Dual Attitudes

What it is

Most accounts of attitudes assume one evaluation per object: you like a brand, or you do not. The dual attitudes idea says that is sometimes wrong. A person can carry an explicit attitude, the one they report when someone asks them, and an implicit attitude, a faster association that comes to mind without being asked for. Timothy Wilson, Samuel Lindsey and Tonya Schooler gave the idea its label in a 2000 model published in Psychological Review.

The two attitudes need not agree. A person can say, and mean, that they do not trust a salesperson while the quicker association to that salesperson stays warm. Neither is the fake one. The explicit attitude is what the person can defend. The implicit attitude is what the person does not usually consult, and it may still be what steers the next small choice.

The reason the idea matters for persuasion is that it separates two things the older picture ran together: whether a person has seen through a tactic, and whether the tactic has stopped working. Seeing through a tactic is an explicit judgement. It is the kind of thing persuasion knowledge produces. Whether the tactic keeps working can depend on the other attitude, which the judgement does not reach.

This is why the concept appears next to the research on pitches, flattery and sponsored content. A person who says they are not taken in is reporting an explicit attitude, and the research asks what is left over once that attitude has been corrected.

In effect

In research: insincere flattery

Elaine Chan and Jaideep Sengupta used the dual attitudes idea in a 2010 paper in the Journal of Marketing Research titled "Insincere Flattery Actually Works". When the ulterior motive of a flatterer was obvious, people discounted the compliments, and their explicit evaluation of the flatterer fell. A favourable implicit attitude persisted alongside it. The authors report that the implicit attitude had more influential consequences than the explicit one. This page read the paper through the American Marketing Association's research summary, not the paper, so the measures and sample sizes are not recorded here.

In everyday life: the guest who hates the invitation

Robert Cialdini quotes a customer of home-party sales in Influence (2021, p. 77): "I hate to be invited to Tupperware parties... But when a friend calls up, I feel like I have to go. And when I get there, I feel like I have to buy something." Cialdini's comment is that such customers appear to be fully aware of the liking and friendship pressures. The quotation is an account, not a measurement of two attitudes, but it has the shape the concept describes: a stated attitude that is negative and a pull that is not.

In commerce: the home party

Cialdini describes the home party as a setting where the more compelling requester is not the representative but the hostess, a friend to every person in the room, who "everyone knows, makes a profit from each piece sold at the party." He cites unnamed consumer researchers for the claim that the strength of the social bond is twice as likely to determine purchases as preference for the product itself; the book gives neither names nor numbers, so the claim is reported here as his, not as a finding. On Cialdini's account, the design puts a visible profit motive next to a friendship, and the friendship does the asking.

What it does not say

The idea does not say that implicit attitudes are the true ones and stated attitudes are a cover. It does not say that every persuasive appeal works on the implicit attitude, or that the implicit attitude always carries the consequences. In Chan and Sengupta's work that was reported for flattery from an agent with an obvious motive, and one summary of one paper is a narrow base.

Implicit attitudes are also measured indirectly, and how well such measures predict behaviour has been argued over for years. Nothing on this page settles that argument. The page also does not claim that Cialdini's home-party customer held two measured attitudes; the quotation is used as an illustration of the pattern.

Why the idea matters for persuasion

Most thinking about persuasion assumes one attitude per person, so a pitch either works or fails. Dual attitudes breaks that assumption. A person can state one evaluation when asked and hold another that works quickly and out of sight. Whatever a seller wants, the stated attitude is the one a survey sees and the implicit one may be the one a purchase follows.

The flattery study

Chan and Sengupta (2010) tested it on flattery from a marketing agent. When the ulterior motive was obvious, people discounted the compliments and their explicit evaluation of the flatterer fell. A favourable implicit attitude persisted alongside it. The authors report that the implicit attitude had more influential consequences than the explicit one. They also examined how the gap between the two can be reduced, but the summary read for this page gives no mechanism, and none is stated here.

The home-party guest

Cialdini's Influence quotes a customer who hated being invited to Tupperware parties, felt she had to go when a friend called, and felt she had to buy something once there. His comment is that such customers appear fully aware of the liking and friendship pressures. Fourteen pages on, he describes an experiment on men in North Carolina in which an evaluator who needed a favour was liked best when he offered only praise, even when the men fully realised that he stood to gain. The book gives neither researchers nor sample size for that study, and neither account tests why.

Dual attitudes offers a reading, and only a reading: the guest can say truthfully that she discounts the host's warmth, and still be moved by it, because what she reports and what she feels are two attitudes.

What it does not say

It does not say that stated attitudes are false or that recognising a tactic is useless. It says that a stated judgement is a poor guide to what the tactic did. It is also a model of the mind, and the evidence for this application comes from one study read at the level of a publisher's summary.

Origin and evidence status

The label comes from Wilson, Lindsey and Schooler's 2000 model in Psychological Review. This site records the evidence for the persuasion application as mixed. See Persuasion Knowledge.

---