Glenn Wilson falls off the wagon?
« previous post | next post »
According to Vaughn at Mind Hacks ("The demon drink", 5/29/2009):
Oh dear. It looks like psychologist Glenn Wilson has fallen off the wagon again. From the man who brought you the 'email hurts IQ more than cannabis' PR stunt before repenting, comes the 'the way you hold your drink reveals personality' PR stunt.
This time it's to promote a British pub chain and God bless those drink sodden journos who have gone and given it pride of place in the science section of today's papers.
In this case, "the way you hold your drink" is not referring to your reaction to various blood alcohol levels, but rather to the position of your fingers, hands, arms, and body, relative to your beverage container.
Here's the BBC's take on Wilson's categories:
There is some genuine science of body language out there. And as the automated analysis of digital video gets better, we may get to the point where there is actually some basis for inducing categories, as opposed to interpreting (reactions to) postures. But this particular incident seems to be another case where there isn't actually any research at all, just a publicity stunt, in this case dreamed up by the PR department of a bar chain.
As the Mind Hacks post observes:
I would point out that it's not published, or even sensical, but is there really any point when the whole premise is so ridiculous that you'd have to be virtually paralytic to take it seriously.
Wilson has actually done a great deal of serious research and is well known for his work on personality but occasionally seems to go on inexplicable media binges on the tab of corporate advertising.
I don't think I'll be so quick to apologize to Prof. Wilson as I was last time.
Here's the Telegraph's illustration for their article:
Which of Wilson's 8 categories do you think she belongs to?
Blake Stacey said,
May 30, 2009 @ 2:27 pm
That's definitely the "I'm being paid to appear in this photograph, and don't you want to buy the product featured in this advertisement?" look.
Tim Silverman said,
May 30, 2009 @ 2:45 pm
"This is washing-up liquid! WTF? I think I'll leave it for later, har, har! I wonder which joker thought this would be amusing. He's toast if I get my hands on him."
ArthurDent said,
May 30, 2009 @ 2:49 pm
Well, I don't know, but it may be worth researching.
John said,
May 30, 2009 @ 3:17 pm
But.. they… b… wh… they all have different drinks! They're in different *stages of drinking* those drinks! They're all sitting/standing in different positions! They're different sexes, in different groups, wearing different clothes and presumably in different establishments and different social situations! How can anyone… who co… why….
*brain melts*
John said,
May 30, 2009 @ 3:23 pm
On reflection: I forgot that putting something in the "science section" of a newspaper doesn't necessarily imply said paper believes the story has any scientific merit. This is of course just one of those "look at what those crazy so-called "scientists" are doing with their time now!" stories. It'd be fine if those were the exception, like sports sections having "The week's cricket as explained by the editor's 7 year old daughter", but they're more like "we can't afford a full-time sports reporter, so for the foreseeable future the editor's 7-year-old daughter will take the helm".
rootlesscosmo said,
May 30, 2009 @ 3:35 pm
The "Does this smell funny to you?" category.
Ray Girvan said,
May 30, 2009 @ 9:11 pm
"This is washing-up liquid!
Or maybe it's that veterinary drink?
Blue Cure-a-Sow.
Heather Rose Jones said,
May 30, 2009 @ 9:40 pm
"email hurts IQ more than cannabis"
Believe it or not, just two days ago I was in a "productivity seminar" at work where the instructor referenced this study. And I said to myself, "This is exactly the sort of sound-bite pseudo-science that Language Log is regularly frothing at the mouth over." I'm tempted to e-mail the instructor the relevant url but I suspect he isn't allowed to deviate from his script.
woodsspoom said,
May 31, 2009 @ 2:28 am
These "category" drawings are actually an example of what Will Eisner calls "expressive anatomy" in his book Comics and Sequential Art.
Eisner points out that because of the economy of comics, artists have to use lots of visual shorthand to convey emotion or personality or intention. (Scott McCloud expands on Eisner in his book Understanding Comics.) If the reader of these little drawings sees an amusing "truth" in them, I think all credit goes to the artist who realized the portrayal!
What makes the "types" recognizable to people with bar or party experience, or especially with experience watching movie and TV actors portray people having bar and party experiences, is that these poses are formulaic (and culture-specific).
If there is scientific study of body language meant to ultimately provide a valid key to our personalities in the poses we strike, researchers will have to carefully tease out all manner of influences affecting the subject and the observer. Interviewing actors and directors and observing acting classes might be an interesting place to look into, as acting is the art of manipulating one's body to convey just such things as "Jack the Ladness", et al.
woodsspoom said,
May 31, 2009 @ 2:52 am
Oops. McCloud takes on body language (and facial expression) in his book Making Comics, actually.
Ray Girvan said,
May 31, 2009 @ 7:07 am
woodsspoom: acting is the art of manipulating one's body to convey just such things as "Jack the Ladness", et al.
And, re this topic, Robert Carlyle used body language of glass-holding in his portrayal of Begbie in Trainspotting:
kip said,
May 31, 2009 @ 9:02 am
Well if no one else will say it, I will. She's clearly closest to the "gossip" position. Which is appropriate, since she's helping to spread lies.
Thomas Westgard said,
May 31, 2009 @ 12:39 pm
You're not wrong, Kip, but I would go with The Browbeater. Also has ties to advertising.
Thomas Westgard said,
May 31, 2009 @ 1:13 pm
The more I look at this study and think about it, the less of a travesty it seems. This blog rightly points to low-quality science reporting, and that's a great service because there are so many shallow stories on deep topics, and few people point that out.
I just don't think this is a good example of that problem. If you look at what's being said, and where and how it's being said, this is a reasonable depiction of preliminary observations.
The problem of bad science writing is severe. Presumably, readers of this blog are familiar with the recent admission by Merck and Reed-Elsevier that they conspired to create six fake "peer-reviewed" journals that were actually secret publications promoting Merck products:
http://www.onthemedia.org/transcripts/2009/05/29/04
http://www.nytimes.com/2009/05/14/business/14vioxxside.html
No doubt we will get some of that good old lawyer-bashing (another journo-trope) when they get sued for false advertising. Point being that even a well-informed reader can't even trust an apparently legit science journal to publish science.
The Walkabout Bars Drink Positioning Study is, as far as I can tell, originally published in popular dailies of general distribution. That is to say, it's not being clothed in a hard-science mantle. What's more, it discloses that it was one researcher observing only 500 subjects in an uncontrolled environment. Finally, it discloses the funder of the research. (As a side note, I observe that I have just proven that the scientific ethics practice of Walkabout Bars is superior to that of Merck or Reed-Elsevier.)
[(myl) The work is presented as if it were the fruit of scientific research. The Telegraph story, for example, is headlined "The way you hold your drink reveals key personality traits, claim psychologists: The way people hold their drink reveals their personality, a new study claims". It's filed in the category of "Science and Technology > Science > Science News" by "Richard Alleyne, Science Correspondent".
And the quality of the work seems remarkably low, though of course it's hard to tell since there's no report other than the press release. In particular, there's no argument that the observed behavior is stable across occasions, as you'd expect a sign of personality to be; there's no argument that there's any intersubjective consistency in categorizing a given drink-holding posture by different observers; there's no argument that the drink-holding categories correlate with any other measures of personality; etc., etc. If this is true, then the work not only wasn't published (other than as a press release), it wouldn't have had a prayer of being accepted by the journals or conferences where Dr. Wilson publishes his real research.
As far as I can judge, the bar chain hired Dr. Wilson to spend a day or two giving them some advice, just as (in the earlier email-is-worse-than-pot case) the computer company hired him to give them some advice; and they then used the results to persuade a bunch of news outlets to give them some free publicity — masquerading as science reporting.
There are at least two problems here. The first one is that "Richard Alleyne, Science Correspondent" and his editors at the Telegraph apparently can't tell the difference between scientific research and publicity stunts, or don't care to do so. The second issue is that Dr. Glenn Wilson, after being mis-used (he says) in an earlier scheme of the same sort, has lent his name to one again. It was the second of these that Mind Hacks (and I) commented on. ]
While individual readers may misuse any information, this study really doesn't look abusive to me. Wilson's drink-positioning research is shallow, but it's not being promoted in a way that is likely to induce reliance by reasonable readers. In theory, at least, it could be a jumping-off point for further research of a more rigorous nature into whether and how drink positioning is an indication of the person's personality or emotional state.
Another contrast – I just watched the first Mr. Bean movie, and was shocked and disturbed to find prominent Marlboro advertising. So much for loving that franchise. There are far greater tragedies than a shallow report on preliminary observations promoted as such.
[(myl) The whole problem of crappy science reporting is pretty far from the top of the list of ills afflicting humanity. But within the context of that problem, the mindless promulgation of press releases is a central contributing factor; it's bad enough when the press releases are misleading accounts of real research; and it's worse if editors allow "science" sections to be filled by articles based on commercial press releases whose scientific background is essentially non-existent, and "scientists" allow their names and qualifications to be used to give these articles a false appearance of credibility. ]
tablogloid said,
May 31, 2009 @ 5:22 pm
Gosh! We have forgottten the crossposers. Of course, the woman in question is " The Flirt/Ice Queen".
[(myl) I'm disappointed that only a few commenters have assigned the Telegraph's photo to one of Wilson's categories. My point is asking the question was to underline the fact that most concrete instances of drink-holding are ambiguous relative to the described categories; and to suggest that different coders would not categorize them consistently.
So far we have "clearly closest to the 'gossip' position", "I would go with The Browbeater", and "Of course, the woman in question is 'The Flirt/Ice Queen'". Others have gone off-script to suggest "Don't you want to buy [my] product?", "This is washing-up liquid?", etc.
I agree with everyone. ]
Tim Silverman said,
June 1, 2009 @ 9:11 am
To clarify, and support what myl said in replying to Tablogloid: my "This is washing up liquid!" piece of interior monologue was not simply there to make people laugh (though I hope some of you did that too), but also to make the semi-serious point that, when asked to explain why person A is engaging in behaviour X, my first thought is, "What is going through their head right now?" rather than, "What sort of person are they?" It's much easier to tell that someone you don't know is nervous right now than to divine that they are habitually nervous (and not all that easy to do even the first, precisely because you don't have knowledge of what they're usually like). Even leaving aside the dubiousness of the idea that there are only eight types of people in the world, you still don't want to go generalising from a sample size of one. Not ever. Just don't do it.
Of course, there are stereotyped signals people can send. However, not only are they often culture-specific, but—what with people possessing, like, brains and therefore having more sophisticated responses to the environment than, say, the average pebble—they can vary what signals they're sending depending on what's appropriate to the environment. Some people flirt more than others, but very few flirt indiscriminately with everybody all the time under all circumstances. Gah! Why am I even saying this? Anybody who judges whether someone's flirting with them based solely on the way they hold their drink is beyond help anyway. <Seethe, foam, gibber!> Basically, What John Said.
Sybil said,
June 2, 2009 @ 12:14 pm
How completely daft. Everyone knows that it's your choice in bar snacks that gives away your personality.
rpsms said,
June 2, 2009 @ 1:53 pm
Apparently, I'm Jack the lad: an ice-queenly browbeating fun-loving gossiping wallflower flirt playboy. Nice to meet you.
I honestly think I am likely to strike every single one of these poses at some point in an evening of drinking.