Friday, July 6, 2007

Test Your Bias

Reading a (Dutch) magazine on the toilet I chanced upon a column about discrimination and people applying for a job. The columnist (it's Ben Tiggelaar on the July 6, 2007 edition of Intermediair) wonders if discrimination is mostly a conscious or subconscious process.

(SF aficionados: cut to Peter Watts's Blindsight.)

So he recommends to do a test. His suggestion: Harvard's Implicit Association Test. I did. Well, that is: I did the first one, the skin tone IAT (they have a total of fourteen demo tests).

Tiggelaar says the following (translation mine):

Everyone who takes the test, is shocked by his/herself. The most people I know, reject discrimination in all its incarnations, but – in this Implicit Association Test – end up in the ‘Wilders’ category.
(Geert Wilders is the leader of the PVV, a Dutch political party at the far right of the political spectrum.)


My result:
I must say that I was relieved: somehow I suspected worse. Of course, there are thirteen more tests to take (which I will do after I finish the May slush).

Now the reason for Tiggelaar's column was that our Home Secretary (our minister in charge of the Home Office, whose title in Holland is Minister, and whose assistant is called Secretary. Confused already? Then go to Brussels...;-) Ter Horst is proposing to use anonymous job applications for government jobs. Which is throwing up a storm of protest and opinions: some think discrimination on the job market should be fought through education and agreements with the employers; others believe there is no discrimination; and some think all Turks should return to Marocco.

The point of his column -- and I must agree -- is that *if* discrimination is mainly a subconscious process, then an anonymous job application procedure, in which the first selection should be done while the name, sex, age, religion, and ethnic background of the applicant are withheld, is needed. Like Ter Horst wants.

Then -- Tiggelaar says -- the final selection committee should be a mirror image of our society (we're talking Holland here): 50% male, 50% female, 20% foreign, and graying hair at the temples.

Food for thought: while I still need to take the other thirteen tests (and don't know if these test actually do show subconscious bias), I do agree that the way we judge total strangers is most probably subconscious, for the most part.

Because I distinctly remember a course about dealing with customers I did some years ago, and the way first impressions work.

Question: At first impression, how fast do we -- on average -- make a judgment about a total stranger solely based on appearance?

Answer: in less than 10 seconds.

Question: How much of this first impression do we keep over time?

Answer: about 90%.

This baffled me at first, but I've come to believe that it's true. I find myself doing it: assessing strangers purely on appearance, without having talked to them. Then, if I find out I'm doing it, I have to make a mental effort to suppress this 'first impression judgment'.

I have to consciously force myself to withhold judgment until I know this person better.

Cutting back to Blindsight (and its theme that most of our actions are subconscious), I theorise that this might be a survival mechanism from the savannah: distinguish friend or foe very quickly. A mechanism that is not appropriate in modern society, but didn't have the time to evolve away (or, to play Devil's advocate, is waiting until this folly of a conscious culture has run its time).

Anyway, it makes me wonder: am I, deep inside, a sexist racist supremist alpha male whose conscious brain needs to supress these tendencies all the time (and probably unsuccesfully), or is my subconscious smarter than that, and is my conscious mind responsible, making quick calculations because it's taking up too much computing power as it is?

I don't know. But sometimes I feel like this:

behind the finer feelings--
this civilized veneer--
the heart of a lonely hunter
guards a dangerous frontier


As Rush drummer Neal Peart worded it on "Under Lock and Key" (from the Hold Your Fire album).

4 comments:

Artemisin said...

I've taken a couple of these tests and wonder how much part our unconscious views on what is morally correct play in the results. I would imagine that you can "push" for a certain answer without really realizing it.

I also wonder how much of this testing is specific to the USA and invalid elsewhere. In the black vs white test, I realized that neither the black nor white subjects looked like what is standard in my country (I'm from Spain, black people here are usually from Africa and much darker, white people are Mediterranean and also darker). In other studies, Spain has always scored lower on racism than other countries (although this is changing rapidly with the wave of immigration and Spaniards are getting more racist instead of less). If you were being machiavellan, you could say that Spaniards score lower on racism because they don't identify with the subjects, either black or white. There is none of that "me or other" that plays such a strong role in discrimination and xenophobia.

Anonymous said...

Going loopy over consciousness

This is a link to David Deutsch's review of
'I Am a Strange Loop'
Douglas Hofstadter
2007 Basic Books
412pp £14.99/$26.95 hb

http://physicsweb.org/articles/world/20/7/5/1

After Peter Watt's 'Blindsight' you should find this Physics World, July 07 book review interesting.
Roy

Jetse de Vries said...

>>>>'ve taken a couple of these tests and wonder how much part our unconscious views on what is morally correct play in the results. I would imagine that you can "push" for a certain answer without really realizing it.<<<<

Sara--

The whole point of these tests is to find out your *unconscious* bias. They should be devised for it!

Of course, I wonder if they're effective. And I do agree that they seem to be aimed at the USA.

I'll happily try a European one if there is one on the net.

In this post, I'm trying to grope how much of our prejudices are conscious or subconscious. I don't know -- although I suspect a lot of it is subconscious, but I want proof -- and I'd like to find out.

Roughly speaking, if our prejudices are conscious, then education might help. If, on the other hand, our prejudices are subsconscious, then anonymous applications (or other methods that force us not to take our first impressions as gospel) are needed.

In general I am very interested in root causes (and I realise these are often extremely difficult to unearth. But we should try to find them nevertheless).

Sorry to hear that Spaniards are gtting more racist (and I'm afraid the trend is the same in Holland).

Roy--

I read David Deutsch's "The Fabric of Reality", which is one of the very best popular science books I've read. All this Interzone malarky robs me of my pleasure reading...;-).

I've also got Hofstadter's famous "Godel, Escher, Bach", but haven't had time to read it, so far.

I'll check the review after the weekend, and it'll probably mean I'll buy another book for which I have no time to read (oh, to quit the day job, and do nothing but my hobbies: come on lottery ticket, you can do it!).

Artemisin said...

>>>>>Roughly speaking, if our prejudices are conscious, then education might help. If, on the other hand, our prejudices are subconscious, then anonymous applications (or other methods that force us not to take our first impressions as gospel) are needed.<<<<<
Hmm, yes this is true for the short term, as solving discrimination _now_, but I wonder if we can't modify our subconscious feelings through training (not just disregard first impressions like you suggested). That would seem to get to the ugly root of the problem, rather than simply finding tricks to bypass it (although I agree, for now, anonymous applications make sense)

By the way, I realized I didn't introduce myself when I made my first post. Sorry about that. I keep forgetting where I have introduced myself and where I haven't. I'm Sara Genge, I got here through your response to an Interzone sub.