Researchers at Brown University in the US found that responses to new odours depended on emotions experienced while the new odour was present.
"As humans, we're not immediately predisposed to respond to a scent and believe that it is good or bad," said Rachel Herz, a visiting assistant professor of psychology at Brown and the lead scientist of the study. "When we like or don't like a smell, that is learned."
Smell is intimately related to how human beings taste food but has long remained the most enigmatic of our senses. The average human nose can detect nearly 10,000 distinct scents, a feat that requires about 1,000 olfactory genes, or roughly 3 per cent of the human genome.
This latest research joins a growing body of fundamental studies seeking to unveil the mystery behind smell.
In October 2004 researchers were rewarded for the studies into the complex olfactory system with pioneering scientists Richard Axel and Linda Buck reaping a joint Nobel Prize in Physiology or Medicine for 2004 from the Swedish jury.
In 1991 the two scientists jointly published a fundamental paper in which they described the large family of 1,000 olfactory genes.
"The basic principles for recognising and remembering about 10,000 different odours were not understood. This year's Nobel Laureates in Physiology or Medicine have solved this problem and in a series of pioneering studies clarified how our olfactory system works," said the Nobel Assembly at Karolinska Institutet.
While electronic nose (e.nose) systems have been available for several years, the food industry has been slow to exploit the technology as part of real time quality assurance systems.
An EU-funded project - e-nose- recently revealed that the new technology can be used to quickly detect bacteria, yeasts, filamentous fungi and off-odours.
Researchers, led by Professor Naresh Magan at Cranfield University in the UK, used different types of e.nose systems - based on conducting polymer sensor arrays or metal oxide sensor arrays - for the 'rapid and early cost effective' detection of undesirable, harmful contaminants, toxins and taints in the dairy and bakery product industries.
Magan and his team are now in the final stages of completing e.nose trials in the food industry together with a cost-benefit analysis in areas of the dairy and bakery industries.
Electronic noses, launched commercially in 1995, are computerised tabletop units with sensors that detect odour molecules. In recent years researchers on both sides of the Atlantic have shown the noses can be useful tools to save costs through their ability to detect the quality of products.
Areas to date touched on by the electronic noses include detecting the flavours of different kinds of cheese, sniffing the quality of wine and coffee, and detecting fishy seafood before it gets to the consumer.
Herz conducted two experiments to test her theory of olfaction. The first included 30 female participants. All were asked to smell five scents and rate them for pleasantness, familiarity and intensity. Most odours were familiar and pleasant - rose, vanilla, lemon and peppermint. But one was new: a mix of odours that included dirt, rain and hot buttered popcorn. The result was a sweet, dank, slightly unpleasant scent.
Participants were randomly assigned into groups. The experimental group entered a room where the new scent was dispersed in the air by a hidden machine. Then they played a humourous card game on a computer and watch a comedy film. Again, the strange scent was gently piped into the room.
The second experiment was similar, but it included 36 participants, both men and women, and tested two new scents. One was slightly floral. The other had a clean, watery smell. All participants rated these scents - as well as rose, vanilla, lemon and peppermint - in a pre-test. Participants rated the new scents as both unfamiliar and pleasant.
Then one group entered a scented room and played a computer card game designed to be frustrating. The second group sat in a scented room and read magazines. In both groups, half were exposed to the floral scent, the other half to the watery one. The third group played the computer game in an odourless room.
In the first experiment, Herz found participants who played the game and watched the film clips were more likely to rate the new odour - even though it was slightly unpleasant - as enjoyable and familiar compared with control groups. Results were similar in the second experiment. After playing the frustrating game, participants were more likely than control groups to score the pleasing new odours as distasteful - a negative association that grew over time.
"When an odour is paired with an emotional event, perception of that odour was altered to fit that association," conclude the researchers.
Full findings of the Herz study are published in the current issue of the International Journal of Comparative Psychology.