Dreaming of Smell

Do we dream in odor? Artists and poets think so—for example, the decadent 19th century writer J.-K. Huysmans described a vivid dream sequence involving scent and color—but the question has rarely been investigated scientifically. Back at the dawn of scientific psychology an instructor at Wellesley College named Mary Whiton Calkins analyzed dream diaries kept by two volunteers over a six to eight week period. Her results, published as “Statistics of Dreams” in the American Journal of Psychology in 1893, showed that smells appear in dreams only occasionally: visual and auditory perceptions are far more common. A follow-up study at Wellesley by Sarah Weed and Florence Hallam in 1896 estimated that odors show up in 15% or less of all dreams.

It was a hundred years before the question was tackled again. In 1998, researchers at McGill University in Montreal asked 164 people whether they had ever experienced sensations of smell or taste in their dreams; 41% of women said yes, as did 35% of men. The McGill team also gave the participants a bedside log in which to write their dreams each morning. They ended up collecting 3,372 dream reports. About half of these mentioned auditory impressions but only 1% mentioned smell. The diaries also revealed a huge sex difference. As in so many other measures of odor perception, women are more tuned in than men: odorous dreams were recorded in the diaries of 20.9% of women but only 2.0% of men. It appears that dream smelling is a widespread but low frequency phenomenon: in other words, it’s something many people have experienced but not all that often.

Can odors from the real world make their way into our dreams? Since the early sleep studies of the 1950s, we have known that external stimuli such as light and sound are occasionally incorporated into dream content. Play music next to someone in the REM stage of sleep, for example, and he may dream that he is conducting an orchestra. In 1988, sleep researchers at Cal State Sacramento exposed people to a smell for five minutes during a REM episode. They woke them a minute later and asked them what they were dreaming. An odor was mentioned about 19% of the time, a rate of dream incorporation similar to other types of external stimuli.

Here’s a typical result from the study:

The subject was presented with a freshly cut lemon while
in REM sleep. The resulting dream was: “I dreamed I was
in Golden Gate Park. I was walking by some gardenias.
They were just opening. All of a sudden, I could smell the
gardenias, but they smelled like lemons instead of gardenias.

The researchers used both pleasant smells (lemon, peanut butter, roses, etc.) and unpleasant smells (match smoke, dog feces, onion, etc.). The pleasant smells were more likely to show up in dreams (on 27% of attempts) than were unpleasant ones (11% of tries). Oddly, the pleasantness of the smell had little effect on the emotional tone of the dream itself: roughly a third of all odor-stimulated dreams were unpleasant in tone.

A similar experiment was reported a few weeks ago at the annual meeting of the American Academy of Otolaryngology in Chicago. A German team found that unpleasant smells produced dreams with predominantly negative emotional content, while nice smells made for pleasant dreams. This yet-to-be-published result is slightly at odds with the Sacramento State finding, but one thing seems clears: The boundary between dreamscape and smellscape is porous. Scents from the real world stimulate our sleeping nose and sneak into the nighttime unreality of our dreams.

The Australian psychologists Richard Stevenson and Trevor Case examined dream smells in a questionnaire study in 2004. Of the 284 participants, 26.4% had had olfactory dreams. When descriptions of these smell dreams were analyzed about half turned out to be food related—with bacon leading the way. Smoke or burning odors appeared in 21% and body or animal odors turned up in another 21%. As in the waking world, smell sensations in dreams were briefer and more emotional than the visual experiences.

Stevenson and Case also had their volunteers fill our questionnaires on visual and olfactory mental imagery ability. (Full disclosure: the olfactory questionnaire was one I co-wrote with Melissa Crouch and Sarah Kemp and published in the Journal of Mental Imagery.) They discovered multiple links between dream smelling and smell ability in the waking world. First, olfactory dreamers experience both visual and olfactory imagery more vividly than non-olfactory dreamers. Second, people with more vivid mental imagery for smells have more vivid smell dreams. A follow-up experiment found a third link: olfactory dreamers are better at identifying odors in a smell test.

All of this suggests to me that some people are simply more tuned into odors than others. Smell-oriented people—those who identify odors accurately and imagine them vividly—tend to dream in smell as well. Olfactory talent shows itself all around the clock.