Cognitive Daily

ResearchBlogging.orgI had a newspaper route up until I was in the ninth grade, and what I dreaded about the job was going door-to-door collecting subscription fees. The worst part was probably the odors in some of the houses. One house emanated a toxic mixture of Lysol, alcohol, pet dander, and cigarette smoke. These people inevitably were out of cash, so I had to return again and again until I finally was able to negotiate payment — sometimes months overdue.

But maybe the smell was prejudicing my judgment. Lots of people couldn’t pay me right away. Why should I only hate the ones with drinking/pet/smoking/air freshener problems? Other than the fact that they had all those problems, they weren’t any better or worse than anyone else (aside from the nice old ladies who baked me cookies I could smell a half-block away).

Hmm… come to think of it, there’s a lot a smell can tell you about a person. Are they overperfumed, undermouthwashed, sweaty, smoky, or infused with motor oil? Different scents clearly have different meanings.

But some smells are too subtle to be detected. You might not be able to discern your wife’s perfume by the end of the day, although traces are still present. What if someone else was wearing her perfume, still at levels you don’t consciously notice? Would that affect your impression of them?

We know from studies on subliminal images and sounds that even when we’re not conscious of these things, they can affect our judgments and actions. But researchers have had difficulty finding any effect of odors that we can’t consciously identify. A team led by Wen Li saw procedural problems in those early studies: An odor that one person can’t detect might still be obvious to someone else. Even the same individual might perceive an odor sometimes and not others (“I thought I smelled smoke, officer!”).

Li’s team tried to correct those problems by giving 31 student volunteers individualized tests of their thresholds for detecting odors. They sniffed bottles containing progressively weaker solutions of three different odor-producing substances: Citral (lemon), Anisole (“ethereal” – a neutral scent), and valeric acid (sweat). For each substance, when each student could no longer detect the odor, a solution 56 percent weaker was chosen for use in the experiment.

Next, these same volunteers were asked to rate photos of faces for likability. Before rating a face, they took a sniff from a random unlabeled bottle (from one of the four weak solutions), and indicated whether they detected an odor. Then they saw the face and rated how likable they thought the person depicted was. So even though the odor was undetectable, did it have an affect on likability? Turns out, it depends on the individual doing the rating. Despite being unable to detect a stronger odor in the pre-test, the responses of 15 of the students indicated (via a d’ measure) that they did notice the smells while rating faces — although they were still extremely doubtful of their abilities, and no one could accurately determine which smell was which. Here are the results:

i-ef472bf5cc40bf320c1f6f9710597d0a-li2-1.png

Those who weren’t conscious of the smells rated faces significantly more likable after smelling the lemon scent than they did after smelling sweat. There was no significant difference for any of the smells for those who were conscious of the odor. Some of the students wore heart rate monitors during the test, and heart rate went up following the sniffs when they were conscious of the odor. But if a student was unconscious of the odor, heart rate was likely to go down after a pleasant or neutral odor, and up after an unpleasant odor.

So it appears that even odors we can’t detect have an effect on our impressions of faces — even when it’s quite clear that researchers are studying the relationship between odors and faces. In fact, Li’s team thinks it might be possible that this knowledge might have caused the reaction by the conscious group: they may have been trying to counter the negative impression of a bad smell with a better rating. It’s only when you’re truly unconscious of what you’re smelling that you can’t consciously manipulate your preferences.

This makes some sense–in the real world, if you meet someone on a smoggy day, you may try to take that into account when judging their personality. But if there’s no other explanation for a foul smell when you meet a new person (or if you don’t consciously notice the smell), then you’re more likely to form a negative assessment of them. Especially if they’re late paying you for their newspaper subscription!

Update: I have a confession to make. Over a year ago, I reported on this exact study. Somehow it slipped my mind as I read the article again a few days ago, and I wrote an entirely new report on the same study. Only when I was Googling this morning for the article’s DOI did I come up with my original report on this study! Someone on Twitter challenged me to post it and see if anyone noticed, so I did. No one noticed after over three hours and hundreds of page views. I guess it’s a matter of collective amnesia — and it suggests that we could all benefit from a reminder about great research every now and then! Here’s the link to the original report. Sorry about that!

Wen Li, Isabel Moallem, Ken A. Paller, Jay A. Gottfried (2007). Subliminal Smells Can Guide Social Preferences Psychological Science, 18 (12), 1044-1049 DOI: 10.1111/j.1467-9280.2007.02023.x

Comments

  1. #1 jay
    February 17, 2009

    This should not be surprising. It is probably the oldest and deepest of our senses (even bacteria have chemical detection). a foul (dangerous) smell can induce nausea and revulsion more than any other sense… it is most deeply wired to our reptillian brain.

    A funny story about this. Years ago my wife worked in animal control and one day had a minor run in with a skunk. There was still a light bit of it on her uniform (not enough to be bothersome) when we stopped at Home Depot after work. I overheard someone discuss how bad a man in the next aisle smelle (he had a somewhat dishevelled look), because they automatically assumed it was him. He ‘looked’ like he should smell bad.

  2. #2 Matthew
    February 17, 2009

    I’m currently dealing with a different form of this – I’ve been dating someone recently who is hypersensitive to smell, often noting when our meal is on its way at a restaurant or a foul street odor well before I detect any sign. I’ve got a white collar research job, so never thought of myself as having much of any particular odor, but find myself being nudged to the shower more and more! Ack

  3. #3 Lilian Nattel
    February 17, 2009

    So the moral of the story is when you go for a job interview, wear dilute lemon juice on wrists.

  4. #4 Ayshela
    February 17, 2009

    given the results of the temperature/favourable response study, I expected this result for the unconscious responses. The conscious responses – that’s the kind of thing I’d hoped for but not really expected to see. ;)

    as for the update – it’s entirely possible that the largest number of people viewing the page haven’t been following you for a year, so wouldn’t have seen the previous posting.

  5. #5 Sagehorse
    February 17, 2009

    I completely disagree. I think the obsession with smells is part of the capitalist plot to overconsume our lives, put us in bankruptcy and give us cancers.

    Get over yourselves.

  6. #6 Nuwan
    February 17, 2009

    Dave, for your older article (in 2007) it stated that “Then, for the actual experiment, bottles that were about 100 times more dilute were used.”

    while for this current article, you state that “For each substance, when each student could no longer detect the odor, a solution 56 percent weaker was chosen for use in the experiment.”

    Does 56 percent weaker mean 100 times more diluted? Further, how did they come up with this 56 percent more dilution as a threshold of undetectable? How are we certain that the brain could even detect (subliminally or not) at 56%?

  7. #7 Donna B.
    February 18, 2009

    One of the most comforting smells for me is a well-mixed combo of grease, dirt, sawdust, and diesel, because this is what my Dad smelled like every evening when he came home and hugged me when I a kid.

  8. #8 NIk M
    February 18, 2009

    I got a social facilitation effect at university using a task completed in isolation but written on paper impregnated with androstenone, so this doesn’t surprise me at all.

  9. #9 Susan Weinschenk
    February 18, 2009

    Proving once again that much (I would claim most) of our decision-making is made “unconsciously”. I am most interested in unconscious decision-making applied to decisions at web sites, but until we get the smell interfaces in common use, I don’t think I’ll be able to apply this to my work or include a chapter on it in my next book! I should probably be glad that so far we don’t have little smellpads hooked up to our laptops that are stimulated via HTML code, right?

  10. #10 Tinkering Theorist
    February 18, 2009

    How did they decide that lemon was positive and anisole was neutral? I think anisole is much more positive of a smell than lemon. Of course, anisole reminds me of my grandmother’s pizelles, but I think it’s more than that. Lemon seems a clean and happy, but desserts are better than cleanliness . . . and surely many people like liquorice or anise cookies.

  11. #11 Alvin
    February 19, 2009

    I have a confession to make. Over a year ago, I reported on this exact study.

    I knew I’d read about this somewhere before! I assumed it was on another blog, or maybe in the Science section of the NYTimes. Funny to see it was right here all along!

  12. #12 kevin
    February 19, 2009

    So, would it be fair to conclude that this would contribute to an overall perception of other people? That you would think people who subliminally smelled like lemons are more agreeable than those who smell like sweat?

    Also, I thought the smell of sweat could be attractive. Is it different between being attractive and causing a positive regard for the sweaty person?

  13. #13 AnnR
    February 20, 2009

    I had a break-in some years ago. I was in the basement and heard noise. When it didn’t stop I carefully went upstairs.
    The first thing I noticed as I climbed the stairs was a smell. It was not identifiable, although it might have had a fried-food tinge. Perhaps the smell of a burglar who was a bit sweaty? It’s not like I kept a perfectly clean house, but I knew this smell was not normal.

    That was enough to cause me to yell and then the kitchen door slammed as the intruder exited.

    So I know that smells register. I was on edge because I was alone and heard a noise. My sniffer went into overdrive.

    It took me years to be comfortable in my house alone after that, but one thing I know is that you should trust your senses because they are working even when you didn’t tell them to!

  14. #14 StellaStinky
    February 22, 2009

    i don’t bathe for 10 to 15 days. i love the way i smell by then. what’s wrong with being ‘au naturel?’

  15. #15 aliana
    May 4, 2009

    I kind of like stagehorses response, but can’t agree entirely. There’s a huge difference between naturally generated plant scents (essential oils and resins) and scents made in a lab (which is nearly every smell you’ll encounter in regular society). I’ve long held that synthetic scents actually damage our limbic region, subsequently hampering our intuition and other more nebulous faculties.

    Love your blog though. Very intresting.

  16. #16 Sonal
    July 27, 2009

    I am not surprised by the finding. Its a known fact that baby animals recognize their biological mother by her smell, though it might not be disctict from the herd otherwise.

  17. #17 ROSE
    January 16, 2010

    After reading this article,am requesting you to mention the resaerch methods used. Thanks for your time.