Troubling the Binary: Critically Thinking about the IAT
I have taken many IATs before and have found that the more I take them, the more curious I am about how they are developed, and what the test designers are hoping to achieve by creating a testing structure that seems to suggest our implicit associations exist in a binary.
According to Harvard’s Implicit Association Testing site, the “IAT measures associations between concepts (e.g., White American and Native American) and evaluations (e.g., American, Foreign). People are quicker to respond when items that are more closely related in their mind share the same button. For example, an implicit preference for White American relative to Native American means that you are faster to sort words when ‘White American’ and ‘American’ share a button relative to when ‘Native American’ and ‘American’ share a button.”
For the Native IAT, the test evaluates one’s ability to categorize White Americans and Native Americans as either American or Foreign. The test used older images of White and Native Americans, as well as natural landmarks in America and other countries around the world to represent ‘American’ and ‘Foreign’ respectively. I was already a bit confused by these tools of measurement before the test began, and was not surprised with my result.
Your responses suggested a moderate automatic association for American with Native American and Foreign with White American.
Alright, well that’s not entirely surprising for me. I was disappointed by the overall results for web respondents, but again, not exactly surprised.
This test posits that our implicit associations are deeply interlinked with visual cues, but when it comes to those visual cues in people, I wonder how much the test seems to only focus on a binary in which the answers are intended to reveal a specific, non-negotiable answer. For the Race IAT, it was White/Black (images) and Good/Bad (words). For the Disability IAT, it was Abled/Disabled (images) and Good/Bad (words). For the Native IAT, it was White/Native American (images) and American/Foreign (images). In the latter test, however, the American/Foreign association images were of natural landmarks in the continental U.S. and other countries and regions around the world.
My initial stumble then, with this association test, was that I associate Indigenous peoples with the stewarding and caring of land and earth around the world. All of my errors on this test were in the section that asked me to sort White Americans with images of American landmarks and Native Americans with images of foreign landmarks. When I did make a mistake, I noticed that my implicit reaction was to associate the American/Foreign landmark images, with the images of Indigenous Americans. (There is a lot to unpack there as well without a doubt.) Now, I am unsure that I can claim that my relationship to colonial history and Indigenous knowledges and worldviews is the reason I received the results I did, however, I feel confident in saying that not even the demographic surveys could have accounted for this difference in cultural context.
To return to the question of test design, or the setting of the stage. There was something else that bothered me about this test, beyond the use of natural landmarks: the Native-Foreign “debate” (if we can even call it that), is a current issue. While the colonization and settling of this country by white immigrants was initiated centuries ago, the Indigenous struggle against foreign association is still very much alive and well. So why not also use photos of Indigenous and White people today? This choice felt anachronistic, to say the least, and seems to further perpetuate the narrative that this is a historical event, rather than a contemporary struggle rooted in a series of violent historical processes.
There is so much to discuss and unpack here that I hope to bring to the table in class, but I will conclude with the following question, or call to action if you will: How could we re-imagine the IAT? Or is critical thinking about the way the test exists now, a more powerful approach?
