Tuesday, 21 January 2014

Nothing to be sniffed at

In our earliest months of life, we explore the world not just with our hands but also our mouths. This drops off a little as we grow older and realize that this is a fast way to ingesting things we perhaps shouldn’t. But it is easy to forget the mouth is a key source of actively seeking tactile information just as much as it is taste and – indirectly – smell (flavour) information. Similarly, we seldom consider our own noses as organs of exploration even if we are aware that some other mammals, dogs and rats for example, are highly active sniffers of their environments. To some extent, our ignorance of the importance of human sniffing is changing in the face of such studies as the relatively recent first demonstration that humans, like dogs, can track a scent in an open-air environment [1]. Other research from this same group (see: http://www.weizmann.ac.il/neurobiology/worg/) has also shed light on the importance of sniffing overall to our sense of smell.

The act of odour localization and identification by sniffing is far more complex than simply pulling air up the nostrils and hoping for the best. The so-called olfactory-motor system functions in many ways just like the visual system. In both cases, the brain uses, and compensates for, head movements in locating an object. There is perceptual constancy too. A moving object does not appear to be bigger as it approaches even though the image size increases on the retina. Similarly, a bigger sniff sends more odour molecules to the receptors, but the odour does not smell stronger [2] as the brain takes the increased sniffing effort into account. There’s even a rough analog of binocular vision in the fact the airflow between the nostrils is always unequal, with each nostril being slightly more receptive to the absorption of certain odourants rather than others. Thus, the nostril each send a qualitatively different picture to the brain [3]. All of this is accompanied by typical search behaviours, in which humans dip in an out of odour “plumes” that travel through the air that we breathe.

Understanding these processes is important because active sniffing is an integral part of maintaining health, even if this is sometimes unacknowledged. Merely letting aromas waft into our noses unaided is – for most of us – not particularly useful in finding out if the milk is off, the gas is on, a fire is burning, or if that odd scent in the air is actually a dead rat under the kitchen floorboards. Active sniffing guides us towards things that require action and away from things best not ingested. Sniffing also ensures that we choose the right product across a vast range of consumer goods, from fruit to shampoo to cleaning products. And when we sniff the wine that a waiter has just poured, we are trying to locate parts per trillion levels of the cork taint, trichloranisole.

Sniffing can also reflect preferences. A recent paper from Shiori Nakano & Saho Ayabe-Kanamura of the University of Tsukuba in Japan showed that, at least when samples were from different categories, the most preferred odour of a group of consumer products (including foods, essential oils, soaps) elicited the longest initial sniff [4]. Interestingly, total sniffing time (participants could spend as much time as they liked sniffing each odour) did not differ between these odours. This combination of findings recalls a much earlier study showing that the first sniff is the deepest when trying to identify odours – all the remaining sniffing is about confirmation of the initial identification [5]. This finding is also consistent with an earlier study from my lab showing that pleasant odours of any type (food/non-food) elicited longer sniffing [6]. Indeed, even when we are simply imagining odour, we tend to sniff, and even then pleasant odors produce higher sniff volumes [7]. 

At first glance, such results look like they have potential for developing into a new behavioural – that is, non-verbal – measure of preferences. This is particularly so since sniffs can also reflect dislikes in that we inhibit sniffing if an odour is unpleasant. How unpleasant an odour has to be before sniff inhibition occurs, or the extent to which the speed of inhibition might reflect degree of unpleasantness is currently uncertain. Even so, could we measure liking in terms of sniff duration – product A gets 0.9 sec, but product B only scores a 0.65? The answer appears to be: not yet. The Japanese group’s paper reported that the discrimination between products due to sniffing was not evident when the products were from the same category (teas) and hence perceptually much more similar. The development of a more fine-grained sniffing measure might address some of the obstacles to a practical measure.

There are other important indices apart from preferences that might influence sniffing. In our study, sniffs were significantly longer and stronger when our participants were hungry, and this was true even when the odours were not related to foods or when there was no odour (a water blank). Although this is counter-intuitive, it makes sense if we think of sniffing as part of an exploration or search for food, which finishes once we have identified things to eat. Sniffing in this context reflects a motivation to search for, and identify, something good to eat. 

The motivational aspect of eating – commonly referred to as wanting, in contrast to liking (see October, 2012: http://prescotttastematters.blogspot.nl/2012/10/learning-to-want.html) – may in fact be a better predictor of what will be consumed when faced with different food types. Hence, an automatic (these responses occur in fractions of a second) and “objective” measure of wanting may develop into an excellent predictor of food choices. As with sniffing as a reflection of liking, the measure requires further development before similar products could be successfully compared. One thing such a measure would not do is eliminate the individual variability associated with other measures or liking or wanting such as rating scales. Adults with high levels of neophobia, for example, show much less vigorous sniffing to food odours, reflecting the wariness with which they approach foods generally [8].


1. Porter, J., et al., Mechanisms of scent-tracking in humans. Nature Neurosci., 2007. 10(1): p. 27-29.
2. Teghtsoonian, R., et al., Invariance of odor strength and sniff vigor: An olfactory analogue to size constancy. J. Exp. Psychol.: Hum. Percept. Perform., 1978. 4(1): p. 144-152.
3. Sobel, N., et al., The world smells different to each nostril. Nature, 1999. 402(35).
4. Nakano, S. and S. Ayabe-Kanamura, Smell Behavior During Odor Preference Decision. Chem. Percept., 2013. 6: p. 140-147.
5. Laing, D.G., Identification of single dissimilar odors is achived by humans with a single sniff. Physiol. Behav., 1986. 37: p. 163-170.
6. Prescott, J., J. Burns, and R.A. Frank, Influence of odor hedonics, food-relatedness and motivational state on human sniffing. Chemosens. Percept., 2010. 3(2): p. 85-90.
7. Bensafi, M., et al., Olfactormotor activity during imagery mimics that during perception. Nature Neurosci., 2003. 6: p. 1142-1144.
8.     Raudenbush, B., et al., Food neophobia, odor evaluation and exploratory sniffing behavior. Appetite, 1998. 31: p. 171-183.

No comments:

Post a Comment