Putting the Freeze on Priming

Abstract

Past research has indicated that individuals with a high need for cognitive closure (NFCC) are more susceptible to priming effects in norm-absent contexts. We proposed that in norm-present contexts, whereby normative information competes with priming in affecting individuals’ understanding of the social environment, the opposite pattern would occur. In Study 1, low- rather than high-NFCC individuals showed greater prime-consistent behavior in a context with a strong norm to comply. In Study 2, when both priming and normative information were manipulated, priming dictated low-NFCC individuals’ behaviors, whereas norms guided high-NFCC individuals’ behavior. In Study 3, the effect of a single priming manipulation was observed in two consecutive contexts. While high-NFCC individuals, compared with low-NFCC ones, were less prime-consistent in the norm-present context, they were more influenced by the same priming manipulation in the norm-absent context. Our findings underscore the importance of NFCC in people’s selection of environmental cues to guide self-regulation.

Keywords

social cognition need for closure norm priming

The large body of research on social priming (see Loersch & Payne, 2011; Wheeler, DeMarree, & Petty, 2007, for reviews) demonstrates that people’s thoughts and behavior in a particular situation can be influenced by the social stimuli they have previously encountered in an unrelated context. Given that social priming represents a prevalent way in which the social environment can affect one’s self-regulation, understanding what factors determine an individual’s susceptibility to priming is of both theoretical and practical importance. Past research has demonstrated that several individual differences, such as self-consciousness (Hull, Slone, Meteyer, & Matthews, 2002), self-monitoring (DeMarree, Wheeler, & Petty, 2005), and need for cognition (Petty, DeMarree, Briñol, Horcajo, & Strathman, 2008) can moderate priming effects.

Following this tradition in identifying important moderators for priming effects, in this article we aimed to explore how individual differences in epistemic motivation moderate priming effects. Because this individual difference dictates the way people select among various cues to understand the social environment, investigating it can generate new insights on how priming interacts with other potential influences in directing individuals’ thoughts and behaviors. Specifically, past research (Ford & Kruglanski, 1995) has demonstrated that individuals who are motivated to achieve cognitive closure (Kruglanski & Webster, 1996) are especially susceptible to the influence of priming. On the other hand, individuals who are motivated to avoid closure tend to be quite impervious to priming. Here, we propose that the reverse would happen in social contexts that have a salient norm, which, despite the temporal primacy of primed constructs, serves as the starting point of epistemic processes.

Priming and Epistemic Motivation

Priming refers to the subtle, implicit activation of mental constructs or representations through situational stimuli, such that individuals are usually oblivious to its influence from the environment. These pre-activated or “primed” mental constructs then influence people’s thoughts and behavior in a subsequent social context. Inspired by initial priming research in the domain of impression formation (e.g., Higgins, Rholes, & Jones, 1977; Srull & Wyer, 1979), social psychologists have amassed an impressive body of evidence that priming mental constructs can lead to prime-consistent thoughts and behavior beyond individuals’ awareness and intention (e.g., Bargh, Chen, & Burrows, 1996; Dijksterhuis & van Knippenberg, 1998). For example, Bargh and his colleagues have shown that individuals who were presented with words related to politeness (e.g., respect, honor) were less likely to interrupt a conversation than those who were presented with words related to rudeness (e.g., bold, aggressively) in an unrelated earlier context. In sum, priming increases the mental accessibility of the social constructs of interest, which in turn guide individuals’ thoughts and behavior.

Nonetheless, not everyone is equally susceptible to the influence of priming, and the way we strive to acquire knowledge about the social environment determines the readiness with which we use highly accessible primed constructs. According to the theory of lay epistemics (Kruglanski & Webster, 1996), knowledge about the environment is acquired via a two-stage process of hypothesis generation and validation, and a stable individual difference among people, namely, the need for cognitive closure (NFCC), determines the initiation and termination of the two stages. NFCC pertains to the degree to which one desires to arrive at a definitive answer to a question as opposed to having a tolerance for uncertainty or ambiguity. Individuals with a high NFCC abhor uncertainty and ambiguity, and thus are motivated to seize on any salient internal or external cues to generate a hypothesis. High-NFCC individuals then freeze on the initial hypothesis and execute the subsequent validation stage of knowledge acquisition quickly, with a strong reluctance to revise their initial hypothesis even in the face of contradictory evidence. By contrast, individuals with a low NFCC avoid arriving at a definitive conclusion too quickly. They are slow in generating a strong hypothesis and actively seek out additional cues in the environment during the validation stage. This individual difference in the way people acquire knowledge about the social environment has been shown to have far-reaching implications for diverse areas such impression formation, attribution, stereotyping, and group decision making (see Kruglanski & Webster, 1996, for a review).

In light of the “seizing and freezing” pattern of knowledge acquisition among high-NFCC individuals, it stands to reason that they, as opposed to low-NFCC individuals, would be more influenced by primed constructs, which achieve a high mental accessibility prior to the context in which one’s behavior is observed. Because the primed constructs enjoy such temporal primacy, Ford and Kruglanski (1995) argued that they act as preexisting knowledge structures that high-NFCC individuals could rely on in an ambiguous context. As high-NFCC people strive for a definitive understanding of the situation, they would adhere to the highly accessible primed construct and use it as the guide for ensuing psychological processes. On the other hand, low-NFCC individuals are motivated to avoid cognitive closure and are inclined to seek alternative cues to the primed construct. As a result, Ford and Kruglanski surmised that their thoughts and behavior would be influenced by a host of other potential cues (e.g., mood, implicit theories, past associations), such that they would be less influenced by priming. The authors used the Donald paradigm (Higgins et al., 1977) to test their hypothesis. After being primed first with words related to either adventurousness or recklessness, participants read risk-taking behaviors performed by “Donald” that could be characterized in either a positive or negative way. Indeed, participants high in NFCC were more strongly biased by the prime in their perceptions of Donald than were low-NFCC individuals.

Priming in Norm-Present Context

In this article, we ask whether the temporal primacy of primed constructs would mean that high-NFCC individuals are always more susceptible to priming than are low-NFCC individuals. There are two main premises underlying such a notion: (a) primed constructs enter the relevant psychological process that leads to cognitive and behavioral output temporally earlier than other cues, and (b) the “seizing and freezing” of high-NFCC individuals always leads them to process and ultimately rely on the earliest cue in the environment. A close scrutiny of the relevant literature suggests reasons to question both of these premises. First, recent theorizing about social priming suggests that individuals may not be influenced by the primed constructs rigidly from the moment it is activated. Instead, individuals flexibly use the primed constructs in their inquiry regarding the social environment (Loersch & Payne, 2011; Wheeler et al., 2007). The situated inference model (Loersch & Payne, 2011), for example, posits a three-stage process of priming. In the first stage, exposure to the prime makes the relevant mental content highly accessible in memory. As people are oblivious to the actual source of the prime, the accessible content is then assumed to be caused by, or misattributed to, the focal target of their attention. Finally, the misattributed construct biases people’s response to the question afforded by the focal target. According to this model, the primed construct can be used in different ways, depending on the type of focal target people pay attention to. For example, following a “politeness” priming (Bargh et al., 1996), if the focal target of the individual is another individual, the highly accessible content will be likely to be applied to the trait or behavior of that target individual (“This is a polite person”). If, however, the focal target of the individual is his or her own behavior or goals, the mental representation will likely to be misattributed to his or her behavior or goals (“I want to be a polite person”). In relation to the current discussion, this suggests that the effect of priming does not happen at the point of its mental activation but applies when the individual’s attention is directed toward a particular inquiry about the social environment. Hence, any cues that are closely associated with the context of interest may be considered and processed at the same time as the primed construct, such that the primed construct would not necessarily take precedence in guiding individuals’ self-regulation of behavior.

Second, although Kruglanski and Webster (1996) emphasized early-cue processing and primacy effects as hallmarks of the “seizing and freezing” principle of high-NFCC individuals, they also acknowledge that “it is quite possible that if the late evidence is particularly compelling . . . participants high in need for closure are pressured to seriously consider it” (p. 270). We would go a step further in arguing that not only would individuals high in NFCC be likely to consider compelling late evidence, but would suggest that high-NFCC individuals would be more likely to seize on the most potent situational cue in the environment, to the neglect of temporally precedent priming, than their low-NFCC counterparts. In this article, we specifically propose that situational norms, real or imagined, can have the potential to override any temporal primacy of priming.

As social animals, we are deeply concerned with the behavior of other humans in the same environment. Others’ behavior either suggests which actions lead to social approval (Cialdini, Kallgren, & Reno, 1991; Fishbein & Ajzen, 1975) or affirms our social reality and the effectiveness of our own behavior (Cialdini et al., 1991). Substantial empirical evidence attests to the potency of normative information in predicting behavior (e.g., Fishbein & Ajzen, 1975; Schwartz & Tessler, 1972). Research has also shown that individuals’ subjective norms predicted their behavior over and above important social factors such as attitudes (for reviews, see Ajzen, 1991; Armitage & Conner, 2001). Indeed, cross-cultural research has shown that high-NFCC were more likely than low-NFCC individuals in utilizing cultural norms in guiding their attributional styles (Chiu, Morris, Hong, & Menon, 2000) as well as conflict judgments (Fu et al., 2007).

Given that norms are another important situational cue that high-NFCC individuals readily achieve cognitive closure on, we propose that when both primes and norms are present, norms might override and replace the primed constructs as the seminal cue for seizing and freezing to occur. In other words, the stronger potency of primed constructs among high-NFCC individuals would not be observed in situations with a salient norm which would replace the primed constructs as the most salient cue for understanding the social environment. In an attempt to end uncertainty and ambiguity, individuals with high (vs. low) NFCC would rely on the salient situational norm as a ready guide for behavior. Subsequently, they should become reluctant to engage in additional information processing, including utilizing the highly accessible constructs of priming. This dismissal of primed constructs should lead to a decreased influence of priming among high-NFCC individuals. Low-NFCC people, on the other hand, are motivated to avoid early closure with the situational norm. As they actively look for additional information to the salient norm, they are likely to access the primed construct and thus be influenced by it. In other words, we predict that in situations with a salient norm, the opposite pattern to the classic finding (Ford & Kruglanski, 1995) would be observed such that high-NFCC individuals would be less, and not more, influenced by priming than low-NFCC ones.

Overview of Research

We tested our conceptualization of how epistemic motivation affects priming in three studies. In Study 1, we sought evidence for the prediction that high-NFCC individuals would be less affected by priming than low-NFCC individuals when a social context has a salient situational norm. In Study 2, we directly manipulated priming and norm together to explore their relative influence on the behavior of individuals who differ in NFCC. In Study 3, we primed both high- and low-NFCC individuals with the same construct and observed their behavior in two consecutive contexts, a norm-present context followed by a norm-absent one. This would allow us to simultaneously replicate past research (Ford & Kruglanski, 1995) in the norm-absent context and find support for a completely opposite pattern in the norm-present context.

Study 1

In this initial study, we sought to provide evidence for the prediction that a reversal of the interaction pattern between priming and NFCC found by Ford and Kruglanski (1995) should happen when a salient situational norm exists. As past research has demonstrated that a powerful norm to comply with the experimenter exists in a lab setting (Milgram, 1974), we used a compliance paradigm to study participants’ behavior. In particular, the experimenter posed increasingly arbitrary and demanding requests to the participants, and the extent to which participants acceded to those requests served as our measure of compliance. At the same time, we primed half of the participants with the construct of self-confidence. We predicted that, in this situation, high-NFCC individuals would rely on the salient situational norm of compliance and behave accordingly, illustrating little if any effect of priming; low-NFCC individuals, however, should explore alternative cues to this salient norm to guide their behavior and thus remain susceptible to priming.

Method

Participants

In all, 111 (75 female, 36 male) introductory psychology students participated in this study to fulfill course requirement.

Materials

We used a word search paradigm (Bargh, Gollwitzer, Lee-Chai, Barndollar, & Trötschel, 2001) as the priming manipulation. We created two 25 by 26 letter matrices, each containing 30 words hidden in the puzzle. A list containing all the words that needed to be found was attached below the matrices. For the confidence priming task, 16 of the 30 words were confidence-related words: brave, persistent, courageous, determined, dominant, confident, firm, gutsy, certain, unafraid, sure, bold, conviction, spirited, assured, and steadfast. The other 14 words were neutral words (e.g., fiddle, crispy). For the neutral priming condition, all 30 words were unrelated to confidence (e.g., battery, grain). Words were randomly arranged in the word bank and in the matrices.

Procedure

Participants completed the experiment in individual cubicles. They first completed Webster and Kruglanski’s (1994) Need for Closure Scale before moving on to the word search task. Depending on the condition, participants received either the puzzle with confidence-related words or the one with all neutral words. Their task was to circle in the matrices all the words listed at the bottom. Once they completed the task, the experimenter collected the word search materials.

In the next part of the experiment, all participants were asked to comply with up to five requests for additional work. The requests were presented one by one, and the next request was only presented when participants agreed to and completed the previous request. If at any point the participant refused any request or made an excuse in an attempt to refuse any request, no more requests were made. The five requests were ordered in terms of increasing effort and arbitrariness. In this way, the dependent measure of compliance was the number of tasks participants agreed to perform.

Specifically, after the word search task, participants were then told a cover story: “We are running behind with data entry on this study and since the other participants are far behind on their word search task, would you be willing to help us out?” The tasks were then presented as follows:

Participants were asked to type one page of the data (20 scores) from their questionnaires into a computer.

If they agreed to and completed Task 1, participants were then asked to type in two sheets of the same data (40 scores) from previous participants.

If they agreed to and completed Task 2, participants were then asked to arrange approximately 25 pages of data in numerical order by subject number.

If they agreed to and performed Task 3, participants were asked to straighten up the cubicle room in which they were seated (trash and paper had been thrown about the room prior to their entry).

If they agreed to and performed Task 4, participants were finally asked to buy the experimenter a coke (coke machines were about a 3-min walk downstairs and 60 cents was given to participants who agreed).

The experimenter, who was blind to the participant’s condition, had been trained to present these requests in a consistent manner without any implied coercion.

To affirm our assumption that a salient norm to comply exists in this compliance setup, we conducted a pilot study on 40 undergraduates. We described the situation that actual participants would go through and asked them “How many requests would you comply with?” Pilot participants reported that they would comply with 2.98 requests on average and 35.3% of people reported that they would comply with all five requests. In addition, out of 40 pilot participants, 38 (95%) reported that they would comply with at least 1 request. These results suggest that individuals recognized that there existed a salient norm to comply with the experimenter in a lab situation.¹

At the end of the experiment, we administered a funneled debriefing (Bargh & Chartrand, 2000) to probe participants’ suspicion of the true purpose of the experiment. No participant came close at guessing the relationship between the priming and the compliance paradigm.

Results

We investigated the moderating role of NFCC in the priming effect by regressing the number of requests participants complied with on the priming manipulation, participants’ NFCC scores, and their interaction term. We coded the priming manipulation as 1 (neutral priming) and −1 (confidence priming), and mean-centered participants’ scores on the NFCC scale for computing the interaction term (Aiken & West, 1991). Likewise, NFCC was mean-centered for all future regression analyses. The regression model (adjusted R² = .14, p < .001) yielded a significant main effect of priming, B = .46, t = 3.74, p < .001, such that participants in the confidence priming condition (M = 3.27, SD = 1.50) complied with fewer tasks than those in the neutral priming condition (M = 4.22, SD = 1.11).² More important, this main effect of priming was qualified by a significant interaction term, B = −.016, t = −2.43, p = .017. The effect of NFCC did not reach significance, B = .009, t = 1.39, p = .17.

To evaluate the nature of this interaction (Figure 1), we conducted spotlight analyses on the effect of priming at one standard deviation above and below the mean NFCC score (Aiken & West, 1991). For high-NFCC individuals, the priming manipulation had no significant effect, B = .16, t < 1, p = .36. On the other hand, low-NFCC individuals in the confidence priming condition complied with significantly fewer tasks than those in the neutral priming condition, B = .75, t = 4.36, p < .001.

Figure 1.

The interaction effect between confidence priming and need for cognitive closure on the number of tasks participants complied with Study 1.

Discussion

Study 1 provided initial support for our hypothesis that under situations with strong behavioral norms, high-NFCC individuals would be less influenced by priming than low-NFCC individuals. As both our pilot test results and participants’ high compliance rate in the control condition strongly suggested, the context for Study 1 participants had a strong behavioral norm of compliance. In such situations, high-NFCC people readily adhered to the norm to satisfy their need for cognitive closure and terminated any further search for other cues to behavior, such as the primed construct of confidence. On the other hand, low-NFCC participants continued to be influenced by priming, presumably because they responded to the norm with an open mind, and continued to search for additional cues to guide their behavior. This continued search afforded them the opportunity to pick up the effect of priming. In other words, the presence of a strong norm leads to the opposite interaction pattern between priming and NFCC, compared with that found by Ford and Kruglanski (1995), where presumably no salient norm exists.

While it is encouraging that we found support for our hypothesis, one potential limitation of Study 1 was that we did not demonstrate directly that high-NFCC individuals adhered to the situational norm and were more influenced by it than low-NFCC individuals.³ That is, while it is true that high-NFCC were less influenced by the prime, our results do not speak to whether they relied more heavily on the situational norm per se. Hence, we manipulated both the prime and the norm in Study 2 to directly compare their influence on low- and high-NFCC individuals in a norm-present context.

Study 2

In Study 2, we sought to demonstrate that situational norms could replace the primed construct as the predominant knowledge structure for high-NFCC individuals to seize on, and for low-NFCC individuals to move on from as they look for additional cues to guide their behavior. Specifically, we sought to demonstrate that not only would high-NFCC individuals be less influenced by priming, but they would also be more influenced by normative information than low-NFCC participants. To achieve this, after priming participants with either cooperation or competition, we presented them with a variant of the Prisoner’s Dilemma (PD) paradigm (Kay & Ross, 2003; Liberman, Samuels, & Ross, 2004), in which they could choose to behave in a competitive or cooperative fashion. We modified the PD paradigm such that a salient situational norm was provided. We predicted that whether the salient norm or the prime prevails in guiding behavior depends on participants’ level of NFCC. While high-NFCC individuals should adhere to the salient norm and behave in a norm-consistent way, low-NFCC individuals should explore other cues for their behavior and be susceptible to the priming manipulation. In other words, we expect to find that high-NFCC individuals, as compared with low-NFCC individuals, would be simultaneously less influenced by the priming manipulation, as we found in Study 1, and more influenced by the norm manipulation.

Method

Participants

Eighty-one (55 female, 26 male) introduction psychology students participated in the experiment in partial fulfillment of a course requirement.

Design

The study had a 2 (Priming: Cooperative vs. Competitive) × 2 (External Norm: Cooperative vs. Competitive) between-subjects factorial design. Participants were randomly assigned to one of the four conditions. Participants’ dispositional NFCC was recorded as an individual difference measure.

Materials

For the scrambled sentence task, a series of 24 non-grammatical five-word sequences were adopted directly from Kay and Ross (2003). Consistent with previous research (Srull & Wyer, 1979), participants were asked to use four of the five words in each sequence to form coherent sentences. In 16 of these sequences (critical items), one word related (depending on experimental condition) either to competition (e.g., tournament, cut-throat) or to cooperation (e.g., fair, alliance) was included.

An unnamed PD game was prepared with the following pay-off matrix: one that awarded 20 points to each player if they both opted for move “A,” 10 points for each player if they both opted for move “B,” and 25 points for the “defector” and 5 points for the “cooperator” if one player opted for move B and the other opted for move A. In this specific context, thus, move B is the more competitive choice and move A is the cooperative choice.

Participants’ dispositional epistemic motivation was measured by the Need for Closure Scale (Webster & Kruglanski, 1994).

Procedure

On arrival to the lab, participants were led to individual computers. After completing the Need for Closure Scale (Webster & Kruglanski, 1994) among other filler questionnaires, participants were told that they would be taking part in a study investigating people’s reaction toward a particular situation. Before working on the experiment, however, they were told that the research team would greatly appreciate their cooperation in completing a short linguistic test. In actuality, the linguistic task was the scrambled sentence task and was described as a newly developed test from the Department of Linguistics that assesses people’s verbal ability.

Participants were then thanked and instructed to proceed to the “actual” part of the experiment. Participants were given the rules of the unnamed version of the PD game (the word “game” was never used), and were presented with the pay-off matrix. They were prompted to consult the experimenter should they encounter any difficulty understanding the matrix.

On top of the standard PD paradigm, at this point, we added an external norm manipulation whereby we fed participants with some normative data.

Although roughly 70% of the participants who have previously surveyed the situation reported that they were likely to choose “A” (“B”) if they were put in the same situation, we would like to know your opinion of the situation.

In this way, participants either received a cooperative norm (70% people choose “A”) or a competitive norm (70% people choose “B”). Once participants understood the situation, they reported their behavioral intention by responding to the question, “Knowing nothing about the identity of the other person other than he/she is an IU student, do you think you would choose move A or B?” along a 5-point Likert scale: Definitely Move A, Probably Move A, 50/50 chance of A or B, Probably Move B, and Definitely Move B. Hence, larger values indicated a more competitive behavioral intention.

Finally, they were given a funneled debriefing (Bargh & Chartrand, 2000) to probe their level of awareness regarding the nature and the possible impact of the priming manipulation, and/or any suspicion they had regarding the relationship between the first and second tasks. No participants expressed suspicion or could report the true purpose of the experiment.

Results

The hypotheses regarding participants’ behavioral intention in general were that (1) high-NFCC participants’ behavioral intentions would be influenced by the norm manipulation and not by the priming manipulation and (2) low-NFCC participants, as they were less motivated to adhere to the norm, would continue to search for alternative interpretations of the situation, during which process they would be influenced by the primed construct.

We tested these hypotheses by regressing participants’ behavioral intention on norm manipulation (cooperative norm = −1, competitive norm = 1), priming manipulation (cooperative prime = −1, competitive prime = 1), and NFCC, and the second- and third-order interaction terms. The regression model (adjusted R² = .22, p = .001) yielded a significant main effect of the priming manipulation, B = .25, t = 2.20, p = .033, and a significant main effect of the norm manipulation, B =.33, t = 2.84, p = .006. On average, participants in the competitive priming condition (M = 3.18, SD = 1.11) were more inclined to choose the competitive move than those in the cooperative priming condition (M = 2.65, SD = 1.17). Participants in the competitive norm condition (M = 3.19, SD = 1.12) also intended to choose the competitive move more than those in the cooperative norm condition (M = 2.60, SD = 1.15). More importantly, the predicted two-way interactions between priming and NFCC, B = −.015, t = −2.29, p = .025, and between norm manipulation and NFCC, B =.023, t = 3.35, p = .001, were both significant. No other significant effects were obtained (all other ps > .38).

To explore the moderating role of NFCC on priming (Figure 2), we conducted spotlight analyses on priming at one standard deviation below and above the mean of NFCC, respectively. At low NFCC, the effect of priming predicted participants’ behavioral intention to be competitive, B =.52, t = 2.92, p = .005. On the other hand, at high NFCC, the effect of priming did not predict behavioral intention, B =.014, t = .08, p = .94. Consistent with our prediction, high-NFCC individuals were less influenced by priming than low-NFCC individuals in this norm-present context.

Figure 2.

The effect of norm, prime, and NFCC on behavioral intention in Study 2.

The moderating role of NFCC on the effectiveness of the norm manipulation (Figure 2) was also explored using spotlight analyses. At low NFCC, the effect of norm manipulation failed to predict participants’ intention to behave competitively, B = −.14, t = −.80, p = .42. At high NFCC, in contrast, the norm manipulation strongly predicted behavioral intention, B = .75, t = 4.11, p < .001. This suggests that high-NFCC participants’ behavioral intention was guided by the normative information to a greater extent than that of low-NFCC participants.

Discussion

Study 2 replicated the findings of Study 1 that high-NFCC individuals were less influenced by priming than low-NFCC individuals in a situation with a salient norm. In addition, we were able to show that at the same time as high-NFCC participants relied less on priming, they relied primarily on the normative information. Together, this finding corroborates our conceptualization that in a context with a salient norm, the situational norm replaces priming as the predominant cue for high-NFCC individuals to engage in knowledge acquisition of the environment (Kruglanski & Webster, 1996). The consequence with regard to NFCC is that high-NFCC individuals cling to the normative information in construing the social situation with an unwillingness to revise this initial understanding. Hence, they were less likely to be influenced by the priming manipulation. On the other hand, low-NFCC individuals are willing to move beyond the normative information and seek additional information (here, in the form of the highly accessible primed construct) as an alternative cue for understanding the social context. It is at this latter stage of processing that low-NFCC individuals are influenced by the primed constructs.

As discussed previously, it is notable that the pattern of results was observed under conditions in which the priming manipulation occurred before the norm manipulation. In other words, the primed construct (cooperation or competition) was already highly accessible before participants received the information about the social norm. Our results suggest that even with a clear temporal primacy, the primed construct did not become the default knowledge structure for high-NFCC individuals to seize on. Again, this may be due to the special importance of normative information to social animals like human beings. But we also think these data provide strong support for Loersch and Payne’s (2011) argument that individuals use the primed construct flexibly such that it only exerts its effect when individuals are engaged in a relevant inquiry about the context (e.g., “How should I behave in this situation?”). In our final study, we aimed to further demonstrate the flexibility of participants’ recruitment of the primed construct in guiding their behavior.

Study 3

If the primed construct is really used flexibly by the individual, it stands to reason that the same primed construct could be applicable to two consecutive social contexts, one with a salient norm followed by one without a salient norm. In this case, the same primed construct should be neglected by high-NFCC individuals in the norm-present context but be utilized by the same high-NFCC individuals to guide their behavior in the norm-absent context. This was exactly what we aimed to show in Study 3. We chose to illustrate the moderating role of NFCC in two well-known contexts in which priming effects have been previously observed. Specifically, after priming participants with politeness-related, rudeness-related, or neutral words, we recorded their behavioral responses in the interruption paradigm (Bargh et al., 1996, Study 1) followed by a variant of the Donald paradigm (Ford & Kruglanski, 1995; Higgins et al., 1977).

In the interruption paradigm, participants’ politeness/rudeness was indexed by the amount of time they waited before interrupting a conversation between an experimenter and a confederate. Bargh et al. (1996) reported that the percentage of control participants who did not interrupt for as long as 10 full minutes was higher than 50%. As a result, the interruption paradigm appears to be associated with a salient norm not to interrupt. We conducted a pilot study to test this assumption. Twenty-five students were given the description of the Bargh et al. (Study 1) situation and were asked to imagine themselves as the actual participant. They were then asked to indicate their perceived norm (“How do you assume most people would behave in this situation?”) on an 8-point scale (1 = definitely interrupt, 2 = most likely to interrupt, 3 = probably interrupt, 4 = inclined to interrupt, 5 = inclined to wait, 6 = probably wait, 7 = most likely to wait, 8 = definitely wait). The average response was 6.08, indicating that pilot participants generally thought that most people would prefer to wait. Indeed, 23 of the 25 (92%) responded with one of the options to wait (within the 5-8 range). Given the strong norm in this situation, we expected that high-NFCC individuals would be more likely to follow the salient norm to wait and be less influenced by priming.

On the other hand, the Donald paradigm, in which Ford and Kruglanski (1995) have previously found that high-NFCC individuals were more susceptible to priming, represented a context without a salient norm. Hence, we expected to replicate the effect in Ford and Kruglanski in this context. Together, the experimental setup would allow us to observe, after processing the prime construct with respect to a norm-present context, whether the same group of participants could immediately process it flexibly and differently, based on their NFCC, in a subsequent norm-absent context.

Method

Participants

Ninety-seven (62 female, 35 male) psychology students participated in the experiment in partial fulfillment of a course requirement.

Materials

The priming materials were adopted from Bargh et al. (1996, Study 1) and were similar to those used in Studies 2 and 3. We assembled three sets of 30 non-grammatical 5-word sequences for the polite, rude, and neutral conditions, respectively. Fifteen sequences from each set contained words related to the intended construct. For the polite priming condition, the critical stimuli were respect, honor, considerate, appreciate, patiently, cordially, yield, polite, cautiously, courteous, graciously, sensitively, discreetly, behaved, and unobtrusively. For the rude priming condition, the critical stimuli were aggressively, bold, rude, bother, disturb, intrude, annoyingly, interrupt, audaciously, brazen, impolitely, infringe, obnoxious, aggravating, and bluntly. For the neutral condition, the critical stimuli were exercising, flawlessly, occasionally, rapidly, gleefully, practiced, optimistically, successfully, normally, send, watches, encourages, gives, clears, and prepares.

For the Donald task, we created a modified passage describing Donald (see Appendix, available online at psp.sagepub.com/supplemental). The passage was tailored such that Donald’s behaviors were ambiguous with regard to whether he was polite or rude.

Procedure

Participants took part in the experiment one at a time. After signing the consent form, they were led to a computer station to begin the experiment. Before they started, the experimenter told participants that he would be in another room down the hallway and that they should look for him when the first part of the experiment was over. They first completed the Need for Closure Scale (Webster & Kruglanski, 1994). After that, they were presented with the “Scrambled Sentence Task,” introduced as a linguistic task developed by the Linguistics Department. Participants were told that their job was to use four of the five words presented to form a grammatically correct sentence. Once they were done with this part, they looked for the experimenter as instructed.

Consistent with the interruption paradigm used in Bargh et al. (1996, Study 1), the experimenter was stationed in another room with a confederate posing as another participant. A second confederate was stationed along the hallway in between the two rooms. The experimenter and both confederates were blind to the condition. As soon as the participant came out from the laboratory, the confederate in the hallway started a stopwatch and signaled to the experimenter and the first confederate, who immediately began a scripted discussion over some bogus experimental materials. Ostensibly, the “other participant” was having trouble understanding the experimental materials and the experimenter was patiently explaining the instructions. The experimenter was standing in the doorway of the room such that his back was facing the hallway, which allowed him to conveniently ignore the participant’s presence. Participants could only wait for the conversation to be over or interrupt the conversation to get the experimenter’s attention. The confederate stationed along the hallway surreptitiously stopped the stopwatch when the participant interrupted the conversation. The staged conversation was stopped automatically after 10 min if participants never interrupted.

Once the experimenter acknowledged the participant’s presence (or the 10 min were up), he led the participant back to the lab to begin the Donald task. The Donald task was introduced as an impression formation task. Participants were asked to form an impression about Donald’s personality on the basis of a short description of Donald. After reading the passage, participants rated their impression of Donald on 8 critical items that were related to the construct of politeness/rudeness: thoughtful, aloof, disrespectful, polite, conceited, insincere, grateful, rude, each along a 7-point scale from 1 (not at all) to 7 (very much). These critical items were embedded among 10 other filler items: obsessive, persistent, independent, listless, stubborn, self-confident, sly, neat, motivated, and assertive. The average score of participants’ ratings on the critical items (α = .73) served as the index of participants’ politeness rating of Donald.

To affirm that the 8 critical items were indeed related to the construct of politeness/rudeness, a separate group of 47 undergraduates were asked to indicate for each of the 18 trait adjectives “whether knowing that a person has the particular trait is able to tell you if the person POLITE or RUDE” on a 4-point scale (1 = not at all, 2 = slightly, 3 = somewhat, 4 = strongly). The 8 critical items had a much higher average rating (M = 3.42, SD = .34) than the control items (M = 1.88, SD = .45), t(46) = 24.09, p < .001. In fact, all but one trait, aloof (4.3%), elicited 0% “not at all” responses (M = 0.5%). On the other hand, the filler items elicited significantly more “not at all” responses (M = 40.0%), χ² = 48.2, p < .001, ranging from 27.7% to 59.6%.

Finally, participants filled up some demographic information before they were thanked and fully debriefed.

Results

Overview

The primary dependent measures were the time participants took to interrupt the conversation between the experimenter and the confederate and their politeness ratings of Donald.

Interruption time

To test whether high-NFCC participants were less affected by priming than were low-NFCC participants in this task, we regressed the number of seconds participants waited before interrupting the conversation between the experimenter and the confederate on priming manipulation, NFCC, and their interaction term. Because we predicted that the effect of priming would follow a linear trend, the priming manipulation was coded accordingly (1 = polite, 0 = neutral, −1 = rude). The regression model (adjusted R² = .23, p < .001) yielded a significant effect of priming, B = 116.39, t = 4.97, p < .001, suggesting that participants’ waiting time in the three conditions—politeness priming (M = 540s, SD = 148s), neutral priming (M = 468s, SD = 182s), and rudeness priming (M = 312s, SD = 233s)—followed a linear trend.⁴

More importantly, the significant interaction term was significant, B = −131.10, t = 2.54, p = .013. To probe the nature of the interaction between priming and NFCC (Figure 3), we conducted spotlight analyses (Aiken & West, 1991) on the effect of priming at low (1 SD below the mean) and high (1 SD above the mean) levels of NFCC, respectively. As predicted, in this norm-present context, the effect of priming was significant among low-NFCC individuals, B = 173.54, t = 5.20, p < .001, and not among high-NFCC individuals, B = 59.23, t = 1.88, p = .07.

Figure 3.

The interaction effect between priming and need for cognitive closure on the time participants waited before interrupting in Study 3.

Ratings of Donald

Similarly, to test whether high-NFCC participants were more affected by priming than were low-NFCC participants in the norm-absent task, we regressed participants’ politeness rating of Donald on priming manipulation (1 = polite, 0 = neutral, −1 = rude), NFCC, and their interaction term. The regression model (adjusted R² =. 087, p = .036) yielded a significant interaction term, B = .55, t = 2.68, p = .009. No other effects were significant, both ps > .28. To probe the nature of the interaction (Figure 4), we conducted spotlight analyses (Aiken & West, 1991) on the effect of priming at both low (1 SD below the mean) and high (1 SD above the mean) levels of NFCC. Consistent with our prediction, priming did not affect low-NFCC participants’ rating of Donald, B = −.14, t = −1.10, p = .30, but did significantly predict high-NFCC participants’ rating, B = .34, t = 2.73, p = .008. This indicates that high- but not low-NFCC participants’ politeness ratings of Donald were significantly influenced by our priming manipulation, replicating the findings of Ford and Kruglanski (1995).

Figure 4.

The interaction effect between priming and need for cognitive closure on the politeness rating of Donald in Study 3.

Testing for a three-way interaction effect

Finally, the fact that we predicted that the same group of participants would respond to the priming manipulation in two consecutive contexts meant that we were expecting a three-way interaction between priming, NFCC, and the two dependent variables as levels of the same within-subjects factor. As a result, a General Linear Model was computed on the standardized scores of the two DVs treating DV as a within-subject factor, priming condition as a between-subjects factor, and NFCC as a continuous individual difference score. Indeed, the analysis revealed a significant three-way interaction, F(2, 91) = 3.19, p = .046, $η_{p}^{2}$ = .066, corroborating the results of the regression analyses.

Discussion

We demonstrated that individuals were able to flexibly utilize the same primed construct, politeness (rudeness), in two consecutive contexts within the same experiment, depending on the presence of a situational norm. In the interruption paradigm, where a salient norm to be polite was present, we replicated the findings from Studies 1 and 2, such that high-NFCC individuals were less susceptible to the influence of priming than low-NFCC individuals. This is presumably because high-NFCC participants adhered to the salient norm and ignored the primed construct, whereas low-NFCC participants sought alternative cues for behavior and were more influenced by priming.

On the other hand, in the Donald paradigm, we replicated the classic finding that high-NFCC individuals, as opposed to low-NFCC ones, were more influenced by priming (Ford & Kruglanski, 1995). Here, in the absence of a salient norm, high-NFCC participants seized on the most readily available information, namely, the highly accessible primed construct, and terminated search for additional cues in the environment. In contrast, for low-NFCC participants, the search for alternative cues to guide their behavior made them relatively less susceptible to the priming effect. The fact that these two opposite findings were observed on different dependent measures within the same study using the same priming manipulation testifies to the flexibility with which individuals use the primed construct in guiding their self-regulation of behavior.

General Discussion

In three studies, we found that the moderating role of NFCC on people’s susceptibility of priming depends on whether a salient norm exists in the social context to compete with the primed construct in guiding people’s behavior. In contexts with a strong situational norm, we consistently found that high-NFCC individuals were less susceptible to priming, a pattern directly opposite to past research conducted in situations without a salient norm (Ford & Kruglanski, 1995). This was presumably because high-NFCC participants seized on the salient norm in the situation to achieve cognitive closure and ceased to consider the more implicit primed construct. In contrast, low-NFCC individuals avoided basing their behavior exclusively on the normative information, and, while searching for alternative cues in the environment, continued to be influenced by the highly accessible priming construct. Indeed, when we manipulated the priming and normative information simultaneously in a PD paradigm (Study 2), we found that high-NFCC individuals were less affected by priming but more influenced by situational norms. This is more direct evidence that in norm-present context, normative information replaces primed constructs as the knowledge structure high-NFCC individuals seize on to achieve cognitive closure.

Finally, in Study 3, we demonstrated that both high- and low-NFCC individuals could flexibly use the same primed construct in two consecutive contexts, one with a situational norm and one without. In the interruption paradigm (Bargh et al., 1996, Study 1), a situation with a salient norm not to interrupt, high-NFCC participants were less susceptible to the priming manipulation than low-NFCC participants. The effect was reversed, however, in an immediately subsequent context, in which we presented participants with a variant of the Donald paradigm (Higgins et al., 1977; Srull & Wyer, 1979). Here, the same primed construct had little impact on low-NFCC participants whereas it biased high-NFCC participants’ perception of the ambiguous social target (Ford & Kruglanski, 1995). This pattern strongly suggests that the primed construct, although it was made highly accessible for all participants, can be used in a highly flexible way and participate in multiple episodes of prime-norm dynamic.

NFCC Guides Selection of Environmental Cues

The current findings underscore the importance of chronic epistemic motivation in individuals’ selection of environmental cues for self-regulation. In any given social environment, there could be multiple cues that individuals might rely on in guiding their behavior. Individuals’ chronic tendency in achieving (high-NFCC) or avoiding (low-NFCC) cognitive closure has been shown to bias their attention to either the most readily accessible cue or to other potential alternatives (Kruglanski & Webster, 1996). Based on the temporal primacy of the primed construct, past research has hypothesized and found that people with a high-NFCC were more likely than low-NFCC individuals to be influenced by these social constructs that had been made accessible in an unrelated, preceding context (Ford & Kruglanski, 1995). Given the prevalence of the environmental influence that priming represents, this classic finding suggested that low-NFCC individuals are a group of individuals who would always be less influenced by a prior social encounter.

The current findings demonstrate that the same seizing and freezing principle associated with NFCC (Kruglanski & Webster, 1996) could be applied to social contexts that have a salient norm, whereby normative information competes with the primed construct in influencing individuals’ behavior. In these situations, we actually predicted and found the opposite pattern such that individuals with high NFCC would be less susceptible to the influence of priming. At the same time, we also demonstrated that the behavior of individuals with high NFCC adhered more closely to the normative information that appeared later in time compared with priming. Thus, relying on the individual difference in NFCC, we were able to demonstrate for the first time in the literature that high-NFCC individuals may not always utilize early cues and show primacy effects, a hallmark of NFCC (Kruglanski & Webster, 1996), in face of a compelling alternative cue such as situational norms. Given that the processes we revealed here were conceptually specific to the construct of NFCC, it is highly unlikely that the current findings are due to other individual differences that correlate with NFCC. This is especially the case when we consider our findings from Study 3, indicating that NFCC led to completely opposite patterns of results in the two consecutive contexts. Hence, the present research testifies to the notion that one’s level of NFCC is a stable individual difference⁵ that plays an important role in one’s interaction with the cues for behavior across different kinds of social contexts. Future research could extend the current findings by examining whether situational moderators of NFCC (e.g., increased cognitive load; Ford & Kruglanski, 1995) would produce similar effects.

Interplay Between Prime and Other Social Cues

By utilizing individuals’ chronic tendencies in NFCC, we were able to reveal an interesting aspect of prime-norm dynamic that has not been shown before. Past research studying the interplay between prime and norm has been focused on how priming can bias individuals’ norm construal in an ambiguous situation. For instance, Hertel and Kerr (2001) primed participants with either the concept of “loyalty” or “equality” before administering a standard minimal group paradigm (Tajfel, Billig, Bundy, & Flament, 1971). After participants reported their resource allocation to their in-group versus out-group, the authors asked them for the perceived norm in the situation. Participants exhibited a greater degree of in-group favoritism in the loyalty priming condition than in the equality priming condition. More important, the authors found that the priming manipulation indeed biased participants’ perceived situational norm, such that participants in the loyalty priming condition felt that their group expected them to use an in-group favoritism strategy while those in the equality priming condition perceived the group norm as one that emphasized equality. This perceived situational norm in turn mediated the prime-to-behavior effect. Our findings suggest that another way in which primes and salient situational norms interact is to act as alternative cues for the individual to guide self-regulation. The use of NFCC allowed us to reveal this prime-norm dynamic such that norms dominated primed constructs (independent of their temporal primacy) in high-NFCC individuals’ inquiry into the social environment.

The revelation of this novel aspect of prime-norm dynamic not only broadens our understanding of how these two important social cues compete to guide behavior, but also has implications for other sources of information that could potentially interact with priming. Although we have argued that the unique importance of situational norms partly contributed to their capacity to replace primed constructs as the initial cue with which high-NFCC individuals begin to acquire knowledge, it is also plausible that other social cues can compete with priming in a similar fashion. Potential candidates for these social cues include norms within a social group (Kruglanski, Pierro, Mannetti, & De Grada, 2006) and stereotypes (Kruglanski & Freund, 1983), constructs to which individuals with a high NFCC, in their perceptual and behavioral output, readily succumb. In the presence of these preexisting knowledge structures, individuals with a high NFCC could quickly end further information processing and be less influenced by priming manipulations or subconscious influences from other stimuli in the environment. On the other hand, low-NFCC individuals, who are motivated to avoid closure, would seek out additional cues and be further subject to the implicit influence of primed constructs. In sum, we believe that the present research is illustrative of a broader and consistent pattern by which individuals’ NFCC influences the interplay between priming and other social cues in guiding their thoughts and behavior.

Conclusion

In the present research, we found that low-NFCC individuals, those who avoid achieving early closure in understanding the social environment, could under some conditions be more influenced by priming. The key difference between the current results and the classic finding that high-NFCC individuals were more susceptible to priming (Ford & Kruglanski, 1995) was the presence of a strong situational norm, which replaces the primed construct as the starting point of individuals’ knowledge acquisition about the social environment. Besides underscoring the importance of NFCC in people’s selection of environmental cues, our findings are consistent with the theoretical perspective (Loersch & Payne, 2011; Wheeler et al., 2007) that highlights the flexibility with which individuals can use the primed construct. Our findings also pave the way for future investigations of the interactive effects of multiple social influences on people’s cognition and behavior.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: SAGE.

Ajzen

(1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179-211.

Armitage

C. J.

Conner

(2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40, 471-499.

Bargh

J. A.

Chartrand

T. L.

(2000). The mind in the middle: A practical guide to priming and automaticity research. In Reis

Judd

(Eds.), Research methods in social psychology (pp. 253-285). New York, NY: Cambridge University Press.

Bargh

J. A.

Chen

Burrows

(1996). Automaticity of social behavior: Direct effects of trait construct and stereotype activation on action. Journal of Personality and Social Psychology, 71, 230-244.

Bargh

J. A.

Gollwitzer

P. M.

Lee-Chai

Barndollar

Trötschel

(2001). The automated will: Nonconscious activation and pursuit of behavioral goals. Journal of Personality and Social Psychology, 81, 1014-1027.

Chiu

Morris

M. W.

Hong

Menon

(2000). Motivated cultural cognition: The impact of implicit cultural theories on dispositional attribution varies as a function of need for closure. Journal of Personality and Social Psychology, 78, 247-259.

Cialdini

R. B.

Kallgren

C. A.

Reno

R. R.

(1991). A focus theory of normative conduct: A theoretical refinement and reevaluation of the role of norms in human behavior. Advances in Experimental Social Psychology, 24, 201-234.

DeMarree

K. G.

Wheeler

S. C.

Petty

R. E.

(2005). Priming a new identity: Self-monitoring moderates the effects of non-self primes on self-judgments and behavior. Journal of Personality and Social Psychology, 89, 657-671.

10.

Dijksterhuis

van Knippenberg

(1998). The relation between perception and behavior, or how to win a game of Trivial Pursuit. Journal of Personality and Social Psychology, 74, 865-877.

11.

Fishbein

Ajzen

(1975). Belief, attitude, intention, and behavior: An introduction to theory and research. Reading, MA: Addison-Wesley.

12.

Ford

T. E.

Kruglanski

A. W.

(1995). Effects of epistemic motivations on the use of accessible constructs in social judgment. Personality and Social Psychology Bulletin, 21, 950-962.

13.

Freedman

J. L.

Fraser

S. C.

(1966). Compliance without pressure: The foot-in-the-door technique. Journal of Personality and Social Psychology, 4, 195-202.

14.

J. H.

Morris

M. W.

Lee

Chiao

Chiu

Hong

(2007). Epistemic motives and cultural conformity: Need for closure, culture, and context as determinants of conflict judgments. Journal of Personality and Social Psychology, 92, 191-207.

15.

Hertel

Kerr

N. L.

(2001). Priming in-group favoritism: The impact of normative scripts in the minimal group paradigm. Journal of Experimental Social Psychology, 37, 316-324.

16.

Higgins

E. T.

Rholes

W. S.

Jones

C. R.

(1977). Category accessibility and impression formation. Journal of Experimental Social Psychology, 13, 141-154.

17.

Hull

J. G.

Slone

L. B.

Meteyer

K. B.

Matthews

A. R.

(2002). The nonconsciousness of self-consciousness. Journal of Personality and Social Psychology, 83, 406-424.

18.

Kay

A. C.

Ross

(2003). The perceptual push: The interplay of implicit cues and explicit situational construals on behavioral intentions in the Prisoner’s Dilemma. Journal of Experimental Social Psychology, 39, 634-643.

19.

Kruglanski

A. W.

Freund

(1983). The freezing and unfreezing of lay inferences: Effects on impressional primacy, ethnic stereotyping, and numerical anchoring. Journal of Experimental Social Psychology, 19, 448-468.

20.

Kruglanski

A. W.

Webster

D. M.

(1996). Motivated closing of the mind: “Seizing” and “freezing”. Psychological Review, 103, 263-283.

21.

Kruglanski

A. W.

Pierro

Mannetti

De Grada

(2006). Groups as epistemic providers: Need for closure and the unfolding of group-centrism. Psychological Review, 113, 84-100.

22.

Liberman

Samuels

S. M.

Ross

L. D.

(2004). The name of the game: Predictive power of reputations versus situational labels in determining Prisoner’s Dilemma game moves. Personality and Social Psychology Bulletin, 30, 1175-1185.

23.

Loersch

Payne

K. B.

(2011). The situated inference model: An integrative account of the effects of primes on perception, behavior, and motivation. Perspectives on Psychological Science, 6, 234-252.

24.

Milgram

(1974). Obedience to authority. New York, NY: Harper & Row.

25.

Petty

R. E.

DeMarree

K. G.

Briñol

Horcajo

Strathman

A. J.

(2008). Need for cognition can magnify or attenuate priming effects in social judgment. Personality and Social Psychology Bulletin, 34, 900-912.

26.

Roets

Van Hiel

(2007). Separating ability from need: Clarifying the dimensional structure of the need for closure scale. Personality and Social Psychology Bulletin, 33, 266-280.

27.

Schwartz

S. H.

Tessler

R. C.

(1972). A test of a model for reducing measured attitude-behavior discrepancies. Journal of Personality and Social Psychology, 24, 225-236.

28.

Srull

T. K.

Wyer

R. S.

(1979). The role of category accessibility in the interpretation of information about persons: Some determinants and implications. Journal of Personality and Social Psychology, 37, 1660-1672.

29.

Tajfel

Billig

M. G.

Bundy

R. P.

Flament

(1971). Social categorization and intergroup behaviour. European Journal of Social Psychology, 1, 149-178.

30.

Webster

D. M.

Kruglanski

A. W.

(1994). Individual-differences in need for cognitive closure. Journal of Personality and Social Psychology, 67, 1049-1062.

31.

Wheeler

S. C.

DeMarree

K. G.

Petty

R. E.

(2007). Understanding the role of the self in prime-to-behavior effects: The active-self account. Personality and Social Psychology Review, 11, 234-261.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.15 MB