0:00:00.780,0:00:04.631
>>Behind every innovation[br]is an inspired scientist.

0:00:04.631,0:00:08.891
And behind every PBS program[br]is a viewer like you!

0:00:08.891,0:00:11.152
Get behind your PBS station.

0:00:11.922,0:00:15.070
Your support makes a difference!

0:00:15.530,0:00:18.550
>>Hello and welcome to[br]Scientific American Frontiers,

0:00:18.550,0:00:19.880
I'm Alan Alda.

0:00:20.451,0:00:24.039
we all like to think that we[br]understand our own minds.

0:00:24.039,0:00:27.958
That we carefully weigh the pros and cons[br]when we make a decision.

0:00:27.958,0:00:29.699
And that after we've made it,

0:00:29.699,0:00:33.870
even if we have cause to regret it, we[br]know why we decided the way we did.

0:00:33.870,0:00:37.923
Oh sure we know some choices[br]are more emotional than rational,

0:00:37.923,0:00:40.285
but even then, we think[br]we're conscious

0:00:40.285,0:00:44.357
of all the conflicting arguments that run[br]through our minds as we make a choice.

0:00:45.371,0:00:49.868
Well, in this program we'll find out[br]how utterly deluded we are,

0:00:49.868,0:00:55.650
as we join in experiments revealing just how[br]sneaky and underhanded our brains can be.

0:00:56.493,0:01:02.992
[music]

0:01:04.159,0:01:06.508
>>I've always thought[br]of myself as a feminist,

0:01:06.508,0:01:09.950
so I'm pretty sure I know how I'll[br]do in this test of my reaction

0:01:09.950,0:01:11.748
to women in the work place.

0:01:11.748,0:01:13.498
I am ready to begin.

0:01:13.498,0:01:19.258
Women, in fact like Mahzarin Banaji, who's[br]a professor here at Harvard University.

0:01:19.258,0:01:22.499
The test is called the[br]"implicit association test"

0:01:22.499,0:01:24.568
and it begins simply enough.

0:01:24.568,0:01:26.657
I have to pair the word in the center,[br]

0:01:26.657,0:01:30.340
with one of the words above by[br]pressing the "e" key with my left hand

0:01:30.340,0:01:32.695
or the "i" key with my right.

0:01:32.695,0:01:34.875
Well, that could be either way.

0:01:34.875,0:01:37.405
Mahzarin has told me to[br]do this as fast as I can

0:01:37.405,0:01:42.146
because it's the time I take to make the[br]associations that's critical to the test.

0:01:44.280,0:01:48.560
Now the target words have changed,[br]but the task remains the same;

0:01:48.560,0:01:52.824
to quickly decide whether the[br]new words belong to the left or the right.

0:01:56.004,0:01:58.286
But things are about[br]to get trickier.

0:01:59.884,0:02:03.043
>>It's the same thing except now any[br]one of these four will show up,

0:02:03.043,0:02:05.295
and when it's "career" or "male"[br]you'll press the left key,

0:02:05.295,0:02:08.315
when it's "family" or "female" you'll[br]press the right key.

0:02:08.825,0:02:11.464
>>"Career" or "male"[br]>>Yes.

0:02:11.464,0:02:14.945
>>So now the categories are[br]described by two words,

0:02:14.945,0:02:17.883
making it harder to decide[br]where the new words belong.

0:02:18.783,0:02:23.273
But since historically, "male" and[br]"career" have gone together....

0:02:23.273,0:02:27.394
This is like, reinforcing this stereotype.[br]>>Yeah, that's it exactly.

0:02:27.954,0:02:32.595
>>The point of the test is to discover[br]if lurking beneath my feminist convictions,

0:02:32.595,0:02:36.866
I actually harbor a hidden bias[br]against women in the work place

0:02:36.866,0:02:41.665
based on all the associations between[br]"man" and "career" and "women" and "family"

0:02:41.665,0:02:44.159
that bathed the culture[br]I grew up in.

0:02:44.159,0:02:49.009
"Family or "male", "career" or "female".[br]Okay, now you're testing me.

0:02:49.199,0:02:53.600
The implicit association test[br]is designed to ferret out any bias

0:02:53.600,0:02:58.389
by seeing if it takes me fractionally longer[br]to figure out where a word belongs

0:02:58.389,0:03:02.989
when the pairing of the target words, in[br]this case: "family" and "male" together

0:03:02.989,0:03:08.578
and "career" and "female", is slightly[br]more difficult for my brain to except.

0:03:08.578,0:03:12.939
Corporation - "family"[br]or...[breaks off in laughter]

0:03:12.939,0:03:16.379
"brother", "management"...

0:03:17.780,0:03:19.251
and we're done.

0:03:20.872,0:03:24.502
>>Slight. I could tell he has a slight.[br]This is very good.

0:03:24.502,0:03:26.952
You are...so let's see,[br]you're showing a slight,

0:03:26.952,0:03:29.551
automatic association[br]between male and career,

0:03:29.551,0:03:34.903
and the...only 12% of the[br]population that takes this test

0:03:34.903,0:03:38.102
shows this bias. What you're[br]seeing is that you're,

0:03:38.102,0:03:41.071
you're showing a much smaller bias[br]then what many other people show.

0:03:41.071,0:03:44.320
I'm up here. I make a strong[br]association between

0:03:44.320,0:03:47.772
"male" and "career" and "female" and[br]"family" even though that's not what I

0:03:47.772,0:03:49.232
consciously express.

0:03:49.232,0:03:50.673
>>Even though that's[br]not what you live, huh?

0:03:50.673,0:03:52.331
>>And that's not what I live,[br]but in my world.

0:03:52.331,0:03:54.781
And I don't know, I'd be[br]interested to know. Maybe

0:03:54.781,0:03:58.680
many years of working in feminist[br]causes made you show this lesson.

0:03:58.680,0:04:01.442
>>So all those years of working[br]in feminist causes

0:04:01.442,0:04:06.490
didn't manage to totally eliminate my lingering[br]association between "female" and "family"

0:04:06.490,0:04:09.542
and "male" and "career".

0:04:09.542,0:04:12.990
What's astonishing though is that[br]Mahzarin is far worse,

0:04:12.990,0:04:16.463
and she's not only enjoying[br]a very successful career-

0:04:16.463,0:04:18.233
she designed the test!

0:04:18.894,0:04:21.854
Another of its designers[br]is Brian Nosek,

0:04:21.854,0:04:24.473
who personally developed[br]the test he's taking now,

0:04:24.473,0:04:27.594
intending to reveal hidden[br]racial prejudice.

0:04:27.594,0:04:29.864
>>These are the same faces[br]you've seen many, many times.

0:04:29.864,0:04:33.633
>>Many times, many times. In fact I helped[br]create the faces, so...

0:04:33.633,0:04:36.974
>>The heart of this test is to see whether[br]it's easier for Brian to match words

0:04:36.974,0:04:40.743
or pictures to the pairing[br]of African American with bad,

0:04:40.743,0:04:43.282
and European American with good.

0:04:43.742,0:04:45.934
Then when the pairings[br]are reversed,

0:04:45.934,0:04:50.424
European Americans with bad and[br]African Americans with good.

0:04:50.985,0:04:55.445
[quiet]

0:04:55.445,0:04:57.533
>>I can predict.[br]>>Yeah [chuckle]

0:04:57.533,0:04:58.583
What do you think?

0:04:58.583,0:05:00.904
>>Moderate. Moderate to strong.

0:05:00.904,0:05:03.455
>>Yeah, I'll say moderate as well.[br]>>Yeah.

0:05:03.455,0:05:06.335
>>Strong preference for white.[br][exclamation of laughter]

0:05:06.335,0:05:09.205
>>You've taken this[br]for the umpteenth time,

0:05:09.575,0:05:14.055
and you still haven't caught on[br]to the fact that you're a little biased?

0:05:14.055,0:05:15.975
[laughter]

0:05:15.975,0:05:21.526
>>I think part of that is the insistence[br]that my conscious beliefs still matter.

0:05:21.526,0:05:25.276
It isn't that-- that fact that I[br]keep showing these implicit biases

0:05:25.276,0:05:28.577
mean that I'm a biased person so[br]I should just accept that and move on,

0:05:28.577,0:05:32.635
it's that-- no, I don't agree with[br]those, I do have them

0:05:32.635,0:05:35.065
and I will admit to having[br]those implicit biases

0:05:35.065,0:05:38.667
but I'm not going to let that rule[br]what I would consciously want to have.

0:05:39.927,0:05:43.335
>>It really is remarkable that here I am[br]in Cambridge, Massachusetts--

0:05:43.335,0:05:45.555
perhaps the liberal[br]capitol of the country--

0:05:45.555,0:05:49.904
with two academics who pride themselves[br]on their enlightened attitudes,

0:05:49.904,0:05:52.246
only to discover that they[br]have latent within in them

0:05:52.246,0:05:56.731
biases they would fervently deny[br]if their own test hadn't revealed them.

0:05:59.581,0:06:01.569
But all may not be lost.

0:06:02.202,0:06:05.794
>>We found in research, and[br]also in testing of myself,

0:06:05.794,0:06:10.685
that if I put myself in a situation where[br]I think about positive black exemplars--

0:06:10.685,0:06:14.402
I think about Michael Jordan[br]and Colin Powell and people...

0:06:14.402,0:06:19.396
Martin Luther King who have had a very positive[br]impact and who are also African American,

0:06:19.396,0:06:23.153
I show much less bias immediately[br]after thinking about those exemplars

0:06:23.153,0:06:25.355
then if I hadn't thought[br]about them before.

0:06:25.355,0:06:29.715
>>Even simple things, like the presence[br]of an African American experimenter,

0:06:29.715,0:06:34.943
reduces race bias. That the presence[br]of a competent woman,

0:06:34.943,0:06:38.085
makes women's attitudes to math[br]become more positive.

0:06:38.085,0:06:40.575
Those kinds of studies

0:06:40.575,0:06:44.515
sit, I think, in contradiction the ways[br]in which people like I thought about this-

0:06:44.515,0:06:46.654
we thought they were learned[br]over long periods of time,

0:06:46.654,0:06:49.143
that they were entrenched,[br]they were rigid, inflexible--

0:06:49.143,0:06:51.224
and that does not appear[br]to be the case,

0:06:51.224,0:06:53.534
and I think that that's where[br]the room for optimism is.

0:06:53.534,0:06:56.805
On the one hand, I would like people[br]to take these data seriously

0:06:56.805,0:07:00.967
when we can show the vast numbers[br]of people who show the bias,

0:07:00.967,0:07:02.945
I think we need to contend with that.

0:07:02.945,0:07:07.224
On the other hand, what this work[br]is showing is that it might be

0:07:07.224,0:07:10.755
not that hard to shape environments[br]that can change these,

0:07:10.755,0:07:13.878
even automatic kinds of attitudes.

0:07:15.188,0:07:19.806
>>You can find out if you harbor unconscious[br]biases you would deny even to yourself,

0:07:19.806,0:07:23.536
by logging on to the[br]Project Implicit website.

0:07:23.536,0:07:27.868
There you'll find tests of attitudes towards[br]such things as age and religion,

0:07:27.868,0:07:33.149
as well as lighter fair like the Harry[br]Potter movies versus Lord of The Rings.

0:07:33.878,0:07:37.256
Give it a try. You may be in for a shock.