1 June 2019 | New Scientist | 43I
QUITOS is a hard place to get to. Nestled
deep in the Peruvian Amazon, the city
can only be reached by air or water – no
roads connect it to the rest of the world. But
for Stéphane Piquart, that remoteness is part
of the appeal. An odour prospector for the
French fragrance company Behave, Piquart
went to Iquitos to search for new aromas. In
particular, he was looking for a fragrant plant
root that the local Shipibo people use in a
love and friendship potion. The root, which
the Shipibo call piri-piri, has a remarkable
fruity-leathery scent that Piquart has
now brought to perfumery.
Visiting isolated places in search of new
smells shows just how keen perfumers are
on striking olfactory gold. That is driven by the
sheer might of the fragrance industry. Always
big business, it is now huge, with annual sales
of $70 billion worldwide for not just perfumes
but everything from soap and shampoo to
candles and air fresheners. That is a lot of
money to spend on nice smells. But what
makes us like the ones we do?
Smell is the least understood of our major
senses, making this surprisingly tricky to
answer. Neuroscientists, psychologists and
even AI researchers are beginning to unpick
the mysteries of how we perceive scent,
while at the same time, fragrance
researchers are devising new ways
to tickle our olfactory neurons.
The first clear lesson in what makes a good
scent is that different cultures have different
ideas about which smells are pleasant. “In
Europe, people love anise,” says Christophe
Laudamiel, a master perfumer based in New
York and Berlin. “In America, I wouldn’t thinkof putting it in a fragrance.” Similarly, North
Americans find the smell of the wintergreen
plant highly pleasant, at least partly because
it is a prominent ingredient in root beer.
Europeans, in contrast, find it unpleasantly
reminiscent of liniment, the stuff you
might rub on aching limbs.
Personal experience plays a key role in
judgements of odour pleasantness. Unlike
the other major senses, the nerve impulses
involved in smell connect directly to the
limbic system, the primitive parts of the brain
responsible for memory and emotion. As a
result, we judge the pleasantness of odours by
the emotional memories they evoke. “A simple
example is the scent of clove,” says Charles
Sell, a chemist now retired from Givaudan, the
world’s largest fragrance company. “I love it,
because when I was a child, my mother used to
make apple pie and put cloves in. But clove oil
is also used in dentistry, so someone who has
had a traumatic dental experience when there
was clove oil will not like clove.”Learning disgust
In fact, Sell says categorically that all odour
preferences – from our fondness for vanilla to
our dislike of decay – are learned. As evidence,
he points to the willingness of young children
to investigate anything. “They will do things
that would disgust an adult, like attempting to
eat the contents of their nappies,” he says. “It’s
the immediate aversive reaction from their
parents that teaches them this is a bad smell.”
It is a view shared by many experts, though
there is surprisingly little research to back it up.
Some studies, however, show that the veryOdour prospectors travelling the world are
working with chemists to answer a deceptively
simple question: what makes a good smell?
Bob Holmes takes a deep breath
same molecule can smell pleasant or
unpleasant, depending on which associations
are triggered. Many cheeses, for example,
contain the same odorous molecule –
isovaleric acid – as sweaty socks, says
Andreas Keller, an olfactory researcher and
independent consultant based in New York.
“Everybody likes the smell when they’re told
it’s cheese they’re smelling. Nobody likes it
when they’re told it’s dirty socks,” he says. “The
smell didn’t change, but the judgement did.
Those judgements, obviously, are all learned.”
Case closed? Not exactly, because there
is another layer to this argument – one that
requires a deeper explanation of how the
sense of smell works. Unlike vision or hearing,
which perceive relatively well-defined aspects
of the world – a narrow band of wavelengths
of light and an even narrower band for sound
waves, respectively – our sense of smell
is tasked with recognising an essentially
infinite number of different molecules
To do that, our noses contain roughly 400
different odour receptors. Scientists still don’t
fully agree on how these receptors recognise
odorant molecules: most think each molecule’s
shape matches a complementary one on the
surface of the receptor, but a few believe at least
some receptors work by sensing the frequency
of vibration in the molecule’s bonds. However
it happens, though, each receptor recognises
a different, often overlapping set of odorant
molecules, and each odorant can trigger just
one receptor or many different ones. What
we recognise as a smell, then, is a particular
pattern of odour receptor activation, like
a chord played on a 400-keyed piano.
The picture gets even more complex >