You must turn off your ad blocker to use Psych Web; however, we are taking pains to keep advertising minimal and unobtrusive (one ad at the top of each page) so interference to your reading should be minimal.

If you need instructions for turning off common ad-blocking programs, click here.

If you already know how to turn off your ad blocker, just hit the refresh icon or F5 after you do it, to see the page.

Psi man mascot

Persuasion and Attitude Change

Advertisers, politicians, and potential romantic partners all aim at persuasion. Persuasion aims to change behavior. It can be accomplished by altered a person's attitude (favorable or unfavorable view of something). That might, in turn, alter the person's disposition to act one way or another.

An attitude (unless it is a deliberate attitude of neutrality) generally involves liking or disliking something. Thumbs up or thumbs down. Attitudes are inclina­tions to favor or disfavor particular beliefs, people, products, or messages.

In North America, the word attitude is used colloquially to indicate a negative disposition. A child who "has an atti­tude" has a disposition to misbehave. This reflects a negative assessment of some situation. The child is against it.

What are attitudes? How do attitudes relate to persuasion?

The words attitude and persuasion are often found together, as in the phrase persuasion and attitude change. Persuasion is an attempt to change people's attitudes and behaviors.

Social psychologists emphasize that an attitude is preparation for behavior. Otherwise, nobody would care about attitudes. An advertiser would not try to induce a positive attitude toward a product unless this was assumed to make you more likely to buy the product.

Hovland's Analysis of Persuasion

The first generation of research on persuasion and attitude change was led by Carl Hovland of Yale University from the late 1940s through the 1950s. Hovland started out his career studying Pavlovian conditioning in the 1930s.

By the late 1940s, Hovland was studying films made by the U.S. Army during WW II. His was asked to find out how effec­tive they were at persuading soldiers.

Hovland and colleagues set out to deter­mine which factors influenced the suc­cess or failure of persuasion. Hovland highlighted three variables.

1. Characteristics of the communi­cator (the person conveying the message) such as whether the person is an expert

2. Characteristics of the communi­cation (what information is con­veyed) such as the arguments used

3. Characteristics of the situation (the circumstances in which the message is conveyed) such as whether the person receiving a message was comfortable

What were the three factors in Hovland's classic analysis?

The first of Hovland's three variables was characteristics of the communicator. Hovland was not the first to study this issue. In 1936, before WW II, Irving Lorge showed that reputation of a source affected people's response to a message (Lorge and Curtiss, 1936).

Lorge presented American students with the statement, "A little rebellion, now and then, is a good thing." If it was attributed to Thomas Jefferson, the students agreed. If it was attributed to Vladimir Lenin, the students disagreed.

Source can have a dominating effect when a message comes from a perceived enemy. Ward and Ross (2002) found that Israeli Jews discounted a peace plan when told it came from the Palestinians, even if the plan was actually proposed by the Israeli government.

Similarly, Israeli Arabs devalued a plan when they were told it came from Israel. In that case, the plan actually came from Palestinian negotiators. The perceived source of the plan had a much greater impact on its acceptance than the content of the plan.

A powerful way to decrease the persuasive power of a source is to attack its credibility. This is called source degradation. It is used in courtrooms when attorneys attack the credibility or qualifications of a witness.

What is source degradation?

Personal characteristics of a communi­cator can also alter the credibility of a message. A variety of social psychology studies showed that a person who was good-looking, neat, sincere, authoritative, or respected was more persuasive, other things being equal.

The Sleeper Effect

The sleeper effect is the tendency of a message from a disreputable source to gain power over time, because people forget the source but remember the message. Hovland, Lumsdaine, and Sheffield (1949) discovered the effect when they studied effects of an army propaganda movie shown to American soldiers.

What is the sleeper effect?

Effects of the movie were weak after five days, but (to the surprise of the research­ers) stronger after nine weeks. Ironically, given the modern view of the sleeper effect, this implies that soldiers had learned to discount army propaganda films as non-credible. When they forgot where the information came from, they gave the information more credence.

Hovland and Weiss (1951) had partici­pants read an article about nuclear sub­marines. One group was told the essay came from Pravda, deemed an untrustworthy source.

A second group was told the informa­tion came from Fortune magazine. That was considered a reliable source in the 1950s but not a place to find expert information about nuclear submarines. A third group was told the information came from a Nobel Prize winning nuclear physicist, a highly credible source.

Source credibility made a big differ­ence at first. However, after a month had passed, the difference in attitudes between the groups disappeared. Now the Pravda group reported as much admiration for the beliefs they had read as the nuclear physicist group. By that time, crucially, people could no longer name the source of the information.

This was called the sleeper effect be­cause it implied propaganda from an unreliable source might be ineffective at first but would do its damage later, like a sleeper cell in a terrorist organization. Eventually–after memory for the source was gone–the information might be persuasive.

Kelman and Hovland (1953) replicated the sleeper effect. But then it became elusive. Pratkanis, Greenwald, Leippe, and Baumgardner (1988) wrote, "Des­pite a long history, the sleeper effect has been notoriously difficult to obtain or to replicate."

In a series of 25 studies, Pratkanis and colleagues (1988) documented and explained the sleeper effect. They found that memory for the "discounting cue" (information that the messenger was unreliable) disappeared more quickly than memory for the message.

Naturally, when the negative source information was forgotten, but the message itself was remembered, the message gained more credibility. This required special conditions, outlined by Pratkanis et al., to make sure the subjects processed the original information as required.

Subjects had to (1) note the important arguments in the message (if they didn't happen to pay attention, no sleeper effect occurred). They had to (2) receive a discounting cue after the message, not before. (That insured that the mes­sage itself was strong, when delivered.)

Finally, subjects had to (3) rate the trustworthiness of the communicator immediately after receiving the dis­counting cue. (That produced the immediate source degradation, reversed over time.)

Under those conditions, but not other­wise, a reliable sleeper effect would occur. This indicates the sleeper effect is not a robust effect. It is the byproduct of special conditions.

In what sense is the sleeper effect "not robust"?

Kumkale and Albarracin (2004) did a meta-analysis of sleeper effect studies in which they verified the finding of Prat­kanis et al. A sleeper effect was most likely to occur when (1) the mes­sage had a strong initial impact, and (2) recipients were motivated to think about it.

They also verified that the sleeper effect occurred only if the discounting cue came after the message. Probably if the cue came first, people did not process the message very deeply.

One attempt at attitude change that might meet these requirements is negative attack advertising during political season. They can have a strong effect on people, and the discounting cue (the message about who sponsored the ad) comes after the rest of ad is over.

Lariscy and Tinkham (1999) found evidence of sleeper effects from attack ads. The information in them gained credibility "substantially" after a few weeks.

How do political attack ads meet requirements for sleeper effects?

Foos, Keeling, and Keeling (2016) took a fresh look at the sleeper effect to see if it might apply to internet advertising. They found the effect in advertisements for packaged foods, such as a gluten-free pizza.

The discounting cue was a mess­age from a fictitious consumer protection group either before or after the pizza ad. The message said benefits of a gluten-free lifestyle were unproven and could cause nutritional difficulties. "Furthermore, customers have complained about lack of flavor of Ciao's pizza."

The researchers found a slight increase in favorability ratings for Ciao's pizza after 11 days, as long as the discounting cue was given right after the ad. As previous researchers had found, putting the discounting cue before the ad eliminated any sleeper effect, and in subjects who saw no discounting cue, opinions were unchanged after 11 days.

How was a sleeper effect found with internet advertising?

The group receiving the discounting cue ended up making higher ratings of the pizza than the other two groups, after 11 days. Some sort of rebound effect might have occurred.

That kind of rebound also happened in the original Hovland research: some­times the less credible source ended up producing more persuasion, after the source was forgotten. Foos, Keeling, and Keeling (2016) concluded that their study replicated the classic sleeper effect in a modern advertising environment.

The Elaboration Likelihood Model (ELM)

In the 1980s, Richard Petty and John Cacioppo developed a theory of persua­sion called the ELM or Elaboration Likelihood Model (Petty and Cacioppo, 1986). They found that people reacted differently to a message depending upon whether they devoted full attention to it.

The ELM suggests that if we do not think about it, irrational "atmospheric" adver­tising can have an effect on us. If we think about it critically, the ad loses its power.

For example, an ad featuring puppies or babies and laughing, attractive people might produce warm, fuzzy feelings. This could raise the ratings for almost any product, even life insurance, but only if a person was not paying much attention, for example, if they were engaged in a distracting task.

If participants were encouraged to scru­tinize the quality of arguments in an ad, they based their responses on whether the arguments were logical and made sense. In that case, atmospherics like beautiful people or uplifting music had little effect.

What was the Petty and Cacioppo ELM model?

This is yet another line of research relating to the two modes of thought discussed in Chapter 3 (States of Consciousness). One summary of the two modes was "the heart" vs "the head" (Epstein (1995).

Ads appealing to "the heart" (pro­cessed rapidly and outside attention) created a positive emotional response. When people evaluated the evidence consciously, engaging "the head" (slower, conscious processing), such positive effects disappeared.

Petty, Cacioppo, and Schumann (1983) found that famous endorsers such as star athletes made advertising more persua­sive only when people were in a low involvement condition. That occurred when they were not very concerned about the product (a fictitious brand of disposable razor).

What experimental manipulation reduced the effect of celebrity endorsers? Why?

When subjects were put into a high involvement condition (expecting to select a brand of razor given to them free) they paid more attention to the ad. Then the celebrity endorsers lost their advantage.

This explains the name Elaboration Likelihood Model. The more likely people are to elaborate on an message (to relate it to other things they know, which means thinking about it) the less likely they are to show an unthinking, automatic positive emotional response to it. The likelihood of elaboration predicts the amount of attitude change.

A successor to ELM was the Persuasion Knowledge Model (PKM) of Friestad and Wright (1994). The PKM said to the extent you know you are being persuaded or propagandized, you are more likely to resist.

People in our society tend to be gullible about advertising as children. We start out naive and unsuspicious. As we grow up, we learn to be increasingly skeptical about advertising and other messages aimed at persuading us. Something like psychological reactance kicks in, and we resist.

Henrie and Miller (2013) verified this. When consumers realized high-pressure sales tactics were being used on them, the tactics lost their effectiveness.

What is the Persuasion Knowledge Model (PKM)?

Campbell and Kirmani (2000) found that the Persuasion Knowledge Model interacted with factors identi­fied in the ELM model. Consumers who detected "ulterior motives" in a persuasion attempt were likely to discount the message...but only if they were not distracted and there­fore had cognitive resources available to pay attention.

What point made in connection with ELM also applied to PKM?

If people did not pay full attention to an advertising message, they did not discount the message. So, in both the ELM theory and the Persuasion Knowledge Model, people have to pay attention, in order to override pressures and persuasions that are irrelevant to the merits of an argument.


Campbell, M. C. & Kirmani, A. (2000). Consumers' use of persuasion knolwedge: The effects of accessibility and cognitive capacity on perceptions of an influence agent. Journal of Consumer Research, 27, 69-83.

Foos, a., Keeling, K., & Keeling, D.I. (2016) Redressing the sleeper effect: evidence for the favorable persuasive impact of discounting information over time in a contemporary advertising context. Journal of Advertising, 45, 19-25.

Friestad, M. & Wright, P. (1994) The Persuasion Knowledge Model: How people cope with persuasion attempts. Journal of Consumer Research, 21, 1-31.

Henrie, K. M. & Miller, D. W. (2013) An examination of mediation: Insights into the role of psychological mediators in the use of persuasion knowledge. In R. Eid (Ed.) Managing Customer Trust, Satisfaction, and Loyalty through Information Communication Technologies, Chapter 7. Hershey, PA: Business Science Reference. doi:10.4018/978-1-4666-3631-6.ch007

Hovland, C. I., Lumsdale, A. A., & Sheffield, F. D. (1949). Experiments on Mass Communication: Studies in Social Psychology in World War II: Volume III. Princeton: Princeton University Press.

Hovland, C. I. & Weiss, W. (1951) The influence of source credibility on communication effectiveness. Public Opinion Quarterly, 15, 635-650. doi:

Kumkale, G. T., & Albarracin, D. (2004). The Sleeper Effect in Persuasion: A Meta-Analytic Review. Psychological Bulletin, 130, 143-172.

Lariscy, R. A. W. & Tinkham, S. F. (1999) The sleeper effect and negative political advertising. Journal of Advertising, 4, 13-30.

Lorge, I. & Curtiss, C. C. (1936) Prestige, suggestion, and attitudes. Journal of Social Psychology, 7, 386-402.

Maoz, I., Ward, A., Katz, M., & Ross, L. (2002) Reactive devaluation of an "Israeli" vs. "Palestinian" peace plan. Journal of Conflict Resolution, 46, 515-546. doi:10.1177/0022002702046004003

Petty, R. E. & Cacioppo, J. T. (1986). The Elaboration Likelihood Model of persuasion. In Berkowitz, L. (Ed.) Advances in Experimental Social Psychology, Vol. 19, New York: Academic Press. (Pp. 123-205).

Write to Dr. Dewey at

Don't see what you need? Psych Web has over 1,000 pages, so it may be elsewhere on the site. Do a site-specific Google search using the box below.