistent with one another. For this reason, for example, we can accept white wines that smell like cat’s pee or cut grass or caramel or petrol.

Clearly a food’s colour provides us with the key information we need to make a decision about whether to sample further by smelling or tasting. Green bananas, yellow lettuce, and grey meat tend to be diverted before they get close to the mouth. But the next
payage on this alimentary motorway is our sense of smell which, at least when we are sniffing, can provide us with information about food at a distance, just as our vision does. This means not only that our senses of smell and vision are used to confirm the information provided by each other, but that particular qualities in each sense (odours; colours) become associated with one another. Hence, repeated pairings of specific qualities in each sense becomes over time a joint signal for the presence of something that can be eaten. Such learned congruency (belongingness) of different sensory signals has a parallel in the way in which specific tastes and odours combine as familiar food flavours and producing, as a results of this integration, odours that smell like tastes [1].

How colour combines with other food sensory signals perceptually or cognitively has been uncertain, although colour effects have received increased scrutiny recently (see for example, [2]). Does colour ‘merely’ act to set up expectations, or should we consider food/beverage colours as an integral aspect of the overall flavour? A just published review [3] by Debra Zellner (Montclair State University, USA) provides an overview, and explanatory model, of colour-odour interactions. Zellner draws together the various findings on odor discrimination and intensity, as well as odour pleasantness, and argues for perceptual – as opposed to response bias – effects of colour on odour in each case. In essence, she is saying that colour can influence odour in much the same way as odours can enhance taste intensity, and for the same associative learning reasons.

Zellner leaves open the possibility of innate (unlearned) responses to odour-colour combinations, but this seems unlikely. Perceptual learning is a flexible process precisely because environmental conditions may dictate the combinations that could signal something edible. Hence, appropriate colour/odour combinations for foods would be expected to vary cross-culturally, and this has been shown in a recent study involving varying drink flavour and colours [4]. One could imagine, however, that colour intensity and odour strength might be intrinsically linked, but this may be an example only of our more general ability to equate intensities across different sensory modalities.

One reason that there has been little consensus on how colour might influence odour perception is because of apparent inconsistencies in these interactions. Thus, while there is evidence that colour can enhance sniffed odour intensity, there are also findings that the intensity of ‘tasted’ odours – flavours – are suppressed by the presence of colour. On the face of it, this seems a bizarre finding. How could an unperceived quality (the colour in the mouth) produce any effect at all on the flavour? To account for this, Zellner invokes the idea of a contrast between the sniffed and the tasted odour. So, if the odour/colour combination is first experienced by sniffing/seeing, then an enhanced odour intensity is produced. However, once the odour source is in the mouth, then there is no further input from the colour, and the intensity is a function of the odour only. The suppression is thus relative, a comparison of the odour enhanced by colour and the same odour with the enhancement suddenly removed once the odour source is in the mouth.

In the light of the Pringles demonstration, it is not surprising that when colour and odour do not match, then the odour/flavour tends to be judged less favourably. An expected combination is also a familiar one that carries with it an implicit recognition that previous experiences are associated with a positive outcome. As with odour/taste combinations, repeated exposure not only produces an integrated perception of a flavour but also has hedonic consequences. Even if an odour/taste combination is not initially liked, the effects of exposure, and the positive effects of the nutrients in the food, produces a liked flavour over repeated pairings.

Intriguingly, colour/odour pairings might also generate an influence in the opposite direction – that is, odours might conceivably enhance colours. While if shown to be the case, this would no doubt have some food applications, those applications that immediately spring to mind relate to effects that could be produced environmentally or artistically. Perhaps, too, this could provide a perfect opportunity for aromatherapists and colour therapists to combine their ‘talents’, hopefully saving their poor consumers some money.


1. Prescott, J., Multimodal chemosensory interactions and perception of flavor, in Frontiers in the Neural Bases of Multisensory Processes. 2012, CRC Press. p. 691-704.
2. Dematte, M.L., D. Sanabria, and C. Spence,
Cross-modal associations between odors and colors. Chem. Senses, 2006. 31: p. 531-538.
3. Zellner, D.A.,
Color–Odor Interactions: A Review and Model. Chemosensory Perception, 2013. 6(4): p. 155-169.
4. Shankar, M.U., C.A. Levitan, and C. Spence,
Grape expectations: the role of cognitive influences in color-flavor interactions. Conscious Cogn, 2010. 19(1): p. 380-90.