Posted: April 30, 2010

Artificial nose gives as good as it gets

(Nanowerk News) Is smell a personal sense, one that is specific to culture? No, says an Israeli team of scientists who have successfully 'trained' an electronic system capable of predicting the pleasantness of new odours. Their study, presented in the journal Public Library of Science (PLoS) Computational Biology ("Predicting Odor Pleasantness with an Electronic Nose"), was supported by the EU through a Seventh Framework Programme (FP7) grant from the European Research Council (ERC). The results could potentially lead to new methods for odour screening and environmental monitoring.
The last 10 years have seen rapid developments in the technology of artificial noses, or 'eNoses' as they are also known. These electronic devices can detect and recognise odours thanks to chemical sensors that are 'housed' inside noses. One of the objectives of artificial nose technology is to report perceptual qualities of new odours.
In this latest study, the scientists from the Weizmann Institute of Science and the Edith Wolfson Medical Center, both based in Israel, contend that the perception of an odour's pleasantness is naturally hard-wired to its molecular structure. Personal or cultural differences are evident only within specific contexts, they said.
For the purposes of their research, the scientists tuned an eNose to human odour pleasantness estimates, and then used the device to predict the pleasantness of new odours. The team asked native Israelis to rate a selection of odours on a scale from 1 to 30, ranging from 'very pleasant' to 'very unpleasant'. The team used the results of these tests to develop an 'odour pleasantness' algorithm that they programmed into the eNose.
The team then used the newly-programmed eNose to rate a number of unfamiliar odours. To find out whether the eNose's understanding of what smells nice chimes with the human nose's opinion, the team got a second group of Israelis (who had not been involved in the first part of the experiment) to rate these new smells.
The scientists found that their device's ratings of odour pleasantness were over 80% similar to the humans' ratings. The eNose's ratings were also more than 90% accurate at discriminating between categorically pleasant or unpleasant odours.
But does our culture influence our opinions of what smells good? To find out, the team tested eNose predictions against a group of Ethiopians who had recently emigrated to Israel. The scientists found that the eNose was able to predict Ethiopians' opinions of a new smell, even though it was 'tuned' to the pleasantness of odours as perceived by native Israelis. According to the team, this suggests a cross-cultural similarity in odorant pleasantness.
'Culture influences the perception of olfactory pleasantness mostly in particular contexts,' explained Professor Noam Sobel from the Department of Neurobiology at the Weizmann Institute of Science. 'To stress this point, many may wonder how the French can like the smell of their cheese, when most find the smell quite repulsive. We believe that it is not that the French think the smell is pleasant per se, they merely think it is a sign of good cheese. However, if the smell was presented out of context in a jar, then the French would probably rate the odour just as unpleasant as anyone else would; that is why the French don't make cheese-smelling perfume.'
The scientists said that the molecular features of an odour stimulate the eNose's chemical sensors, triggering a unique electrical pattern - what the team calls something like an 'odour fingerprint' - that characterises the specific smell.
The eNose must be 'trained' with odour samples in order to build a reference database, according to the researchers. Where human noses and the eNose differ is that the former can recognise and even classify a novel odour whose fingerprint is not available in the database.
The Israeli team's innovation was to train the eNose to predict whether an odour would be perceived as pleasant or unpleasant, or even something in between. 'The uniqueness of this approach was that rather than learning singular odorant objects such as 'rose' or 'skunk', their eNose learned an axis, and could then place novel objects anywhere along the axis it learned,' the researchers explained.
'These findings suggest that unlike in vision and audition, in olfaction there is a systematic predictable link between stimulus structure and stimulus pleasantness,' the authors of the study wrote. 'This goes in contrast to the popular notion that odorant pleasantness is completely subjective, and may provide a new method for odour screening and environmental monitoring, as well as a critical building block for digital transmission of smell.'
Source: Cordis