Credit: iStockphoto
MELBOURNE: In a step toward the digital communication of smell, scientists trained a machine to predict the pleasantness of an odour and showed that we all perceive smells in the same way.
Certain smells can either repulse us or appease us, but how a collection of molecules manages to evoke such emotional responses has been under scientific debate until now.
"Odours are not only in the eye of the beholder, we think there is a common basis to perception of smell to all humans - this is why all people do not like the smell of decay but love the smell of lilies - this common thing is the molecular structure," said neurobiologist Rafi Haddad from the Wiezmann Institute, the lead author of the study published in PLoS ONE.
A rose is a rose … no matter the nose
"This finding of hard-wired odorant pleasantness is in contrast to the popular notion that odorant pleasantness is both subjective and learned," the researchers wrote in their paper.
Researchers from the Weizmann Institute and Edith Wolfson Medical Centre in Israel trained an electronic nose to predict the pleasantness of an odour it encounters for the first time.
The machine could predict whether a smell would be pleasant or not, even if it had not been exposed to that odour before. Therefore something common in the odours structure that makes it pleasant or unpleasant, according to the researchers.
Electronic noses
Artifical noses, or eNoses, have been developed and used over the last decade to detect and identify a particular odour that they had previously 'learned'. As an odour passes through an eNose, the molecular structure of the odour stimulates certain chemical sensors within the eNose, producing a unique 'odour fingerprint'.
eNoses can be used to identify, as a whole, the mixture of components that together form an odor. This is different to detectors such as gas chromatographs, which identify components of an odour.
eNoses have proved useful in monitoring environmental pollution and even in diagnostics - mainly presence/absence of a particular odour. Until now, they have never been used for reporting odour quality.
Haddad and his team trained an eNose, using 76 odours, to rank an odour it had never been exposed to before along an axis of pleasantness - meaning they trained an eNose to predict how pleasant a new smell is.
