Home About me Research Vita Try my studies

Research Overview

In my research I focus on what are most commonly called 'implicit attitudes', which are those evaluations that exist relatively outside of our conscious awareness, are activated automatically, and are difficult to actively control.

I am particularly interested in how we can enable people to be more aware of their implicit evaluations (i.e., more able to report them to themselves and to others), what behaviors are related to implicit evaluations, and how implicit evaluations can be changed.

Most measures of implicit evaluations operate based on the idea that concepts that are related in the mind automatically activate each other. So, when you hear the word 'Doctor', the word 'Nurse' likely comes to mind quite quickly. And the stronger this association is for you, the more quickly it will come to mind.

Most of the measures we use to index implicit evaluations, then, are based on how quickly you can respond to, for example, faces of African Americans paired with good words compared to bad words. Often, the easiest way to understand the measures is to try them. You can do that by completing a Demonstration task or signing up for Research at Project Implicit.

Relationship between Implicit and Explicit Evaluations

One way to understand what implicit evaluations are is to understand what they are related to. The relationship between implicit and explicit evaluations varies widely based on a number of factors surrounding the attitude object and the individual involved (e.g., Hofmann, Gschwendner, Nosek, & Schmitt, 2005). In my own work, I have found that implicit evaluations are more related to the attitudes that people report when they report them very quickly, or when they report their 'gut reactions' as compared to their 'considered responses' (Ranganath, Smith, & Nosek, 2008). This implies that our initial, gut-level responses give us more information about our implicit evaluations as compared to the attitudes we report when we've given a lot of thought to our response.

I have also found that the attitudes that people report are more related to their implicit evaluations when they focus on their feelings and emotions as compared to their thoughts and beliefs (Smith & Nosek, 2011) which I take as evidence that implicit evaluations are more related to emotional processing than to higher-order thinking. Additionally, implicit and explicit evaluations are more related when people are in a good mood than when they are in a bad mood (Huntsinger & Smith, 2009).

Tying these findings together, having people focus on their gut-level emotional responses while completing an implicit measure (Affective Misattribution Procedure) leads to larger effects on that measure (De Houwer & Smith, in press).

Relationship between Implicit Evaluations and Behavior

One reason (though far from the only reason) we study evaluations is because they are related to behavior. Implicit evaluations have most often been shown to be related to the types of behaviors that we perform relatively automatically (like seating distance or speech errors). In my own work, I have mostly looked at the relationship between implicit evaluations and voting behavior, which is a much more thought-out type of behavior. Indeed, in two elections (2008 U.S. Presidential and 2009 national Parliamentary election in Germany) we found that implicit evaluations were an important predictor of voting behavior both for decided and undecided participants, but explicit evaluations were even more predictive (Friese, Smith, Plischke, Blumke, & Nosek, 2012).

In another direction, we found that people who showed more anti-Black bias on implicit measures of race were more likely to vote for John McCain than for Barack Obama in the 2008 U.S. Presidential election (Greenwald, Smith, Sriram, Bar-Anan, & Nosek, 2009).

Changing Implicit Evaluations

Given that people have less awareness of the content and operation of their implicit evaluations, many of us worry about their impact. In other words, we can say we're unbiased, and we can be 100% honest in believing that, but there are huge portions of our minds that we have very little access to. Thus, I believe that research on changing implicit evaluations is integral to reducing the bias that people unknowingly engage in and consciously experience in their lives.

To this end, I have conducted research showing that the source of a message matters; implicit evaluations are more affected by a source high in expertise and trustworthiness (Smith, De Houwer, & Nosek, 2013). In addition, people prefer an identical political plan more when it is proposed by a member of their own political party (Smith, Ratliff, & Nosek, 2012). Also, aspects of the message are important; people who smoke are more persuaded (on implicit measures) by an anti-smoking message that was emotional in content than one that was focused on facts about smoking (Smith & De Houwer, under review).