Implicit Association Task - Guidelines
In the Implicit Association Task (IAT) participants are asked to categorize words into categories. The task is designed to measure the strength of automatic associations between concepts in a person’s memory. The IAT is often used to explore implicit biases or attitudes that individuals may hold unconsciously.
Visit Project Implicit for examples and complete at least one experiment there to understand the flow of an IAT.
For your IAT experiment, you can use one of the associations listed on the Project Implicit site or come up with your own. You may not use family/career and women/men for your association as we will use that in our examples described below.
Example IAT screens:
Demo
View run through of IAT demo project...
Structure
You will have 4 different blocks of trials. E.g.:
- One block classifying women vs. men words
- One block classifying family vs. career words
- Combination of women or family vs. men or career
- Combination of women or career vs. men or family
Across participants, block 1 and 2 will come in a random order, and 3 and 4 will come in a random order (e.g., sometimes 2,1,3,4; other times 1,2,4,3, etc.). This flow is set up for you in the conditions array described below...
Conditions
To structure your experiment, use the following conditions file: iat-conditions.js (click Raw to see a version of the code you copy).
Here are the steps you should take to work with this file:
- Add it to your project and make sure it’s imported via your
index.html
file beforeexperiment.js
. - Skim through the entire contents of the file to get a broad understanding of how it works.
- Load your project in the browser - you should see the resulting
conditions
array from that file output to the console. Carefully study the structure and data of theconditions
array so you understand it, as you will be building your experiment around it. - Update the top of the file with values that are applicable to your project. Refresh the browser as necessary to check the resulting array meets your expectations.
Once you have studied and modified the conditions file, you can move on to programming your experiment around it.
Mechanics
Between each block should be an instruction screen introducing the new categories and re-iterating the instructions. E.g.:
Between each word should be a fixation screen that lasts 250ms, e.g.: