The Role of Humor in the Persuasion of Individuals Varying in Need For Cognitionm

Maggie Geuens, Vlerick Leuven Gent Management School, Ghent University
Patrick De Pelsmacker, Universiteit Antwerpen Management School, Ghent University
ABSTRACT - Empirical research is contradictory concerning the impact humor has on individuals low and high in need for cognition (NFC). An empirical study with 510 Belgian citizens using humorous and non-humorous advertising stimuli shows that humor has a positive impact on the attitudes of both high and low NFC-individuals, but that attitude formation takes place in a different way. For individuals low in NFC a direct effect of humor on attitudes is found, while for individuals high in NFC an indirect influence via biased cognitions can be observed.
[ to cite ]:
Maggie Geuens and Patrick De Pelsmacker (2002) ,"The Role of Humor in the Persuasion of Individuals Varying in Need For Cognitionm", in NA - Advances in Consumer Research Volume 29, eds. Susan M. Broniarczyk and Kent Nakamoto, Valdosta, GA : Association for Consumer Research, Pages: 50-56.

Advances in Consumer Research Volume 29, 2002     Pages 50-56


Maggie Geuens, Vlerick Leuven Gent Management School, Ghent University

Patrick De Pelsmacker, Universiteit Antwerpen Management School, Ghent University


Empirical research is contradictory concerning the impact humor has on individuals low and high in need for cognition (NFC). An empirical study with 510 Belgian citizens using humorous and non-humorous advertising stimuli shows that humor has a positive impact on the attitudes of both high and low NFC-individuals, but that attitude formation takes place in a different way. For individuals low in NFC a direct effect of humor on attitudes is found, while for individuals high in NFC an indirect influence via biased cognitions can be observed.

Since the eighties the use of emotional advertising has risen substantially. In some countries the use of emotional appeals has increased from about 45% to 70% of the ads over the period 1975-1995 (De Pelsmacker and Geuens 1997). Within the category of emotional advertising, humorous appeals have become very popular. For instance, in a study of 33 different countries McCullough (1992) found that, regardless of the country, about 35% of magazine and outdoor advertisements was humorous.

Together with the increasing use of emotions in advertising, research interest in this field boomed. A multitude of studies have compared the relative impact of emotional versus informational advertising appeals on cognitions, attitudes and behavioral intentions (Fabrigar and Petty 1999). The same holds for humorous versus non-humorous appeals in particular (see Weinberger and Gulas 1992 for an elaborate overview). Despite the popularity of humor in advertising, its impact is not unequivocally positive. In their meta-analysis Weinberger and Gulas (1992) indicate that several variables moderate the impact of humorous ads: for example, product category, the newness of the product and the brand, the type of humor, and the relationship between the product or the brand on the one hand, and the humor on the other. Other examples of moderating variables are audience involvement (Zhang and Zinkhan 1994), self-monitoring (Lammers 1991) and afect intensity (Geuens and De Pelsmacker 1999). These results show that humor can lead to more positive responses in some individuals than in other.

The relationship between audience characteristics or individual differences like affect intensity and need for cognition, and responses to advertising has already been studied frequently, and becomes more and more relevant with the increasing use of direct marketing communication techniques (LaBarbera et al. 1998). Indeed, in their search for more communication effectiveness companies try to tailor their communication messages to the individual characteristics of the members of their target groups. More specifically, the impact of individual differences in need for cognition (NFC) on the responses to humorous advertising is not clear yet. Inconsistent results have been encountered as to whether humor can persuade individuals who are high in need for cognition. The proposition in this study is that humor influences the attitudes of both high and low NFC-individuals in a positive way, but that in respondents low in NFC humor has a direct effect, while in individuals high in NFC humor exerts an indirect effect by biasing ad-evoked cognitions. After reviewing the relevant literature, an experimental study designed to investigate the moderating impact of need for cognition in the processing of a humorous message is described, and the results of the analyses are discussed.


Persuasion and need for cognition

Cacioppo and Petty (1982) define Need for Cognition (NFC) as "a tendency to engage in and enjoy thinking" (p.116). Individuals high in NFC are motivated to elaborate on a message and scrutinize messages very carefully, while their low NFC-counterparts usually are not willing to devote a lot of cognitive resources to process a message and are more likely to use heuristic processing. Over the past two decades NFC has been studied in various settings and appeared to account for significant variance in information processing and attitude formation (see Cacioppo et al. 1996 for an elaborate review). As opposed to their low NFC-counterparts, individuals high in NFC have been shown to follow the central rather than the peripheral route to persuasion (Elaboration Likelihood Model, Petty and Cacioppo 1986) implying that they focus more on message-relevant cues and the strength of the arguments used. More specifically, research results indicate that individuals high in NFC (a) focus more on central information cues (Baker and Lutz 2000), (b) process a larger range of information cues and accept wider price ranges if these prices seem justified by relevant cues (Inman et al. 1990, Suri and Monroe 2001), (c) process information more thoroughly than their low NFC counterparts (Haughtvedt et al. 1992, Peltier and Schibrowsky 1994, Mantel and Kardes 1999), (d) collect information on more aspects of a problem and excel in problem-solving tasks (Nair and Ramnarayan 2000), (e) form attitudes that are more resistant and persistent over time (Haughtvedt and Petty 1992, Areni et al. 2000), (f) develop more topic-relevant cognitions (Haughtvedt et al. 1992), and (g) respond more positively to factual appeals, informationally dense ads and ads containing high quality arguments (Cacioppo et al. 1996). Individuals low in NFC are more likely to be impacted by peripheral cues, esecially when they are unfamiliar with the product or brand and attitudes still need to be developed (Hamilton et al. 1993). Research has shown that people low in NFC respond to peripheral cues such as: (a) celebrities (Ul Haque and Bahn 1996), (b) an attractive endorser (Cacioppo et al. 1996), (c) promotional cues (Inman et al. 1990), (d) the number of arguments (Cacioppo et al. 1996), and (e) message framing (Smith and Levin 1996).

NFC and biased information processing

Although the information processing of individuals high in NFC might appear very rational and objective, this is not always the case. They indeed focus on the merits of the information they are exposed to, but their information-processing activity can in some circumstances be biased in different ways. One source of bias may arise from mood (Cacioppo et al. 1996). Several researchers studied the impact of mood on persuasion, and the general conclusion seems to be that a positive mood leads to a greater persuasive impact (McGuire 1985). Petty et al. (1993) review some of the explanations that have been proposed to account for this fact. A first explanation is that mood can influence attitudes directly through a simple association or inference process. A second explanation concentrates on the impact of mood on cognitions. The assumption here is that mood can have an effect on attitudes by biasing the cognitive responses evoked by the communication in a mood-congruent direction. In other words, a positive mood may facilitate retrieval of positive information from memory, induce positive cognitions, lead to a positive interpretation of the message and to less scrutinizing of the message (because one feels safe and sees the world positively). On the other hand, a negative mood may facilitate the retrieval of negative information, lead to more negative cognitive responses, result in a negative interpretation of the message, and in a more careful consideration of the arguments (Isen 1989, Schwarz and Bohner 1996).

According to Petty et al. (1993) this biased-thinking effect is more likely for individuals high in NFC because they are more prone to engage in elaborate processing. Research results seem to support this view. Petty et al. (1993), for example, found that inducing a positive mood led to an equally positive effect on the attitude towards a foster care program for both high and low NFC-individuals. However, positive mood did not evoke more positive cognitions in individuals low in NFC, while it led to an increase in the proportion of positive thoughts of high NFC-respondents. Furthermore, in low NFC-individuals positive mood had a direct effect on attitudes that was not mediated by thoughts, while for their high NFC-counterparts no direct, but an indirect effect of mood on attitudes via cognitions could be observed. Wegener et al. (1994) studied the impact of mood on the persuasive effects of a message emphasizing the benefits of following a recommendation or the costs of not following a recommendation. Individuals in a positive mood were more persuaded by the 'benefits’ message, while the 'cost’ message had a greater impact on individuals in a negative mood. For high NFC-respondents the influence of mood on attitudes was mediated by likelihood judgements, while for low NFC-respondents no mediation effect was encountered. Areni et al. (2000) provide other evidence that a peripheral cue (in this case not mood, but reported group opinions) can influence low NFC-respondents in a peripheral way and high NFC individuals in a central way by biasing topic-relevant thinking. The conclusion of their study was that reported group opinions affect the attitudes of low and high NFC-individuals in a similarway, but the effect is mediated by ratings of topic-relevant arguments for high NFC-individuals.

The preliminary conclusion is that high NFC-individuals are likely to be more affected by mood induced biased thinking effects, although their attitudes do not seem to be influenced in a way different from low NFC-individuals.

NFC and humor

Two studies have investigated the moderating influence of NFC on the response to humorous appeals. Zhang (1996) tested a humorous and a non-humorous print ad containing weak or strong arguments for a fictitious brand. His conclusion is that humor only mattered for low NFC-individuals and the strength of the arguments only for high NFC-respondents. On the other hand, Geuens and De Pelsmacker (1998) tested TV commercials for existing brands and distinguished three different levels of humor (no, moderate and high). They found that humor (and especially a high level of humor) affected feelings, attitudes and purchase intention of both high and low NFC-respondents. No mediation effects were investigated. Since both studies lead to contradictory results, the question remains if humor can also persuade high NFC-individuals.

Research hypotheses

The objective of this study is to find out a) whether humor in advertising leads to more positive responses, and to direct and/or indirect persuasive effects, and b) to what extent they are different for high and low NFC-individuals. Therefore, following hypotheses are advanced:

H1: The use of humor has a positive effect on ad-evoked feelings, ad-related and brand-related cognitions, the attitude towards the ad (Aad), and the attitude towards the brand (Ab), regardless of whether individuals are low or high in NFC.

The positive effect of humorous messages on ad-evoked feelings, Aad and Ab has been found frequently in previous research (Weinberger and Gulas 1992). Also for low NFC-individuals a positive effect of humor on cognitions is hypothesized. The reason for this is that in the literature discussing the effects of humorous appeals it has been frequently stressed that humor attracts significantly more attention (Lammers 1991, Weinberger and Gulas 1992, De Pelsmacker and Geuens 1996) and evokes a significantly higher cognitive response (Chattopadhyay and Basu 1990). Therefore, in the case of humorous appeals, we assume that also low NFC-individuals will generate more cognitions. Furthermore, the findings of Petty et al (1993), Wegener et al. (1994) and Geuens and De Pelsmacker (1998) suggest that the effects of humor do not differ between individuals high and low in NFC.

H2a: There is a direct effect of humor on the attitude towards the brand of low NFC-individuals that is not mediated by cognitive responses.

H2b: The effect of humor on the attitude towards the brand of high NFC-individuals is mediated by cognitive responses.

Since humor can be considered a peripheral cue that evokes positive feelings in respondents (Geuens and De Pelsmacker 1998), we propose that humor has a similar effect as mood (Petty and Cacioppo 1993, Wegener et al. 1994). We expect that humor impacts the attitudes of both high and low NFC-individuals, although the effect on high NFC-respondents is assumed to be mediated by biased cognitive responses. Hypotheses 2a and 2b refine the analysis of Zhang (1996) in that the effect of humor on individuals with varying NFC is related to the cognitive impact of the message.


Advertising stimuli

Eight fictitious advertisements made by a professional designer were used. To minimize possible confounding effects of the product category a humorous and a non-humorous version was created for four different products: paper handkerchiefs (low involvement, informational product), insurances (high involvement, informational product), a snack (low involvement, transformational product), and holidays (high involvement, transformational product). The humor is of the incongruity type (McCullough and Taylor 1993) and consists of a picture of a dog standing in front of its dog kennel. The dog has just found an enormous bone. Holding the bone in its mouth the dog does not seem to be able to figure out a way to enter its kennel. The headlines differed for each product category, but all alluded to size. For the handkerchiefs "A more convenient size, a smaller price" was used, while "Large, tasteful and healthy. For a strong appetite." formed the copy for the snack, "Tailored insured. Call our free number for more information on our extremely advantageous insurances. For a care-free future." for the insurance company, and "More holiday for less money. Tailored holidays." In the non-humorous versions no pictures were used, but the slogans were kept the same.


As part of a larger investigation, 600 respondents were invited to the University of Antwerp (Belgium) to measure their responses to various advertising appeals. 510 valid responses were obtained. Of these 510 respondents 49.5% was male and 50.5% female. 45.7% was aged between 20 and 29, 25.3% between 30 and 39, and 29.1% between 40 and 55. 14.7% only held a primary degree, 32.7% a secondary degree, and 52.6% held a college or university degree. The group was randomly split in four subgroups. Each group saw four ads, of which two were test ads and two were filler ads. The test ads were for different product categories, and the order of the ads was randomized to avoid sequence effects.


Manipulation control.

Two items ("I think this ad is amusing" and "I think this ad is funny") measured on a 7-point Likert type scale (1 = I totally disagree, 7 = I totally agree) were used to find out whether the manipulation was successful. Since the correlation between the two items was high (r= .65, p<.001), the mean of the two items was calculated ('perceived humor’). The manipulation proved to be successful: the respondents perceived the humorous ads as significantly more humorous than the non-humorous ads (t-test, 4.99 versus 2.39, p<.001).

Need for cognition.

Need for cognition was measured by means of the 15-item Dutch scale, tested and validated by Pieters et al. (1987). Principle components analysis with Varimax rotation indicated that one main factor underlies the structure of the NFC scale, explaining 27% of the variance (coefficient alpha = .70). These results resemble the ones obtained by other researchers (Pieters et al. 1987, Nair and Ramnarayan 2000). Subjects were rank ordered and split into thirds. Subjects whose NFC score fell in the middle group were not included in the analyses (cf. Inman et al. 1990, Zhang and Buda 1999, Suri and Monroe 2001). The mean NFC score for low NFC-respondents is significantly lower than for their high counterparts (t-test, 2.90 versus 3.88, p<.001).

Affective responses.

The 17-item Dutch scale of Pieters and De Klerk-Warmerdam (1996) was used since this scale is substantially shorter than the scale of Edell and Burke (1987), itcontains all factors distinguished in other affective responses scales such as in the Standardized Emotional Profile (Holbrook and Batra 1988), and unlike the SEP it also contains items often encountered in advertising research such as "feeling irritated" and "being bored". The 17 items were measured by means of a 7-point Likert type scale. Principal components analysis with Varimax rotation indicated four factors explaining 64% of the variance. The first factor comprises items such as feeling curious, excited, active and playful, and is labeled "upbeat feelings" (coefficient alpha = .81). The second factor "sentimental feelings" comprises items such as feeling sentimental and mollified (correlation = .69, p<.001). The "negative feelings" factor consists of items such as feeling guilty, sad and scared (coefficient alpha = .75), while the fourth factor "critical feelings" is composed of the items feeling bored, uninterested, annoyed and suspicious (coefficient alpha = .78). For each factor the mean of the composing items is calculated.

Cognitive responses.

Respondents were given a few minutes time to write down all the thoughts that crossed their mind when watching the ad. Afterwards, these cognitive responses were classified by the researchers as positive or negative, and ad- or brand-related (cf. Miniard et al. 1990, MacKenzie and Spreng 1992).

Attitude towards the ad (Aad).

Seven items based on previous work (Miniard et al. 1990, Olney et al. 1991, Cho and Stout 1993) were used: pleasant / unpleasant, likable / unlikable, unfavorable / favorable, persuasive / unpersuasive, informative / uninformative, believable / unbelievable, effective / ineffective. All items were measured on a 7-point scale. Since coefficient alpha is comfortably high (.87), a mean of the seven Aad-items is calculated to indicate the Aad score [While there is no generally accepted way to create composite scores, different approaches may lead to different results. As suggested by one of the reviewers, besides averaging the different items to form the composite, also a single factor CFA model is conducted. The weights determined for each item are used as a multiplier for the respective items after which all items are summed into a composite score. The correlations between the firstly and secondly created composite scores are very close to one. As a consequence, no differences in results can be observed due to a difference in how the composite scores were created.].

Attitude towards the brand (Ab).

The following five items were measured on a 7-point scale: favorable / unfavorable, nice / awful, unappealing / appealing, useful / useless, satisfactory / unsatisfactory (Batra 1986, Cho and Stout 1993). Also for this scale coefficient alpha is high (.84). Therefore, the Ab score will be represented by the mean of the five Ab items1.

In between the affective responses, cognitive responses and attitude measurements, respondents filled out personality scales and unrelated other questions. The total procedure took about one hour and a half.


To test the first hypothesis two-way ANOVA’s were carried out with NFC (low, high) and humor (yes, no) as independent variables and affective responses, cognitions and attitudes as dependent variables (see Table 1). No significant main effect of NFC on affective responses, and attitudes could be detected. NFC did not exert any influence on the number of positive ad- and brand-related cognitions either, but significantly impacted the number of negative ad- and brand-related cognitions. As encountered elsewhere, high as compared to low NFC-individuals developed significantly more negative cognitions, indicating that high NFC-individuals may be more skeptical (Cacioppo et al. 1996).

As predicted, the use of humor had a positive impact on all affective responses except on the negative feelings. When exposed to a humorous as compared to a non-humorous ad, respondents experienced significantly more intense upbeat and sentimental feelings, and less intense critical feelings. Furthermore, in line with the hypothesis of less critical and mood-congruent processing when being in a positive mood or experiencing positive affect, the use of humor significantly increased the number of positive ad-related cognitions, while the number of negative ad- and brand-related cognitions decreased significantly (Isen 1989, Schwarz and Bohner 1996). Surprisingly, the number of positive brand-related cognitions decreased as a result of the exposure to humorous ads. The reason for this can be that humor attracted a lot of attention to the ad at the expense of the brand. As a result, more ad- than brand-related cognitions may have been formed reducing both the number of positive and negative brand-related cognitions for the humorous as compared to the non-humorous ads. As hypothesized and in line with previous results, humor had a significantly positive effect on the attitude towards the ad and the attitude towards the brand (Weinberger and Gulas 1992, De Pelsmacker and Geuens 1996).





Since no significant interaction effects between humor and NFC were detected, it can be concluded that all the results described above hold regardless of the level of NFC, providing support for hypothesis one. Indeed, humor seems to have a uniformly positive impact on both low and high NFC-individuals. This contradicts the findings of Zhang (1996) and to a certain extent confirms the findings of Geuens and De Pelsmacker (1998).

To test H2a and H2b, separate path analyses for high and low NFC-individuals were carried out using AMOS, an SPSS-based tool for structural equation modeling. The paths depicted in Figure 1 and Figure 2 were simultaneously estimated.

For low NFC-individuals only a direct effect of perceived humor on the attitude towards the brand can be observed. Humor exerts a significant influence on positive and negative ad-related cognitions, as well as on negative brand-related cognitions, but these cognitions have no significant impact on the attitude towards the brand. On the other hand, the attitudes of high NFC-individuals are influenced directly by positive and negative brand-related cognitions, and indirectly by the effect of perceived humor on negative ad-related cognitions. These results lend support to both hypothesis 2a and 2b, and confirm the findings by Petty and Cacippo (1993) and Wegener et al. (1994).

However, it should be noted that the direction of the impact of cognitions on Ab is counterintuitive. Negative ad- and brand-related cognitions are positively correlated with Ab, while positive brand-related cognitions are negatively correlated with Ab. A possible explanation for this result can perhaps be found in the contrast effect. An affective context cue, such as a happy mood (or as in this case the use of humor), may influence object evaluations in two different ways. First of all, it may lead to an assimilation of object responses toward the affect found in the context (this is the assimilation effect or mood-congruent effect described above). Another possibility is that individuals partial out the effect of the contextual affect. By doing so individuals often over adjust their object evaluations resulting in a contrast effect (Meyers-Levy and Tybout 1997). Whether an assimilation or contrast effect will occur depends on how appropriate the contextual cue is considered to be for evaluating the object, and how much cognitive resources the individual is able and willing to expend to processing the message. With respect to the latter, persons high in NFC have been proved to be more likely to show a contrast effect than low NFC-individuals. However, Meyers-Levy and Tybout (1997) observed that in case there is no semantic relation between the context and the object (when it is easy to realize that the contextual affect is inappropriate for judging the object), both high and low NFC-respondents show a contrast effect. In the present study there clearly is no evident relation between the humor (the dog and the bone) and the product advertised. Therefore, it may be possible that the respondents adjusted their evaluation of the brand to partial out the positive affect and positive thoughts that were evoked by the humor in order to come to their "true" evaluation of the brand.




Understanding individual customers and building good customer relationships have become key to remain competitive. Therefore, studies investigating individual differences in perception, appreciation and persuasion of communicative messages can give valuable insights into how to communicate with different target groups and individual customers. Need for cognition is an individual difference variable that has been shown to moderate the responses to various types of messages and persuasive appeals. However, unlike the results of previous studies, the current results suggest that individuals low and high in NFC respond in a similar way to humorous appeals. Although humorous messages lead to more positive affective and cognitive reactions and attitudes than non-humorous ones, no difference between the attitudes of low and high NFC-individuals could be found. However, the attitudes seem to have been formed in a different way. Attitude formation of low NFC-individuals seems to be based on a simple association or inference process, while attitudes of high NFC-individuals seem to be formed more as a result of biased processing. Indeed, for individuals low in NFC only a direct effect of perceived humor on brand attitude was found, while for high NFC-respondents no direct, but an indirect effect via cognitive responses was encountered. However, reality may be more complex than that. Taking a closer look at the valence of the regression coefficients of the cognitive responses, it may be that over and above the previous effects, for both high andlow NFC-individuals a contrast effect took place. Possibly, respondents considered the humor (the dog holding a bone that was too large to fit in his dog kennel) as an irrelevant affective context cue and therefore corrected their brand attitude to partial out the effects of the humor. Previous research on the impact of humorous appeals already stressed the importance of the link between the humor and the brand for advertising effectiveness (Weinberger and Gulas 1992). If indeed a contrast effect occurred, the present results support the latter view. More research is needed to explore this phenomenon further. Future research could compare the effects of humor related and unrelated to the brand advertised and investigate whether for related humor assimilation effects can be detected and for unrelated humor contrast effects. Related to this, it would be interesting to investigate which types of context cues are considered as appropriate and inappropriate for different individuals, in order to get a better idea when assimilation and contrast effects can be expected.


Areni, C.S., M.E. Ferrell and J.B. Wilcox (2000), "The Persuasive Impact of Reported Group Opinions on Individuals Low vs. High in Need for Cognition: Rationalization vs. Biased Elaboration? Psychology & Marketing, 17 (10), 855-875.

Baker, W.E. and R.J. Lutz (2000), "An Empirical Test of an Updated Relevance-Accessibility Model of Advertising Effectiveness", Journal of Advertising, 29 (1), 1-14.

Batra, R. (1986), "Affective Advertising: Role, Process, and Measurement", in The Role of Affect in Consumer Behavior. Emerging Theories and Applications, ed. Peterson, R.D., W.D., Hoyer, and W.R. Wilson, Massachussetts, Lexington Books, 53-85.

Cacioppo, J.T., and R.E. Petty (1982), "The Need for Cognition", Journal of Personality and Social Psychology, 42 (1), 116-131.

Cacioppo, J.T., R.E. Petty, J.A. Feinstein, W. Blair, and G. Jarvis (1996), "Dispositional Differences in Cognitive Motivation - The Life and Times of Individuals Varying in Need for Cognition", Psychological Bulletin, 119 (2), 197-253.

Chattopadhyay, A., and K. Basu (1990). "Humor in Advertising: The Moderating Role of Prior Brand Evaluation", Journal of Marketing Research, 27, 466-476.

Cho, H.G., and P.A. Stout (1993), "An Extended Perspective on the Role of Emotion in Advertising Processing", Advances in Consumer Research, 20, 692-697.

De Pelsmacker, P. and M. Geuens (1996), "The Communication Effects of Warmth, Eroticism and Humour in Alcohol Advertisements", Journal of Marketing Communications, 2 (4), 247-262.

De Pelsmacker, P. and M. Geuens (1997), "Emotional Appeals and Information Cues in Belgian Magazine Advertisements", International Journal of Advertising, 16 (2), 123-147.

Edell, J.A. and M.C. Burke (1987), "The Power of Feelings in Understanding Advertising Effects", Journal of Consumer Research, 14, 421-433.

Fabrigar, L.R. and R.E. Petty (1999), "The Role of the Affective and Cognitive Bases of Attitudes in Susceptibility to Affectively and Cognitively Based Persuasion", Personality and Social Psychology Bulletin, 25 (3), 363-381.  

Geuens, M. and P. De Pelsmacker (1998), "Need for Cognition and the Moderating Role of the Intensity of Warm and Humorous Advertising Appeals", Asia Pacific Advances in Consumer research, 74-80.

Geuens, M. and P. De Pelsmacker (1999), "Individual Differences and the Communication Effects of Different Types of Emotional Stimuli: "Affect Intensity" Revisited", Psychology and Marketing, 16 (3), 195-209.

Hamilton, M.A., J.E. Hunter, and F.J. Boster (1993), "The Elaboration Likelihood Model as a Theory of Attitude Formation: A Mathematical Analysis", Communication Theory, 3, 50-64.

Haugtvedt, C.P. and R.E. Petty (1992), "Personality and Persuasion - Need for Cognition Moderates the Persistence and Resistance of Attitude Changes", Journal of Personality and Social Psychology, 63 (2), 308-319.

Haughtvedt, C.P., R.E. Petty, and J.T. Cacioppo (1992), "Need For Cognition and Advertising: Understanding the Role of Personality Variables in Consumer Behavior", Journal of Consumer Psychology, 1, 239-260.

Holbrook, M.B. and R. Batra (1988), "Toward a Standardized Emotional Profile (SEP) Useful in Measuring Responses to the Nonverbal Components of Advertising", in Nonverbal Communication in Advertising, ed. Hecker, S., and D. Stewart, Lexington, MA, Lexington Books, 95-110.

Inman, J.J., L. McAlister, and W.D. Hoyer (1990), "Promotion Signal: Proxy for a Price Cut?", Journal of Consumer Research, 17, 74-81.

Isen, A.M. (1989), "Some Ways in Which Affect Influences Cognitive Processes: Implications for Advertising and Consumer Behavior", in Cognitive and Affective Reactions to Advertising, ed. Cafferata P., and A. Tybout, 91-117.

LaBarbera, P.A., P. Weingard and E.A. Yorkston (1998), "Matching the Message to the Mind: Advertising Imagery and Consumer Processing Styles", Journal of Advertising Research, (September/October), 29-43.

Lammers, H.B. (1991), "Moderating Influence of Self-Monitoring and Gender on Responses to Humorous Advertising", Journal of Social Psychology, 131 (1), 57-69.

MacKenzie, S.B., and R.A. Spreng (1992), "How Does Motivation Moderate the Impact of Central and Peripheral Processing on Brand Attitudes and Intentions?", Journal of Consumer Research, 18, 519-529.

Mantel, S.P., and F.R. Kardes (1999), "The Role of Direction of Comparison, Attribute-Based Processing, and Attitude-Based Processing in Consumer Preference", Journal of Consumer Research, 25 (March), 335-352.

McCullough, L.S. (1992), "The Use of Humor in International Print Advertising: a Content Analysis", Working paper, Miami University, Oxford.

McCullough, L.S., and R.K. Taylor (1993), "Humor in American, British, and German Ads", Industrial Marketing Management, 22 (1), 17-28.

McGuire, W.J. (1985). "Attitudes and Attitude Change", in The Handbook of Social Psychology, ed. Lindzey, G. and E. Aronson, New York: Random House, p. 233-346.

Meyers-Levy, J., and A.M. Tybout (1997), "Context Effects at Encoding and Judgment in Consumption Settings: The Role of Cognitive Resources", Journal of Consumer Research, 24, 1-14.

Miniard, P.W., S. Bhatla, and R.L. Rose (1990), "On the Formation and Relationship of Ad and Brand Attitudes: An Experimental and Causal Analysis", Journal of Marketing Research, 27 (3), 290-303.

Nair, K.U., and S. Ramnarayan (2000), "Individual Differences in Need for Cognition and Complex Problem Solving", Journal of Research in Personality, 34, 305-328.

Olney, T.J., M.B. Holbrook and R. Batra, (1991). "Consumer Responses to Advertising - The Effects of Ad Content, Emotions and Atitude Toward the Ad on Viewing Time", Journal of Consumer Research, 17 (4), 440-453.

Peltier, J.W., and J.A. Schibrowsky (1994), "Need for Cognition, Advertisement Viewing Time and Memory for Advertising Stimuli", Advances in Consumer Research, 21, 244-250.

Petty, R.E., and J.T. Cacioppo (1986), "The Elaboration Likelihood Model of Persuasion", Advances in Experimental Social Psychology, 19, 123-205.

Petty, R.E., D.W.Schumann, S.A. Richman and A.J. Strathman, (1993), "Positive Mood and Persuasion: Different Roles for Affect Under High and Low Elaboration Conditions", Journal of Personality and Social Psychology, 64, 5-20.

Pieters, R.G.M., and M. De Klerk-Warmerdam (1996), "Ad-Evoked Feelings: Structure and Impact on a Ad and Recall", Journal of Business Research, 37 (2), 105-114.

Pieters, R.G.M., B. Verplanken, and J.M. Modde (1987), "Neiging tot Nadenken: Samenhang met Beredeneerd Gedrag", Nederlands Tijdschrift voor de Psychologie, 42, 62-70.

Smith, S.M., and I.P. Levin (1996), "Need for Cognition and Choice Framing Effects", Journal of Behavioral Decision Making, 9 (4), 283-290.

Suri, R., and K.B. Monroe (2001), "The Effects of Need for Cognition and Trait Anxiety on Price Acceptability", Psychology & Marketing, 18 (1), 21-42.

Schwarz, N., and G. Bohner (1996), "Feelings and their Motivational Implications: Moods and the Action Sequence", in: The Psychology of Action: Linking Cognition and Motivation to Behavior, ed. Gollwitzer, P.M., and J.A. Bargh, New York, The Guilford Press.

Ul Haque, E., and K.D. Bahn (1996), "Exploring the Role of the Reading Situation and the Cognitive Type of Individual on the Impact of Advertising", Marketing and Research Today, 24 (2), 72-82.

Wegener, D.T., R.E. Petty, and D.J. Klein (1994), "Effects of Mood on High Elaboration Attitude Change: The Mediating Role of Likelihood Judgments", European Journal of Social Psychology, 24, 25-43.

Weinberger, M.G., and C.S. Gulas (1992), "The Impact of Humor in Advertising - A Review", Journal of Advertising, 21 (4), 35-59.

Zhang, Y. (1996), "Responses to Humorous Advertising: the Moderating Effect of Need for Conition", Journal of Advertising, 25 (1), 15-32.

Zhang, Y., and R. Buda (1999), "Moderating Effects of Need for Cognition on Responses to Positively versus Negatively Framed Advertising Messages", Journal of Advertising, 28 (2), 1-15.

Zhang, Y., and G.M. Zinkhan (1994). "Audience Involvement and Persuasion in Humorous Advertising", working paper, University of Houston.