Some Help in Evaluating Subconscious, Implicit, System 1 Measures

If we claim a measure is “implicit”, let’s define the implicit features we’re measuring.

Editor’s Note: This post is part of our Big Ideas Series, a column highlighting the innovative thinking and thought leadership at IIeX events around the world. Paul Conner will be speaking at IIeX North America (June 12-14 in Atlanta). If you liked this article, you’ll LOVE IIeX NA. Click here to learn more.

By Paul Conner, Founder & CEO of Emotive Analytics

Ah, the subconscious! Many marketers are searching for what’s there because the subconscious can greatly influence consumer behavior.

There are many ways to see what’s in the subconscious. Psychophysiology (e.g., brain scans, biometrics, facial coding, etc.), metaphor elicitation and analysis, and implicit measurement are three prominent approaches. I can’t address all of them here, but I will comment on implicit measurement.

In marketing and consumer research, the subconscious is also referred to as nonconscious, implicit, and System 1. When it comes to implicit measurement, I’ll suggest the term “implicit” needs more clarification and should not be completely synonymous with subconscious, nonconscious, and System 1.

In our (Emotive Analytics’) implicit measurement work, we see confusion and disagreement on what is called implicit measurement. Specifically, we see measures characterized as “true implicit” versus “fast explicit”. The primary difference between the two comes from whether the stimulus of interest (SOI; e.g., a brand, ad, package design, etc.) is directly and consciously evaluated in the data collection process.

What are called “true implicit” measures use “indirect measurement” and sometimes, but not always, reaction time in their implicit scores. For these techniques, indirect measurement means that respondents are not directly, not consciously evaluating the SOI, but the SOI is incidentally influencing implicit measures.

What are called “fast explicit” measures use “direct measurement” and reaction time in their implicit scores. For these techniques, direct measurement means that respondents are directly, consciously evaluating the SOI. Implicit associations with SOIs are those that occur very fast (the exact threshold not universal), before explicit processing (a.k.a., “thinking” or System 2) has time to kick in.

Users of each approach call their technique “implicit”, implying that it measures subconscious, implicit, System 1 content. The true implicit camp claims that fast explicit approaches aren’t truly implicit because conscious reflection on an SOI, no matter how fast its evaluation, makes the process explicit.

Let me negotiate a truce. Jan De Houwer and Agnes Moors have extensively studied implicit processing and measurement. They suggest that implicit processing can possess any or all of the following features: uncontrolled, unintentional, goal independent, nonconscious (in one of several ways), efficient, and fast. Any or all allows many approaches to be called implicit, including those, like fast explicit, that involve direct, conscious evaluation of SOIs.

However, De Houwer and Moors (2012) strongly suggest that researchers identify the specific features that make up their implicit measure. In that way, clients can know, and make decisions based on, the features involved. For instance, if truly nonconscious associations are important, then fast explicit measures are not appropriate. However, if fast conscious reactions are OK, then fast explicit measures are OK, too, and they can be called implicit.

Let’s try following De Houwer and Moors’ advice. If we claim a measure is “implicit”, let’s define the implicit features we’re measuring. Furthermore, with those features defined, let’s make sure we use implicit measures with features consistent with our clients’ applications.

De Houwer, J. and Moors, A. (2012). What are Implicit and Explicit Processes? In R. Proctor & J. Capaldi (eds.), Psychology of Science, Implicit and Explicit Processes in the Psychology of Science.

You can leave a response, or trackback from your own site.

Leave a Reply

*