The Elaboration Likelihood Model is the most recent theory that tries to explain attitude change and persuasive communication. It was introduced by Richard E.Petty and John T.Cacioppo during the 1980s. The basic idea of Petty and Cacioppo's theory is that the efficacy of persuasion, in terms of endurance, depends on "the likelihood that an issue or argument will be elaborated upon (thought about)."( Petty & Caccioppo, 1981). When the arguments used in a message are of importance (in terms of involvement and motivation toward the issue) to the message recipient, the expected attitudinal change will be greater than if the message is of little or no relevance to the receiver. If the receiver of the message is interested in the issue and has the ability to process the persuasive message, that person will follow the central route to attitudinal change. On the other hand, if the receiver is not motivated by the arguments of the message and/or does not posses the capacity to process the message, then he or she will follow the peripheral route to attitude change.

-What is an attitude?

     An attitude can be defined as a "general and enduring positive or negative feeling about some person, object or issue" (Bem, 1970; Insko & Schopler, 1972; Oskamp, 1977).Statements like "I love X" or " Y is horrible" can definitely be considered attitudes. This is because they express a feeling toward something--one of them positive, the other negative, but both attitudes.

     Attitudes have to be differentiated from beliefs. Beliefs are ideas, whether factual or only an opinion, that people have about some other people or issues. Another differentiation to be made is between attitudes and behaviors. Behaviors are not beliefs or attitudes, but actual actions.

     Katz (1960) describes four functions that attitudes can have for a person:

     In the last 30 years, many studies have taken place examining methods by which the consumer’s attitudes are created and modified. Attitudes have become a main interest in consumer behavior research. In the different approaches made to the study of attitude change in consumer and social psychology, there is a common link among most of them. This is that these theories use one of two routes to attitude change: the central route or the peripheral route.

     The central route to attitude change depicts attitude change as a thoughtful consideration of information that is crucial to the features of a defined attitudinal position. The second route, the peripheral route, uses positive or negative cues to associate the cues to the attitude issue or object. The person makes a plain inference about the "merits of the advocated position based on various simple cues in the persuasion context"(Petty, Cacioppo & Schumann).

-Central Route

      As explained by Petty and Cacioppo(1981), the attitude change in the central route is the result of a deep examination of the argument and a diligent consideration of the issue-relevant arguments. The central route to persuasion will be used by the message recipient when there is a high elaboration situation and when the recipient's involvement with the topic at hand is important. Therefore, the central route stresses the information that a person posses about the attitude issue under consideration. The output of this process appears to be a very rational one. The message recipient follows a path in which, first of all, pays attention to the message arguments; secondly, attempts to understand them; and finally, makes an evaluation of them .After this process, the message recipient will integrate the information received into a reasoned position, whether favorable or unfavorable, toward the issue. Although, even a very rational process is not always an objective one, since the arguments that are good for one person may be bad for another. Despite this certain lack of objectivity under the central route, persuasion lies on the logical and rational consideration of the objects of study.

-The Peripheral Route

      The peripheral route portrays quite a different explanation of the persuasion process. The picture that this approach portrays is not a very thoughtful one. Attitude changes that occur via this route do not occur after a process of diligent consideration of the pros and cons of the argument, but rather because the attitude is associated with cues presented in the persuasion context. According to this approach, attitude change is affected by such variables as the consequences, rewards, or punishments associated with the message; the judgmental distortions in the perception of the message; or the inferences the recipient draws about why the sender of the message advocated a certain position. So, if a message is associated with an attractive source or a pleasant feeling, it is accepted; while if the message is linked to a discrepant position, it will be rejected regardless of the arguments presented.

      The difference between the central and peripheral routes to persuasion is not based in the fact that the central is rational whereas the peripheral is based in emotion. The apparent rationality of the central route is misleading, since it only has to make sense to the person who uses it. Furthermore, the peripheral approach can somehow also be considered rational, since it is logical to be attracted to things that lead to rewards or to agree with matters that an expert on the matter claims. The main difference between the central and peripheral routes to attitude change is that with the central route, the most relevant determinant predictor of the direction and amount of attitude change is the process of thinking about the relevant issue, while in the peripheral route, persuasion cues indicators. Persuasion cues can be defined as "factors or motives inherent in the persuasion that are sufficient to produce an initial attitude change without any active thinking about the attributes of the issue or the object under consideration"( Petty & Caccioppo, 1981). These cues allow the recipient to adopt an attitude without having to go through a deep analysis of the issue under consideration.

-Anticipatory Attitude Changes

      We can define an anticipatory attitude change as the attitude that happens ( a person is expecting a message but has not yet received it )before the person has received a message but is expecting it. It seems that the attitude of the receiver toward the message becomes more or less extreme depending upon how involved the recipient is with the topic of the message. When the message is not very involving for the recipient , attitudes generally become more moderate; whereas when the person is very involved with the expected message, attitudes become more extreme.

      When the messages that are expected to be received are personally relevant to the recipient , the person is not only more likely to engage in more extreme positions, but the person will also have greater support of his attitude than the recipients of less involving messages.

      When the impending message is on an issue of low involvement, the recipient takes a more moderate position which requires less thinking about the arguments while being easier to defend, and also gives the image of open-mindness. On the other side, people are more concerned with defending their positions when the issue is more relevant to them--more interested than in showing a favorable image of themselves. Defending one’s position obviously requires a certain level of thinking.

      It is also true that the people that use the central route model will experience more permanent changes in attitude than people that process a message using the peripheral route.

     In cases where the involvement is high, subjects’ attitudes will be modified by the nature of the issue-relevant arguments, while the source from which the message emanates will not be of significant importance. On the other hand, when low involvement issues are presented, the quality of the arguments is disregarded and the attitude change will be based on the degree to which the message receiver considers the source likable. Evidence of these statements can be found in a study carried on by S. Chaiken(1980).

     -Theories of Involvement
     Regarding the effects of involvement on persuasion, the dominant notion stems from Sherif’s (Sherif et al. 1965) social judgment theory. This theory explains that highly involved subjects present more negative evaluations of a message because high involvement is associated with and "extended latitude of rejection"(Sherif et al. 1965) . Therefore, highly involved messages are likely to be rejected because it is a greater probability of them falling within the unacceptable range of the subject’s implicit attitude continuum. Another theory that has been widely recognized is the one that Krugman(1965) presents. He states that incrementing the level of involvement does not enhance resistance to persuasion but provokes a shift in the sequence of communication impact. Hence, in a highly involved situation the sequence will be the following: cognition, attitudes and behaviors, whereas in under a situation of low-involvement behaviors would be affected before attitudes are.

     The general trend in social psychology states that the variables that affect persuasion are different depending on the level of involvement the subject has of the issue treated. Under highly involved conditions subjects appear to thoughtfully evaluate the relevant arguments presented in the persuasive message, and their attitudes will be a result of this sentient and argument processing effort. In the case of low involvement conditions, the process explained above does not happen and the attitudes are formed by the reactions of the subject to the cues presented in the message.

     Summing up, the elaboration likelihood model presents two routes to attitude change. The central route results when the message is evaluated thoughtfully by the receiver and hence persuasion comes from carefully thinking about the subject under consideration. The second route, the peripheral route, is taken when persuasion is sought by non-relevant arguments about the topic under consideration. These non-relevant arguments are diverse: the attractiveness of the source, social role, etc. Furthermore, the lasting effects of persuasion are significantly higher when the process of persuasion has come about through a certain amount of thinking (central route) rather than when the persuasion has been accomplished by persuasion emotional cues. In this second case, persuasion will only remain while the cues remain prominent. Hence, as noted in the introduction, "enduring attitude change appears to depend on the likelihood that an issue or argument will be elaborated upon (Thought about)."( Petty & Cacioppo, 1981).

      Every person receives an incredible number of messages daily, and certainly, we do not carefully pay attention to every single one. The great majority of them are not worth our time and will completely be dismissed. Though, for the ones that will be considered one of the two routes will be taken. A diagram adapted from (Petty & Cacioppo, 1981) is presented to show the path that the receiver goes through when receiving a persuasive message under the ELM.

     The ELM can be a very useful tool when used in advertising strategic decisions, but we have to be somewhat flexible when applying it. Despite working in the same framework, rules have to be adapted to the specific features of the advertising product. Despite its clear- cut structure, it is important to understand a key factor that is often overlooked: Central factors influencing persuasive success vary from person to person, and from situation to situation. A degree of persuasive effectiveness held true by one person may not stand strong in the mind of another.

     -Empirical Evidence
     In order to test and prove the empirical applicability of their theory for advertising purposes, Petty and Cacioppo(1986) used mock magazine advertisements as stimuli. The experiment took place among college students who were presented a booklet of six ads, five of them of existing but not very well known products and one of a fictitious product (Villance shampoo).
     -Variables
     Three variables were used to examine the different reactions of the students used as test subjects. Thus , a booklet with the six ads, with a brief description of the purpose of each ad, was handed to the students. The description of the real products read all the same , but for the bogus products the description was varied to create one with low-personal involvement condition and the other with a high involvement. In the low involvement ad, the description stated that the shampoo was being tested in American universities to be evaluated but it would be only marketed in European countries, thus, low-involvement subjects were lead to think that was very unlikely for them having the chance of purchasing the product . On the other hand, the high involvement description presented the product as about to be sold in their community.

     The Villance ad was presented in four different manners, though all four followed a standard format. The ad portrayed a young man and a woman explaining the reasons why they liked the shampoo.

     The variables that were manipulated in the ad were the cogency of the reasons for liking Villance shampoo, as well as the attractiveness of the couple. Hence, the ads designed to create a higher attraction by the audience depicted a couple that had previously rated as very attractive whereas the low attractive featured a couple previously rated somewhat unattractive. In manipulating the cogency of the ads’ argument, statements about the composition of the shampoo, previously rated as very compelling and persuasive, were used for the strong-argument version. Unpersuasive arguments as the color of the shampoo were used for the weak argument versions. The four different ads combined these variables.

     The audience, following the exposure to the complete ad booklet, responded several questions about the ads. The attitudes toward the shampoo were measured on a series of semantic scales representing the average of subjects’ ratings.

     It is not surprising that the ads presenting the attractive couple and the strong arguments lead the subjects to like the product more than for the unattractive couple and the weak arguments. "Of great interest, however, was significant argument-quality-by-involvement interaction. It was revealed that the quality of the arguments had greater impact on attitudes toward the product when the ad was of high relevance rather than low relevance, then using the central route. However, the source-attractiveness-by involvement interaction revealed that the source cue did not make any difference between the low and high relevance ad(peripheral route).

     A variety of possible explanations for not finding evidence for the theory of the peripheral can be found, although the most compelling is that the models were not only viewed as peripheral cues but their actual physical appearance was another argument rather than only a cue.

     Trying to provide a more accurate test of the two routes to persuasion, a second study was conducted similar to the first one but with some relevant modifications. Once again undergraduate university students were asked to examine a booklet of ads that this time contained twelve ads. Among the twelve ads one was for a fictitious product, Adze disposable razors. One of the notorious changes from the first test was that a gift was promised to the students. The students who were going to be exposed to the high-involvement ads were said that they would given a variety of disposable razor products, making the Adze ad high involving. On the other hand the students exposed to low-involvement ads were promised to be able to select their gift among a variety of toothpaste brands.

     Four different versions of the ad for Adze were created. The ad presented the design of the razor, a man and a woman endorsing the product, and six statements defining the product. In the same manner as in the shampoo ads, the strong-arguments version of the ad presented statements previously rated as compelling whereas the weak-arguments version used statements rated as unpersuasive. Regarding the endorsers used, some ads were endorsed by two well-known and liked professional athletes and the headline read "Professional athletes agree: Until you try new Adze disposable razors you’ll never know what a really close shave is." The other ads were endorsed by ordinary citizens and the headline read "Bakersfield, California, agrees: Until you......". In order not to run into the same problem that occurred in the former experiment, being the physical appearance of the endorsers relevant to the product information, in this case it was considered that for the majority of people the status of the endorsers will not be relevant to an evaluation of the qualities of a disposable razor. On the other hand, the status of the endorsers could serve as a strong positive cue. Following the presentation of the twelve ads booklet, the students were asked a series of questions. The attitude toward Adze razors was evaluated by having students respond to a series of semantic differential questions. Later on, these responses were averaged to create an index of attitude.

     The results showed the following patterns. Regarding the quality of the product-relevant information displayed in the ad, it had a much stronger impact on attitudes when an imminent decision about the product had to be made(high involvement) rather than when no decision was imminent (low-involvement). The impact on attitudes depending on the nature of the endorsers was significant under low involvement conditions while no effect was reported under high involvement. Thus, the different kinds of information processing activities postulated in the Elaboration Likelihood Model were demonstrated in the second experiment. However, not only attitudes were measured in the experiment. The likelihood of the subjects purchasing Adze disposable razors the next time they needed a product in this category was also tested. The conclusions refuted the ELM postulates as well. Under high involvement conditions, attitudes toward Adze were proven to be a better predictor of future behaviors than under low involvement conditions. Thus, attitude change in the central route was a better predictor of behavior than attitudes changed via the peripheral route.

     In order to fully understand the ELM theory, a brief review of the critiques it has received is definitively needed.

     Stiff and Boster (Petty et al. 1987) argued that the ELM model presented a bipolarized model of processing messages, either using the central or the peripheral route, but never explained the possibility of channeling the messages through a combination of both routes. Since they believe this middle route exists, they point out that Petty and Cacioppo fail to give an explanation the case in which both routes were used. Stiff and Boster also argue the statement made by Petty and Cacioppo that the same cue can be used at the same time in central or peripheral route. By their understanding, this view allows an undefined number of outcomes, which is not only practically impossible, but also renders the model incapable of making predictions. Hence, Stiff and Boster conclude by describing the ELM as a very limited and incomplete theory to explain persuasion.

     Hamilton, Hunter and Boster's (1993) criticism of the ELM is that, although is similar to the information processing model , it does not reflect discrepancy nor information comparison-- two aspects that the authors believe crucial for a persuasion model. "Emotional conditioning model of the effects of thoughts " is used instead of information comparison. This model assesses that too much counter-arguing will lead the attitude away from the direction advocated by the source, creating the boomerang effect. If this assumption is to be true, the boomerang effect should occur quite often. But empirical evidence demonstrates that it does not appear as often as the model claims, leading to the thought that the empirical evidence to date has not been able to find evidence of emotional conditioning.

     ELM tends to concentrate on the receiver's ability to be persuaded rather than on the message itself. ELM can not determine what makes a case weak or strong apart from the effect on the listener. ELM's strength, though, is the fact that answers some of the communication questions theorists have asked. For example, it explains that few people have the motivation and ability to do the mental work required to change their attitude.

     By bringing ELM to real world situations, it is clear that the central route is really hard to accomplish, although if we are able to channel our targeted recipient, then persuasion can be big and enduring. The elements that the central route requires, motivation and the ability to process information, are very difficult to bring together outside of a experimental environment. First, motivating people to pay attention to what we are going to say is not an easy task. Second, constructing a highly persuasive message without it becoming too complicated to be understood is also not an easy job.

     The peripheral route is easier to handle. It is relatively simple to connect with the audience. presenting them with other things that we already know they find compelling. The problems in this case will be twofold. First, when the person has prior information about the issue to be treated they will not take into consideration the peripheral cues. Second, the peripheral route will not produce a long lasting persuasive effect, since its efficacy will only remain while the persuasive cues remain salient.

     The ELM theory is a good model in which the professionals can base their understanding about the people they want to reach with their persuasive messages. The ELM is based in solid empirical evidences and it is well and logically rationed. The next step in ELM theory should be to find the different arguments that may work better in different product categories, which would turn ELM in a powerful advertising tool.