Will you still trust me when you're angry?

When I was a kid, like most kids, I used to wait for my mom to be in a good mood before I asked her for something I wanted. I thought I was being pretty sophisticated—I understood that her decisions might be different if she was in a bad mood. However, I might not have been as sophisticated as I thought. Jennifer Dunn and Maurice Schweitzer of the University of Pennsylvania have found that emotion is a more important influence on trust than mood ("Feeling and Believing: The Influence of Emotion on Trust," Journal of Personality and Social Psychology, 2005).

Mood is a simpler concept than emotion. Generally moods last longer but are less directed than emotion. For example, the emotions anger and sadness are both associated with a bad mood, but anger tends to be directed at others, while sadness tends to be directed internally. If I had to choose, would I rather confront my mother when she was sad, or angry? Similarly, happiness and gratefulness are both positive emotions. Which is more likely to elicit trust? Is there anything my mom could have done to mitigate this crass manipulation by her son?

Dunn and Schweitzer stopped people in a train station and asked them to fill out a survey in exchange for a candy bar. The survey was actually designed to induce a particular emotion in the respondents and then evaluate how that emotion affected trust. The method used to induce emotion was simple but effective. To induce sadness, for example, they asked the respondents to write down the three things that make them most sad. Then they instructed them to describe the one situation that made them the saddest, in such a way that whoever read the description would also become sad. Though a few couldn't complete the survey because they had a train to catch, most respondents wrote extended narratives about very sad things like when someone in their family died.

Next respondents answered a second survey to determine how much they trusted a coworker selected randomly. They rated, on a scale of 1 to 7, statements about things like how likely this person would be to show up for a meeting, do a favor, or return borrowed money. People who were "happy" were significantly more trusting than those who were "sad." Sad people, in turn, were more trusting than "angry" people. Dunn and Schweitzer performed a separate test which indicated that the emotion-inducing test actually induced the intended emotions.

In a separate study, Dunn and Schweitzer tested college students. They used a similar method to generate emotions of anger or gratitude, or gave a neutral task. Then the students answered the trust questionnaire for either someone on campus they knew very well or were merely acquainted with. Here are the results:

Emotions don't appear to affect trust when the person in question is a close friend, but play a strong role when the person is only an acquaintance. So not only was I apparently wrong to consider only my mom's mood when I was a child, but because my mom knew me very well, she probably didn't change her level of trust in me even when she was feeling sad, angry, or grateful.

Dunn and Schweitzer did one further study—they found that when people were made aware that their emotional state affected trust, they could compensate accordingly. They suggest that this finding could be important in business. For example, an executive who's angry after a difficult meeting should recognize that this emotion could affect how she treats her coworkers. Or, alternately, a salesman who's happy because he's just hit a great golf shot should realize that he might be overly trusting and ink a bad deal with the client he's entertaining.