In the first volume of his autobiographical novel "In Search of Lost Time" (or, if you prefer, "Remembrances of Things Past"), Marcel Proust famously wrote about how the odor (and taste) of a small French cake (petite madeleine) dipped in tea evoked highly emotional memories of his childhood.

In recent years, scientific studies have confirmed a strong link between odor and memory. Last week, a team of Israeli researchers reported in the journal Current Biology that the reason we associate particular odors with childhood may be because something unique goes on in our brain during our first, but not our subsequent, encounters with a smell — something that may leave a lasting impression, for better or for worse.

Researchers at the Weizmann Institute of Science in Israel presented adult volunteers with images of 60 objects. During each presentation, a machine called an olfactometer emitted either a pleasant or an unpleasant odor. The volunteers were then asked to look at the images a second time and to try to recall which odor was associated with which image. During this part of the experiment, the volunteers underwent fMRI imaging, which measures neural activity in various areas of the brain. Finally, the entire process was repeated — with the same images, but with different odors for each.

A week later, the volunteers were brought back to the lab to have their brains scanned again while reviewing the images and trying to recall the odors that had been emitted when they saw them before.

Unique brain-activity pattern created
The researchers found that people remembered the earlier of the two image-odor associations best if it had been unpleasant. But — and this was the really interesting part — they also found that the first image-odor association created a unique pattern of brain activity, even if both odors were equally remembered. This "signature" activity appeared in two brain structures: the hippocampus (key to memory formation) and the amygdala (crucial to the processing of emotions).

So strong was the brain activity pattern that the researchers could accurately predict from the fMRI readings taken on the first day of the experiment which associations a person would remember the following week.

The researchers repeated the study using sounds, but found that auditory experiences had no similar imprint on the brain.

Proust was on to something. Our first associations with an odor may be the ones we carry with us all our lives.

(On a completely non-health-related — but definitely Proustian-related — topic, I recommend Germaine Greer’s curmudgeonly article, “Why Do People Gush over Proust? I’d Rather Visit a Demented Relative,” in last weekend's Guardian newspaper.)