Are men more attracted to white smiles?

A press release recently went around saying that toothpaste manufacturer Colgate had commissioned a study looking at how people react to white smiles. This study involved neuroscience - BRAINS WERE MEASURED - and found that "in a social context men are 48% more engaged by women with a whitened smile", that "a woman is 24% more likely to be committed to memory in a job interview if she has a whitened smile", and that "Aussies 30 years and younger are 34% more likely to remember someone with a whitened smile versus an unwhitened smile."

 This press release seemingly hasn't gotten much media play; the only thing I could find was Alison Larsen at Popsugar, saying that "in a social setting, 48 percent of men were more engaged by women with a white smile." Larsen seems to have misread the press release; it claimed that men are on average 48% more engaged by smiles, rather than that 48% of men caring about white smiles (and presumably the other 52% not giving a toss). One suspects, anyway, that these statistics will probably end up being said by a woman in a reassuring sounding voice on a Colgate ad. So - does it stack up? Should you be whitening your teeth if you care about what others think of your appearance?

Neuroscience is golden, right now.  Barack Obama recently announced a $100 million brain research initiative, presumably because he needs the good PR right now (hi there, NSA bots!); it's only good PR because people will read it and go, "yeah, that seems like decent money for interesting science." People are fascinated by brains, fascinated by fMRI studies and MRI scans. Newspapers and science magazine regularly feature studies saying that this or that is caused by the posterior frontal lobe or the ventral anterior temporal cerebellum. Most of the people who read the newspapers have no idea what these words mean, but they sound good, don't they? (I did make up the 'ventral anterior temporal cerebellum', by the way). In fact, a 2008 study by Donna Weisberg suggested that when people see the words "brain scans indicate" they're more likely to rate another otherwise quite poor explanation as a good one. I mean, anybody using phrases like 'ventral prefrontal temporal cerebellum' must know what they're talking about!

So, seeing how impressed we are with neuroscience, it's not surprising that Colgate want in on the action. "Neuroscience says that" in a press release definitely sounds better than "surveys say that". So Colgate engaged a Melbourne-based neuromarketing firm, Neuro-Insight, to give them some cold hard neuroscience that they could put in a press release, and then use in ads. Hence the statistics cited earlier. I was curious, though, about where such numbers actually come from, so I've been emailing with the very helpful Neuro-Insight people trying to get details of the study. 

So, the details: the Neuro-Insight people had 100 participants from the Melbourne area, and got them to wear a headset while they looked at pictures of women on a screen. The headset they were wearing was a sort of electroencephalograph (EEG); this is a relatively old technology, where electrodes glued to the person's scalp can pick up the electrical impulses of the brain. In contrast, much of the brain research you've heard about recently uses a newer technique, fMRI, which picks up the changes in chemistry that occur when the brain uses energy. fMRI has pros and cons compared to EEG, of course. EEG has much higher temporal resolution (you can figure out when things are happening down to the millisecond), whereas fMRI has much higher spatial resolution (you can generally pinpoint which part of the brain does what with higher accuracy). In particular, the headsets the participants wore in this study were set up to use an EEG technique called steady-state topography; basically, as people watched things on a screen, the screen flickered in such a way that meant that the researchers could fairly accurately calibrate what's on screen to the electrical impulses in the brain.

In the study, adults (18-54 years old) sat in a group looking at pictures on a screen. The pictures presented women who were either natural-looking or wearing more obvious makeup, in three separate contexts (indicated by the backgrounds of the pictures - 'dating', 'social' or 'professional'). Teeth were whitened by 'a few shades' in some versions of the photos and not in others. The participants weren't told anything about what the researchers were looking for, and they didn't have to do anything other than watch the screen. Different participants got the pictures in different orders, so that none of the pictures were seen more recently than any other (order effects can often affect an experiment's result).

So what does this all mean? Basically this: when the press release says (for example): "in a social context men are 48% more engaged by women with a whitened smile", what it means is that, when men in the study saw a smiling women with a background context that was 'social', there was 48% more electrical activity being captured in the electrodes nearer to the parts of the brain that the researchers associate with 'engagement' (the left and right prefrontal and frontal lobes, in particular). However, it's hard to know what this 48% means, exactly. Firstly, I'm not certain that this 48% is the kind of real, statistically significant difference that scientific journals expect to see when they're trying to judge the paper as being worth publishing - Richard Silberstein, one of the researchers, implied to me in an email that this 48% was actually not a statistically significant difference once they had adjusted their statistics in accordance with best practice (though other findings in the study were statistically significant - he pointed at how whitened teeth seemed to significantly affect the brain's electrical signals in people under the age of 30).

But more importantly, even if this is a real effect, 48% more brain activity across a few electrodes, on average, doesn't necessarily translate to a guy being 48% more interested in a woman when she has bright white teeth. 48% more brain activity doesn't necessarily mean 48% more interest, for starters; there's not going to be a one-to-one ratio between increased brain activity and increased subjective feelings of engagement. Remember that EEG techniques like steady-state topography have relatively poor spatial resolution. If there is more brain activity from around the area of the prefrontal cortex, it could mean that the men in the study are simply paying more attention when there's a flashy white smile. However, it could also mean a bunch of other stuff. That area of the brain plays a role in lots of stuff! It plays a role in deciding what to do, in learning and categorising, and in short-term memory (trying to remember things for the very near future - think trying to remember a phone number while you find paper to write it down on). So, there might be increased brain activity here because the person really is more engaged, really is paying more attention when the woman in the picture has whitened teeth. Either that, or they're making a mental note that they might have to remember something about the picture later, or they're consistently categorising that photo as "teeth whitened" (presumably they see each photo a few times, and so the difference in teeth shade stands out). Or maybe seeing the white teeth make them think to themselves about how they should brush their teeth when they get home, because their teeth aren't that bright and it's obviously something that matters (the prefrontal cortex also has something to do with such internal monologues).

It's hard to tell if any of these possibilities were more likely than others, anyway, because there wasn't a 'behavioural' aspect to this study. They didn't ask the participants to fill out a survey about how much they say they love white teeth (or not), and didn't try and figure out if those white pictures changed people's behaviour in any way (e.g., by getting people to rate pictures for attractiveness, and only showing some people the ones with whitened teeth, to see if, overall, they rate them differently). I suspect the researchers would justify this lack of behavioral tasks by pointing at how other research has previously shown that brain activity in their 'engagement' area (the prefrontal/frontal cortex) has been correlated with such tasks previously. I'd argue, however, that brain activity in their 'engagement' area probably correlates with quite a few different behavioural tasks, and that they need to show more conclusively that it's the 'paying more attention to whitened teeth' and not some other thing that's causing the brain activity for me to be convinced.

Of course, other researchers have done such behavioral research in the scientific literature; they've had versions of faces with whitened teeth or without, and asked people to rate them for attractiveness. Interestingly, a German study and an American study both found that whitened teeth had no effect on people's ratings of facial attractiveness. In contrast, a Finnish study found that those with little education were very slightly more likely to value whitened teeth than those with more education, and that young people very slightly cared more about it than old people (these correlations were about r = 0.1, which is small but apparently significant). It is possible that participants in these studies knew too much about what was going on, and didn't want to admit that they cared about teeth. It's also possible that, while they may not find people with whiter teeth more attractive, they may pay more attention to them nonetheless. It also seems equally possible to me that people just don't care that much about teeth being a few shades whiter. More research probably needs to be done!

In the end, the Neuro-Insight study isn't a bad study from the scientific perspective. It is interesting that there is more brain activity when people look at whitened teeth compared to unwhitened teeth, especially in young people. I don't think they've conclusively shown what exactly that brain activity is correlated with, but hey, human behavior (and the brain stuff that's behind it) is complicated, and there's lots of things this brain activity could be correlated with. There's lots of things that could affect results, and much of science is about systematically trying different things to see what happens. I'm happy to see this study as part of the of advance of science, as part of scientists slowly coming to figure more about stuff through systematic experimentation (assuming that the Neuro-Insight eventually publish the research in a scientific journal). But, I'm not quite sure it shows what Colgate want it to show; in the end, I haven't seen much evidence here that you should worry too much about the whiteness of your teeth. Of course, maybe I'm just saying that because my teeth are a little yellower than I'd ideally like them to be.


Tim Byron

profile of TimByron