Identical messages can have different impacts depending on whether they are couched as “I think” or “I feel,” says Stanford Graduate School of Business Marketing Professor Zakary Tormala.
STANFORD GRADUATE SCHOOL OF BUSINESS —Suppose someone says, “I think it’s the right thing to do.” Or “I feel it’s the right thing to do.” It’s the same thing, right?
Well, not exactly. Ask Zakary L. Tormala, a Stanford Graduate School of Business associate professor of marketing, who has found that substantively identical messages can have a different persuasive impact depending on whether they are couched in terms of their source’s thoughts or feelings.
“Even without changing your actual arguments, you can make subtle framing changes by saying ‘I think’ or ‘I feel’ to make your message more persuasive,” said Tormala, who studied volunteers with Nicole D. Mayer of the University of Illinois at Chicago. “We find that people who are emotionally oriented respond more favorably to messages that begin with ‘I feel,’ whereas cognitively oriented or thinking-oriented individuals respond more favorably to messages that begin with ‘I think,’ even when everything else that follows is exactly the same.”
In one study, after determining whether participants tend to rely more on their emotions or thoughts in making decisions, each person received a message with several arguments in favor of blood donation. These arguments were identical except that they were framed in terms of the source’s thoughts or feelings.
For instance, one message, entitled “My Feelings About Blood Donation,” started with, “I feel that donating blood is one of the most important contributions I can make to society.” It went on to include several more arguments framed in terms of the source’s feelings — for example, “I feel that blood donation is the most fantastic thing I can do with 30 minutes of my free time.”
In a different condition, the message was entitled “My Thoughts About Blood Donation,” and opened with, “I think donating blood is one of the most important contributions I can make to society,” and went on to frame the exact same arguments in terms of the source’s thoughts — “I think blood donation is the most fantastic thing I can do with 30 minutes of my free time.”
Aside from the use of the word “feel” or “think” throughout the message, the content of the arguments was identical, yet those more emotionally oriented were more impressed with (and persuaded by) the “feel” arguments, while those more cognitively oriented liked the “think” arguments better. These are group results, of course; no subject read both the “think” and “feel” arguments.
Tormala and Mayer also found differences between the sexes. “Generally speaking, women tend to self-identify as being more emotionally attuned than do men, and this plays out in persuasion,” said Tormala. “In one study, we found that women were more persuaded by an ad for a new movie when it quoted reviews beginning with ‘I feel.’ Men, however, were more persuaded by the same basic ad when it quoted reviews beginning with ‘I think.’” Tormala warned against interpreting these findings as having anything to do with intelligence or intellect. “Our studies simply show that people have different preferences for persuasive arguments that appear to reflect their source’s thoughts or feelings.”
Ultimately, these “think versus feel framing effects,” as Tormala calls them, could be applied to promote different brands or products more effectively. IBM, for example, has long had a slogan of “Think,” so consumers might have mostly cognitive or rational associations with the brand. On the other hand, Tormala said, Apple computer ads tend to emphasize creativity, trying to make a more emotional connection with consumers. “Although we didn’t test this specific hypothesis, our findings suggest that think and feel messages might be differentially effective in promoting these two brands.”
— Dave Murphy
- “‘Think’ versus ‘Feel’ Framing Effects in Persuasion,” Nicole D. Mayer and Zakary L. Tormala, Personality and Social Psychology Bulletin, forthcoming
Also on Stanford Knowledgebase: