What are implicit measures?

Implicit measures is the general name for measures that were developed to measure social cognition (attitudes, self-concept and stereotypes) indirectly. It is plausible that implicit measures capture the effects of unintentional activation of social cognition. Unintentional processes might occur very quickly, with no need for much cognitive resources, and without awareness. However, much is still unknown about the psychometric properties of implicit measures, and they are still a matter of active research.

What can implicit measures do for me?

Investigating, improving, and developing indirect measures of psychological constructs might be a useful pursuit for improving human knowledge of these constructs and about the automatic and non-automatic processes that underlie them. That is the main reason that our lab investigates implicit measures. If that kind of research seems attractive to you, you are welcome to join us.

Many researchers are interested in implicit measures because they search for better measures of the constructs that they study. This is a reasonable and common research direction. However, it should be clear that it is very difficult to know what these measures capture. When you are the first to study a topic or a mental construct with indirect measures, your research would be exciting because it would be characterized by many uncertainties, and you would be the first to collect evidence that can reduce those uncertainties. However, that kind of pioneering research is less suitable for those who hope for a simple and clear indirect measure of the mental construct they seek to study. Further, using implicit measures requires some expertise in analyzing reaction time data. It also requires a certain capacity for using appropriate technologies to implement those measures within a computerized study.

The implicit measures

The best known implicit measure is the Implicit Association Test (IAT). To learn about the IAT, it is best to first visit the websites of Tony Greenwald, Mahzarin Banaji, and Brian Nosek. A great source for best practices is this paper.
The IAT has many weaknesses, but it is currently the best implicit measure. There are more than a dozen other implicit measures. To learn more about some of them, you can read these chapters by Gawronski & De Houwer (2014), and De Houwer & Moors (2010). Another source for much knowledge about many implicit measures is this empirical research project from our lab.

If you want to exprience a few implicit measures, here are online examples from our lab.

If you want your students to experience the IAT, send them to Project Implicit’s demonstration website.

Project Implicit

Project Implicit is a non-profit organization and international collaboration between researchers who are interested in implicit social cognition - thoughts and feelings outside of conscious awareness and control. The goal of the organization is to educate the public about hidden biases and to provide a “virtual laboratory” for collecting data on the Internet. Our lab is involved in maintaining and improving Projec Implicit’s technologies (mainly the software that enables building online studies).

IAT in Israel

During her time in my lab, Mayan Navon led a new translation of Project Implicit’s website to Hebrew. If you want your students to experience the IAT, send them there.

Mayan Navon also processed the data collected in Project Implicit’s Hebrew website from 2009 to 2019. Read her post to learn more about it.

How to program implicit measures

You can program reaction time tasks with most of the programs that were developed for this purpose (e.g., Open Sesame and Psychopy). Michael Pinus, a PhD student in our lab, has created a few Open Sesame IAT scripts. You can find them here.

If you want Project Implicit to help you plan, program and carry out your study, please contact services@projectimplicit.net.

Project Implicit’s software, Minno.js, is open source, and you can see it here (documentation of our player).

There are currently five methods for creating and running studies using Minno.js:

  1. Run MinnoJS studies for free, using free services such as Github Pages, DataPipe, and the OSF. You can read more about it in this post.
  2. Install our whole platform. The platform includes a website for authoring and managing web studies. This option requires installing the platform on your own website (the documentation includes a guide how to get such a website for free or for cheap). It has not been thoroughly tested, but some have used it successfully. This is the most recommended option for a long-term use of our platform.
  3. Install on a PhP server. This option would allow you to run web studies, but it does not include a website for creating the study. You will need to program the studies on your own computer and upload them to this servier.
  4. Run one task from Qualtrics. We have stopped maintaining the code for those solutions, but the code still works. If you have a Qualtrics account and you just want to add a reaction-time task to your study, you can read some guides how to add to your Qualtrics study Project Implicit’s scripts for the IAT, Single-Target IAT, Brief-IAT, Evaluative Priming, and the Affect Misattribution Procedure.
  5. Run Minno locally. This will allow you to run studies programmed by Minno.js on your computer, not as a web study.