Implicit Association Test - Mahzarin Banaji

Source - serious-science.org/videos/433
Harvard Prof. Mahzarin Banaji on detecting the strength of associations, indirect measures of cognition, and how to find the truth about our mind

[@naziratolobekkyzy2723]

In my opinion, because of the first option putting together coloured people's faces with bad things and then changing it makes it difficult for our brains, not that some people are biased, because of the structure. I would assume if, in the beginning, white people with bad feelings, things and then changing it will demonstrate otherwise. But overall, I believe that most of us, if not all, are biased depending on our cultural, educational or ethnic backgrounds and outside institutions' influences.

[@nancyjaramillo1080]

I thought that too. What would the results be if we started off with black and good/ white and bad...and then switched it?

[@beatrizm9451]

I'm pretty sure they are smart enough to have done it both ways and have come to the same conclusion

[@immaghost2109]

yeah I hope they don't always do it in that order

[@lignodrenful]

From the IAT website. I see it after taking the test, but I am sure it appears elsewhere. "Does the order in which I took the IAT matter? The order in which you take the test can influence your results, but the effect is small. We minimize this effect by giving practice trials after the categories switch sides. We also randomly assign the order of the IAT so that some people get one order and other people get the reverse order."

[@paulag1805]

I would really like to see the results when the pairing starts off the other way around first. Re-learning an already learned pattern takes more time than just learning it in the first place, no?

[@lignodrenful]

From the IAT website. I see it after taking the test, but I am sure it appears elsewhere. "Does the order in which I took the IAT matter? The order in which you take the test can influence your results, but the effect is small. We minimize this effect by giving practice trials after the categories switch sides. We also randomly assign the order of the IAT so that some people get one order and other people get the reverse order."

[@BarnardClangdeggin]

Has Minaji yet publicly stated that the reliability and validity of the IAT makes it an unworthy psychometric test?

[@supernova743]

The mind works on patterns. Associating white with good and black with evil first will train your brain to put those together. Then when you switch to Good with Black and White with evil it will take longer as you first have to untrain the earlier association. The same with the cards, separating red and black is easy as you only have two options. You take twice as long to sort the cards when the colors don't match because you're not sorting two colors but 4 different shapes. You have to stop to identify the shape instead of just recognizing the color. Doing the color test first will influence the second test at the start as you have the same situation (sorting cards) but you need to unlearn the previous sorting rules. I really can't see how you could tell anything from these association tests other than the person's relative ability to change their own though processes.

[@Phaze252]

+supernova743 This is something that I find very odd that they completely overlooked because IIRC Psychologists are the ones that helped figure this out, while they completely neglect this as an explanation for the time difference.

[@DarkProvidence420]

+supernova743 They first acknowledge that mistakes are made when making associations concerning card suits, a situation from which you can determine no bias. Then when mistakes are made concerning associations that could contain biases, they attribute those mistakes as confirmation of bias, even though they have already established that mistakes are to be expected.

[@porky5567]

+DarkProvidence420 Their just incredibly dumb

[@EpicUXnl]

+supernova743 Yeahhhhhhhh you think they haven't thought of that?

[@Phaze252]

Florence Gee Any evidence that they did?

[@SamyTheSheep]

So surely the next step for the test (assuming we're yet to know) is to determine whether there is a relationship between high performing card sorters and lower implicit bias.

[@Cliffhouse97]

The IAT does not meet commonly accepted standards of validity and reliability as a psychometric test. She either does not know that or she does and has an agenda that lies outside science. Given her curriculum vitae, the former is unlikely.

[@jimiekuester7298]

You are correct it has been brought to her attention and clearly she has a agenda.

[@gabbyscott8181]

What would be the independent variable in this test?

[@Tobi-pn2xs]

What the good and what the bad words are associated with

[@emmanuelalfredoruizacacio4474]

Ciertamente sufrimos en psicología por las medidas de autoreporte, ya era hora de que apareciera algo mejor para medir, ahora es cuestión de seguir y seguir haciendo investigación con este tipo de aproximación. Espero poder aportar en este campo. Gracias por el vídeo.

[@rogermaris5]

What about the other scientist who developed this and found it has NO positive effect. Also the test looks for prejudice towards the same group NOT prejudoce against an out group. Please respond to this.

[@darlenepoisson4659]

Do I need to take a test to reveal my bias regarding her glasses?

[@SamyTheSheep]

certainly. please do share the results.

[@sonyaberle797]

I am confused by the methodology used here and the connectivity provided by the exam format. The test feels, to me, quite remiss in accounting for intellect and layering of daily experiences. "Good and bad" cannot so easily be categorized, I believe. As well, I'm not interested in pressing a button that connects "Black and bad." I did not complete the test. I feel internalized racism is present in the test. My apologies for the comment and I support the critically important research surround racism. I'm just not sure this test contains adaptability for variation of thought and doesn't include objective enough questions, in my opinion. I feel it can reinforce racism.

[@SamyTheSheep]

Indeed that may well be the purpose of the test, to scientifically confirm the presence of racism. I think we can safely say after a quarter of a century that this venture has failed. The speaker even said so herself, that another measure is likely to come along and consign the IAT to the scrap heap.

[@Jasnogorodskij]

Jordan Peterson brought me here. Not only the fact, that we all have certain prejudices is trivial and vacuous but the measures, which had been described in the video are completely indistinguishable from novelty response. Why not measure how fast Arabs can take 'jewish' and 'good' together? Because specific groups won't find the answer nice and politically correct.

[@The_Real_Devil]

Ah I did wonder why so many of the comments here are ridiculous, like "she's a marxist" and "she has an agenda". Thank you for explaining what the cause of that is.

[@felixms]

The dark or "black" face test vs good or bad thing has a strong bias to the general concept from yin yang, black magic, dark side, etc. rather than person stereotypes. Is there another test to figure out where the connectors or implicit association is coming from? I imagine that would better describe an individual.

[@CanadianPrepper]

Yeah and what impact do you think calling black people "black" has then. Think about it.

Canadian Prepper you cant be serious.

[@dandobrowolski5818]

The problem with IAT is it ignores how attitudes are informed by experience. We cannot presume that everyone's experience with a particular group is equally positive or negative based on real world experiences. The assumption with the IAT is implicit bias is human fault that must be corrected...as opposed to a normal response to external experiences. For instance, if you lived in a neighborhood dominated by group x and it's accompanying cultural norms you may have positive association with that group if your experiences with that group are positive....if your experience is negative with group x then you will exhibit a bias against that group....the bias measured maybe completely justified based in the individual experience.

[@AbCDef-zs6uj]

This doesn't seem like serious science.

[@luketravels4472]

Can you comment on the reliability and validity of these tests? You already said that it is a ‘rough measure.’ This is an extreme misuse of psychology. You only promote this as you het laid and get cited a lot. The fact that you or Anthony Greenwald will not comment on anything to do with reliability or validity proves the fact that you a promoting something that ridiculous. Plenty of evidence has surfaced proving that this test is not reliable and not valid. A diagnostic test must have a minimum of 0.8 reliability.... this test is way below this. To ONLY view the world in binaries is stupid. Do binaries exist? Absolutely! Do they exist across all dimensions? Absolutely not! You’re playing fast and loose with the term and really pushing post-modernist rhetoric with a Marxist basis. You should probably have your credentials revoked for promoting such bad, unreliable science.

[@The_Real_Devil]

I was thinking you might have a point until you said "Marxist" lol.

[@paradiddlemcflam7167]

The implicit association test is not serious science, at least not in the way it has been politicized, promoted, and made to seem far more reliable and valid than the evidence suggests.

[@Cliffhouse97]

Agreed. But if you point that out, you are accused of being a racist.

[@zadeh79]

But biases are also what make people smart

[@TheNameIwantedWasTkn]

funny, bother her and and anthony greenwald have both said that these test results are not repeatable or reliable...

[@napoleon_bonaparte2462]

This research has been invaluable to me, though I was aware of my own personal bias, this emphasis has made me realize how much I value Western culture. It's like Banaji has pointed out that I have a penis, so I should more consciously act like a man, that is very encouraging, thankyou.

Oh..its become quite popular. Thats a solid scientific premise. Does she admit that its 'popular' because it attempts to confirm a presupposed ideology? Does she admit further that it is unsuccessful and inconclusive at doing so? This bullshit COULD NOT BE ANY MORE UNSCIENTIFIC. She is proof that 'arguing from authority' is fallacious, at best, and more likely dangerous. This woman, and many other social psychologists, sociologists, et al are all members of highly influential groups, such as American Psychological Assoc., a Fellow on suchnsuch, Trustee of thisnthat, etc. Let that sink in. This garbage has invaded HR depts throughout the business world now....how many more red flags are needed?

[@Zeno2Day]

Normative science

[@rogersyversen3633]

I dont agree that this is not a measure of racism. people that say it does not believe in the ability of rationality to override other functions of the mind. some introspective knowledges are self enforcing ond some knowledges are enforced by the self.

[@mademsoisellerhapsody]

There is inherent bias in the Implicit Association Test. Seriously what do you associate "joy" with and your choices are "white" or "black" when everyone knows "joy" is golden.

[@pawew8264]

This is wrong on so many levels... As always psychology never fails to impress. Test designed on a assertion with no back up. And up on this we build our hypothesis... Just great...

[@CanadianPrepper]

You have absolutely no idea what you are typing about.

[@zo1dberg]

Great response, clearly you proved Pawel wrong. kzbin.info/www/bejne/hperaXx9bKaqoZIh23m5s

[@pumpkinking365]

Im with you on this Pawel, these type of people came up with the idea of giving electric shocks to the head, thinking it will fix people. Not many people argued with them then & not many people will argue with them now.

[@ibperson7765]

It’s wrong because being correct was never part of the goal. Attacking white people was and is the goal. The left has become racist; they use science not to seek truth but to seek power, and merely dress it up to be sciency. There was never any chance of this field providing accurate ways of understanding or measuring the world.

[@jimhalfpenny442]

@@pumpkinking365 On a side note, though, ECT has actually made a comeback and rightfully so. It has clinical value in some limited applications. Though, obviously it was overused in the past and was ineffective for many patients.

[@andrewhewitt5370]

So we're are measuring a ghost basically since we are unable to see it. Pseudoscience perhaps?

[@renoesmaeilian9489]

And when you say “how fast” what does that exactly mean. My idea of “fast” is not the same as yours.

[@phuongtam951]

it's reaction time, a method used in psych testing

[@ronmaest]

Too inconclusive. Too much hyperbole. Too much convenience.

[@shaikhnbake]

Is it only me, or do her glasses piss you off

[@dalebolger1969]

i want her glasses, they are amazing

[@The_Real_Devil]

Is it because they kinda look like a rainbow flag? And that upsets you on an implicit level?

[@shaikhnbake]

@@The_Real_Devil bruh why are you replying to comments that are 4 years old ahahahahah

[@nickduffy8919]

Ummm, using card category that is used in poker or deck make up as unconscious decisions compared to color is such BS. Color has been around more then 100,000 years in our evolution. A deck of cards a couple hundred years maybe. And they call that objective science wow!

[@michaelbeard3393]

What complete and utter nonsense. People don't seriously consider this science do they? You can conclude nothing from such a test other than what you proclaim it can tell you, without appeal to anything else. Sounds more like some sort of Christmas drinking game to me.

[@PlaybackMansion]

All cops should have to take this type of test to determine if they are menace to humanity.