ACR 2010 Special Session Proposal Session Title: Priming Susceptibility: The Role of the Self in Moderating Nonconscious Priming Effects List of Titles and Speakers: (speakers denoted by *) 1. The Role of the Self-Concept in Susceptibility to Priming Influences on Behavior Dirk Smeesters*, Rotterdam School of Management, Erasmus University S. Christian Wheeler, Graduate School of Business, Stanford University 2. Priming Susceptibility: The Moderating Role of Self-Brand Connection on Nonconscious Priming Effects Heather Johnson*, University of Maryland Amna Kirmani, University of Maryland 3. The Effect of Action and Inaction Goal Primes on Consumers’ Information Processing Juliano Laran*, University of Miami 4. Individual Susceptibility to Priming Effects Stacy Wood*, University of South Carolina Cait Poynor, University of Pittsburgh Tanya Chartrand, Duke University Session chair: Heather Johnson, University of Maryland, will serve as session chair and will lead the 15-minute discussion on big picture implications of understanding individual (consumer) susceptibility to nonconscious priming effects. All speakers have agreed to serve if the session proposal is accepted. SESSION PROPOSAL General orientation and objectives: Recently researchers have examined individual difference factors that play a critical role in consumers’ reactions to nonconscious primes. Developing an understanding of when and how individual and situational factors can moderate effects of nonconscious primes is key to advancing our knowledge of when consumers are likely to respond to primes in real-world environments. The papers in this session suggest a set of individual difference variables that moderate or attenuate the effect of nonconscious primes. Issues to be covered: All four papers share a common focus on deepening our theoretical understanding of factors that influence consumers’ susceptibility to marketing-relevant nonconscious primes (cues), and how these primes can influence downstream consumer behaviors. The first presentation by Smeesters and Wheeler discusses the role of the selfconcept in driving behavioral intentions upon encounters with nonconscious cues. Their results provide initial evidence of the capability of self-consciousness and chronic selfconcept to interpret the meaning of nonconsciously perceived stimuli. The second paper by Johnson and Kirmani demonstrates the role of self-brand connection in shielding consumers from the effects of environmental primes. The third paper, by Laran presents evidence that consumers’ goals automatically invoked by peripheral cues can increase (decrease) their susceptibility to marketing messages. The last paper by Wood, Poynor, and Chartrand offers a 6-item Susceptibility-to-Priming index to explain magnitude of priming effects. “Big picture” discussion of these papers will address questions such as, “Do individuals differ in their susceptibility to nonconscious priming effects?” and, “If so, who is most vulnerable to priming influences?” Importantly, discussion will also address how 1 consumers can shield themselves from being influenced outside of their awareness, as well as suggestions for marketing practitioners in developing initiatives to protect their customers from the influence of nonconscious cues related to competitors’ products. Potential audience: Researchers interested in the automatic psychological underpinnings of consumer behavior and the downstream implications of these influences for marketers. Contribution: This research makes substantial theoretical contributions to contemporary research in consumer psychology, grounded in relevant consumption phenomena. These four papers all feature factors related to individuals’ susceptibility to priming effects, shedding light on the role self-related factors play in driving actual behavioral intentions in line with a cue encountered outside the consumer’s awareness. Taken together, these four papers enrich our understanding of the automatic psychological processes underlying behavior caused by peripheral environmental cues. State of completion: Paper 1 provides a new conceptualization of the authors’ recent empirical findings. Data has been collected for the studies described in papers 2 and 3, while paper 4 is a working paper. SHORT ABSTRACTS The Role of the Self-Concept in Susceptibility to Priming Influences on Behavior Dirk Smeesters, Rotterdam School of Management, Erasmus University S. Christian Wheeler, Graduate School of Business, Stanford University Several factors related to self are discussed that determine the susceptibility of the self-concept to priming influences. These factors include the accessibility of the chronic self-concept, private self-consciousness, and the extent to which primed content is confused with one’s actual self-views. This demonstrates the importance of the self-concept in driving behavior. Priming Susceptibility: The Moderating Role of Self-Brand Connection on Nonconscious Priming Effects Heather Johnson, University of Maryland Amna Kirmani, University of Maryland This paper explores an important moderator of nonconscious priming effects: selfbrand connection, a construct capturing personal relevance of brands. Findings from two experiments point to the moderating role of self-brand connection to both target and competing brands on nonconscious priming effects, suggesting that self-brand connection shields one from incidental encounters with environmental cues. The Effect of Action and Inaction Goal Primes on Consumers’ Information Processing Juliano Laran, University of Miami Action and inaction goals were activated by exposing consumers to information associated with these goals. The studies demonstrate that allowing consumers to pursue an action or inaction processing goal makes them more susceptible to the marketer’s influence and more likely to perform decision-congruent behaviors. Individual Susceptibility to Priming Effects Stacy Wood, University of South Carolina Cait Poynor, University of Pittsburgh Tanya Chartrand, Duke University We suggest that individuals with high cognitive capacities in attention and associative thinking may be more strongly influenced by primes. We demonstrate the efficacy of our proposed 6-item Susceptibility-to-Priming index in explaining magnitude of 2 priming effects, thus contributing theoretical insight into the priming mechanism, policyrelevant insight regarding the use of priming stimuli, and an important empirical tool. LONG ABSTRACTS The Role of the Self-Concept in Susceptibility to Priming Influences on Behavior Dirk Smeesters, Rotterdam School of Management, Erasmus University S. Christian Wheeler, Graduate School of Business, Stanford University Considerable research has demonstrated that behavior is influenced by construct accessibility, and there is rapidly growing evidence suggesting that susceptibility of the self-concept to influence can determine the susceptibility of individuals to primed constructs. In this presentation, we present several studies showing effects self-related processes can have in determining whether primed constructs will affect behavior. A first factor that determines susceptibility to priming influences is the accessibility of the chronic self-concept. In the context of economic decision-making, two experiments tested whether participants with more accessible self-concept representations would be less likely to be influenced by primed constructs, but more likely to act in line with their chronic self-characteristics (Smeesters et al. 2009). Results of the first experiment showed that those who had highly accessible social value orientations acted in line with those selfrepresentations and not with the primed constructs. Those who had less accessible social value orientations did not act in line with those self-representations, but rather acted consistently with the primed concepts. In a second experiment, the accessibility of the selfconcept was manipulated. The primes affected participants whose self-concept was not made accessible by self-related words. However, when the accessibility of self-concept was increased, participants were no longer susceptible to the primed constructs. A second factor that can determine susceptibility to priming influences is private self-consciousness. Past research on this dimension has found opposite effects: some demonstrated that private self-consciousness increased priming effects (Hull et al. 2002), whereas other showed that it decreased priming effects (Dijksterhuis and van Knippenberg 2000). Self-consciousness is a multifaceted construct containing two important dimensions: internal state awareness and self-reflectiveness (Wheeler, Morrison, DeMarree, and Petty 2008). Participants who score high on the internal state awareness factor of the private self-consciousness scale should be less susceptible to priming influences. Self-reflectiveness, on the other hand, might increase susceptibility to priming influences, as those who engage in self-reflection tend to process external feedback in terms of its implications for the self, thereby leading it to have greater effects on self-views and behavior (Wheeler et al. 2008). These predictions were tested in a study in which participants’ scores on both internal state awareness and self-reflectiveness were measured (Wheeler et al. 2008). Results revealed independent interactions between primes and both internal state awareness and self-reflectiveness. A third factor that can determine susceptibility to priming effects concerns the extent to which the context facilitates confusion of the primed content with one’s actual self-views. When the context signals that a prime is “not me”, it can diminish the prime’s effect on behavior. In one experiment (Wheeler, Sleeth-Keppler, and Morrison 2010), nonAsian participants were subliminally primed with intelligence or not via a lexical decision task. Results indicated that the intelligence primes had an effect only when not preceded by Chinese characters, presumably because the Chinese characters signaled to non-Asian participants that the intelligence primes did not apply to them. In a follow-up study, 3 participants who identified or disidentified with math were subliminally primed with the elderly stereotype (or not). Either math symbols or a row of XXX served as forward and back masks to the priming stimuli. These results are consistent with the idea that the math symbols served as a signal as to whether the prime was self-relevant or not. These studies contribute to a growing body of research showing the important role that the self-concept can play in prime-to-behavior effects and emphasize the importance of the self-concept in driving behavior. Priming Susceptibility: The Moderating Role of Self-Brand Connection on Nonconscious Priming Effects Heather M. Johnson (Robert H. Smith School of Business), University of Maryland Amna Kirmani (Robert H. Smith School of Business), University of Maryland This paper proposes that self-brand connection serves an important moderator of priming effects. Self-brand connection reflects the extent to which a brand fits with one’s self-concept (Escalas and Bettman 2005). Counter to prior research demonstrating main effects of nonconscious cue exposure, we predict an inverted U-shaped effect: consumers with a moderate level of self-brand connection to a target brand are more susceptible to a nonconscious prime’s influence on their subsequent purchase intentions toward the target brand. For these individuals, a nonconscious prime renders related constructs accessible and this fluency effect will increase purchase intentions toward the product, consistent with main effects findings in the nonconscious priming literature (Berger and Fitzsimons 2008). However, consumers’ high and low self-brand connection to a target brand shields them from influence of nonconscious cues. In addition, we predict that consumers’ self-brand connection to a competing brand also influences behavioral intentions toward a prime-congruent (target) brand after encountering a nonconscious prime related to that target brand. For example, if John has high self-brand connection to Nike he may be resistant to the influence of a Puma-related incidental cue on his propensity to purchase Puma. We show that this prediction holds, and interestingly that a nonconscious prime may even prompt individuals high in self-brand connection to a competing brand to express decreased purchase intentions toward the target brand. Two experiments demonstrate the moderating effect. In the first study, we measured self-brand connection for both target (Puma) and competing brands prior to participants’ encounter with a dog prime, and show that individuals with low or high selfbrand connection are less susceptible to the prime. Study 2 employs a novel priming paradigm to provide generalizability to a different product domain, laptop computers. Participants again provided self-brand connection ratings for target (Apple) and competing brands in the category (e.g., Dell) and were exposed to an incidental cue related to the target brand (the flavor orange). We again show that by increasing the accessibility of the target brand for consumers moderate in self-brand connection, the prime can lead to increased purchase intentions, while consumers low or high in self-brand connection for the target or competing brands in the category do not express increased purchase intentions for the target brand. Interestingly—and importantly for marketing practitioners—we find that individuals with high self-brand connection for Dell express decreased purchase intentions for Apple laptops upon encounters with the Apple-related prime. These empirical findings extend recent main effect studies demonstrating a 4 nonconscious prime’s influence on purchase intentions by showing that consumers’ susceptibility to these influences is based on self-brand connection. The Effect of Action and Inaction Goal Primes on Consumers’ Information Processing Juliano Laran, University of Miami Action and inaction primes have been shown to influence motor output, such as doodling on a piece of paper and eating, and cognitive output, such as recall and problem solving (Albarracin et al. 2008). These primes can influence decision satisfaction, choice satisfaction, and ultimately the likelihood that consumers will be willing to perform behavior congruent with a marketer’s intent (i.e., buy a given product). In other words, action and inaction primes make consumers more susceptible to performing any behavior intended by a marketer if the goal can be satisfied. I predict that priming of an action goal will make consumers willing to evaluate an increased amount of product information and be highly satisfied with the decision process when choice sets offer a lot of attribute information. Alternatively, priming of an inaction goal will make consumers willing to evaluate fewer pieces of product information and be highly satisfied with the decision process when choice sets offer only few pieces of attribute information. Satisfaction with the decision process may influence choice satisfaction and the likelihood to buy the chosen brand. All studies were composed of three phases. In phase 1, participants were told that they would participate in a sentence recall test. Participants were exposed to ten sentences, which could be action-oriented (e.g., “you can do it”), inaction-oriented (e.g., “you deserve to relax”), or neutral, depending on the goal priming condition. In phase 2, an ostensibly unrelated study, participants were asked to make choice between two brands of digital cameras. In the low load condition, I presented information about 4 attributes, whereas in the high load condition, I presented information about 9 product attributes. In phase 3, participants were asked a series of questions, including satisfaction with the process of deciding which brand to choose and satisfaction with the chosen option. In study 1, participants in the action goal prime condition showed higher decision satisfaction when making a choice from a choice set offering high information load than when making a choice from a choice set offering low information load. Participants in the inaction goal prime condition showed higher decision satisfaction when making a choice from a choice set offering low information load than when making a choice from a choice set offering high information load. There was no influence of information load in the neutral prime condition. Choice satisfaction was found to be fully mediated by decision satisfaction. In a similar study, decision satisfaction also led to an increased likelihood that people would actually purchase the brand. Study 2 examined whether these results would extend to actual behavior. Participants primed with an action goal were willing to donate more money to their favorite charity when a lot of information was presented. Participants primed with an inaction goal were willing to donate more money to their favorite charity when little information was presented. These results suggest that allowing people to process an amount of information congruent with a primed action or inaction processing goal makes them more susceptible to an external agent’s influence outside of their awareness. Study 3 manipulated goal achievement by having participants perform an action (i.e., solve GRE problems) vs. an inaction (i.e., close eyes and relax) task after being exposed to the goal primes. In the action (inaction) goal prime condition, after having performed an 5 action (inaction) task, participants were more satisfied with the decision process and showed higher likelihood of purchasing the chosen brand when the choice set offered low (high) information load. Individual Susceptibility to Priming Effects Stacy Wood, University of South Carolina Cait Poynor, University of Pittsburgh Tanya Chartrand, Duke University The purpose of this research is to test the susceptibility concept and identify the characteristics that would contribute to individual differences in susceptibility to priming (STP). While priming researchers would benefit from the creation of an STP scale because it would serve as a useful covariate, consumer behaviorists are likely to be interested in the concept of STP because of its implications for advertising efficacy. We posit that, counter to intuition, people who are both attentive to their environment and have a tendency to strong associational processing are those who are most susceptible to priming. Indeed, the mechanisms by which priming work suggest that primes require 1) physical (but not necessarily aware) exposure and 2) associative processing, suggesting two paths to priming susceptibility. We build on this conceptual development to generate a simple 6item reflective STP measure. Three studies demonstrate that our STP index does predict the magnitude of an individual’s response to a prime. Study 1 demonstrates that priming individuals with concepts of the elderly (akin to the “walking to the elevator” study by Bargh, Chen, and Burrows [1996]) can influence susceptible consumers to “move slowly” in consumer domain, specifically in adversely affecting their attitude toward the speed of high tech product innovations. In study 1, we expose participants to a modified version of an “elderly prime and test the impact of this prime on responses to a survey about attitudes toward the speed of technological change; specifically, we predict that an elderly prime will serve to decrease reported attitudes toward speedy innovation. We anticipate that the STP index will moderate the effect of this prime, such that priming effects are strongest for those with higher STP. Participants engaged in two ostensibly unrelated tasks that were, together, the focal task. In the first part of this task, participants filled out a survey that asked them to look at a list of brand names for products (e.g., Dasani water, Kleenex tissues) and to identify which brands they thought were American brands and which brands were Canadian brands. In the elderly prime condition, six fictional brand names were words associated with the elderly (e.g., Bingo cookies, Slow-Pour syrup). In the neutral condition, these words were replaced with unrelated words (e.g., Bell cookies, Sweet-Pour syrup). Participants then moved on to another separate survey about their attitudes about high tech products intended to tap into participants’ attitude toward the pace of technological change. We first created an STP index for each participant by averaging across the 6 STP items (α = 0.76, M = 5.57, SD = .74). We then analyzed the data using the STP index and the prime condition of the participant as predictors of attitudes towards technological change. The main effect of the prime (F(1, 151) = .14, p = .71) was not significant. However, there was a significant overall interaction of priming condition and STP scale score (F(1, 151) = 4.08, p = .04). We then probed the nature of this interaction using spotlight analysis as suggested by Aiken and West (1991). This analysis indicates that the prime was ineffective for low and moderate STP individuals, who did not rate technology differently than did their unprimed counterparts (moderate: Mprime = 4.92, Mno prime = 4.85, F(1,151) = .14, p = .71; 6 low: Mprime = 4.77, Mno prime = 5.09, F(1,151) = 1.41, p = .49). However, among high STP individuals, the prime did have an effect. Higher STP individuals exposed to the elderly prime were significantly more likely to describe the pace of technological change as too fast (M = 5.08) than those not exposed to the prime (M = 4.61, F(1,151) = 3.46, p = .05). Studies 2 and 3 further support our hypotheses. Study 2 demonstrates that a subtle work-oriented prime can alter prime-susceptible college students to report increased intentions to study. Study 3 demonstrates that priming the concept of promotion versus prevention (Aaker and Lee 2006) within an advertisement can increase susceptible consumers’ valuation of a promotion-congruent product attribute. These studies provide insight both into susceptibility to primes and to the nature of the prime itself: In some cases, the use of the index provides more nuanced interpretation of an apparently robust prime, showing that its overall effect is actually localized to only a subset of the population. In other cases, the use of the index allows a weak or marginally-significant aggregate effect of a prime to be understood as quite strong if only among a section of the sample. References: Albarracín, Dolores, Ian M. Handley, Kenji Noguchi, Kathleen C. McCulloch, Hong Li, Joshua Leeper, Rick D. Brown, Allison Earl, and William P. Hart (2008), “Increasing and Decreasing Motor and Cognitive Output: A Model of General Action and Inaction Goals,” Journal of Personality and Social Psychology, 95, 510-23. Berger, Jonah, and Grainne Fitzsimons (2008), "Dogs on the Street, Pumas on Your Feet: How Cues in the Environment Influence Product Evaluation and Choice," Journal of Marketing Research, 45 (1), 1-14. Chartrand, Tanya L., Joel Huber, Baba Shiv, and Robin J. Tanner (2008), "Nonconscious Goals and Consumer Choice," Journal of Consumer Research, 35 (August), 189-201. Escalas, Jennifer Edson, and James R. Bettman (2005), "SelfâConstrual, Reference Groups, and Brand Meaning," Journal of Consumer Research, 32 (3), 378-389 Hull, Jay G., Slone, Laurie B., Karen B. Meteyer, and Amanda R. Matthews (2002), “The Nonconsciousness of Self-Consciousness,” Journal of Personality and Social Psychology, 83, 406–424. Smeesters, Dirk, Vincent Y. Yzerbyt, Olivier Corneille, and Luk Warlop (2009), “When Do Primes Prime? The Moderating Role of the Self-Concept in Individuals’ Susceptibility to Priming Effects on Social Behavior,” Journal of Experimental Social Psychology, 45, 211– 216. Smeesters, Dirk, S. Christian Wheeler, and Aaron C. Kay (2010), “Indirect Prime-toBehavior Effects: The Role of the Self, Others, and Situations in Connecting Primed Constructs to Social Behavior,” Advances in Experimental Social Psychology, 42, 259-317. Wheeler, S. Christian and Kenneth G. DeMarree (2009), “Multiple Mechanisms of Prime-toBehavior Effects,” Social and Personality Psychology Compass, 3, 566–581. Wheeler, S. Christian, Kimberly R. Morrison, Kenneth G. DeMarree, and Richard E. Petty (2008). Does Self-consciousness Increase or Decrease Behavioral Priming Effects? It Depends,” Journal of Experimental Social Psychology, 44, 882–889. 7