Bragging in the right context: Impressions formed of self-promoters who create a context for their boasts

Abstract

Two studies examined the impression formed of self-promoters who actively work on creating the right context for their boasts. According to previous research, self-promotion in response to a question is perceived more positively. The current research claims that people commonly encourage their interactants to ask them a question pertaining to their success, and investigates the resulting impressions created. In Study 1 speakers were perceived more negatively when they raised the issue relevant to their self-promotion only when their interactant did not ask them a question concerning their success. Study 2 provided evidence regarding the cognitive process underlying the perception of the self-promoter. The results are discussed in terms of the correspondence bias, the mindlessness hypothesis, and models of reconstructive memory.

Keywords:

• Impression formation
• Impression management
• Self-promotion

Since the early writings of Goffman (1959) and Jones (1964) it has become clear that people frequently engage in impression management in order to influence how others perceive them. The tactics people use in order to be liked are commonly called ingratiation (Jones, 1990). These tactics include opinion conformity, helping behavior, and promotion of the other. In order to be perceived as competent, on the other hand, individuals have to engage in self-promotion by highlighting their personal achievements and unique characteristics (Giacalone & Rosenfeld, 1986).

Although there is abundant research documenting the various impression management tactics (e.g., Leary, 1995), researchers have paid much less attention to their perceptions (e.g., Carter & Sanna, 2006). Researchers in the field of person perception tend to ignore perceivers’ knowledge regarding impression management (Vonk, 1998, 1999). They assume that perceivers, as a result of the correspondent bias, tend to ascribe people's traits to their behavior, ignoring the social context that may have led to such behavior and thus disregarding impression management motives in their perception (Vonk, 1999).

Research that did examine perceptions of impression management showed that perceivers do pay attention to the social context in an attempt to understand the actors’ motives. Paying attention to the social context resulted in a greater chance of perceiving an ingratiator as engaging in impression management. For example, helping another person was perceived as an ingratiation attempt when this person was one's superior (Vonk, 1998).

On the other hand, paying attention to the social context might reduce the possibility of perceiving a self-promoter as engaging in impression management. This was shown to be the case in a study conducted by Holtgraves and Srull (1989). These authors studied the impact of a conversational context on the perception of self-promoters. Participants read a fictional conversation between two students and were asked their opinion of one of them. This target person was perceived more positively when his or her positive self-description followed a direct question by his or her interactant.

So, people can praise themselves and still be liked by their audience. They just have to wait patiently for their interactant to ask them about their accomplishments. However, what happens if the interactant does not ask the desired question? According to Holtgraves (1990, 2002), speakers may actively work to establish the right context for their positive self-statements. For example, they can raise the general issue, encouraging their interactant to ask them the awaited question. But engaging in such a behavior might be perceived, by itself, as an impression management attempt. Is it worthwhile adopting such an approach? Doesn’t the audience see through such tactic? And what is the underlying cognitive process that explains the audience's perception? The current research examines these questions.

It is argued here that self-enhancers are perceived more positively when they answer a direct question because of the attribution made for their behaviors. When people initiate a behavior, in our case a positive self-description, it can more easily be attributed to their internal characteristics (as braggers) than when the same behavior is a response to external factors (Jones & Davis, 1965) such as a direct question (Holtgraves, 2002). However, if the question is the result of an active effort by the self-promoters who raise the issue in order to initiate a question, the audience may realize the motivation behind this action and view them as manipulative.

The present research hypothesizes that self-promoters who raise the general issue relating to their success will indeed be perceived as boastful, but only if they do not succeed in making their interactants ask them the awaited question. If a question is asked, their active effort in creating the right context for their boast will not harm their perception.

This argument is based, in part, on the mindlessness hypothesis, according to which most everyday interactions are processed automatically without paying attention to the details of the messages (Langer, Blank, & Chanowitz, 1978). As a result of frequent exposure to similar situations, an individual pays attention only to the “scripted cue points” and tends to ignore the specific semantics of the message (Langer et al., 1978, p. 638). This mindless processing occurs as long as there is no violation of the expected norm (Burgoon, Berger, & Waldron, 2000). Since self-promotion in response to a question is normative, it is predicted that, in such a case, perceivers will not analyze the conversation mindfully.

However, since the question comes after the raising of the issue, how can it affect the processing of this earlier information? Previous research has documented the impact of individuals’ expectations at recall time on retrieval (e.g., Carli, 1999). A scheme that is activated after the encoding of information tends to bias the memory to fit the scheme. Based on this evidence, it is argued here that the existence of a question regarding the achievement of the interactant activates a scheme for a typical polite conversation in which people describe their success only when asked to do so. This scheme results in a reconstructed memory of the previous parts of the conversation in which the raising of the issue by the self-promoter is forgotten because it is inconsistent with the expectation (Hirt, 1990). Thus I hypothesized that a positive self-description following a question would result in a relatively positive perception of the target person, irrespective of whether the question was initiated by the fellow interactant or was the result of an active effort by the target person. However, if there were no question preceding the self-promotion, the modesty norm might be violated, leading the perceivers to analyze the conversation more mindfully. In such a case the perceivers would see through the active effort to create the context for self-promotion. Consequently they would perceive the target person in a more negative light when he or she encouraged the other interactant to ask the question than when the question was initiated by the fellow interactant.

These hypotheses were examined in an experiment based on the methodology of Holtgraves and Srull (1989). The participants were asked to form an impression of a target person who described his success in a fictional conversation with a fellow student. This conversation varied based on two factors: whether the target person or his interactant created a pre-context for the positive disclosure and whether the interactant asked the target person a question regarding his success.

Study 1

Method

Participants

The participants were 105 undergraduate students of a university and a college in Israel, of whom 72% were female, 26% were male, and 2 did not indicate their gender. Their mean age was 23.64 (SD = 1.73). The participants volunteered to participate in the experiment.

Materials and procedure

Participants were randomly assigned to four experimental conditions. The conditions were based on two levels of topic initiator (the target person or his interactant) and two levels of the existence of a question (whether a question was asked or not). In all conditions, participants received a paper that included the manipulations and measures of the dependent variables. The participants were informed that the questionnaire was part of a study on interpersonal communication.

Following the introduction, participants were asked to imagine the following scenario: Avi and Danny, two acquaintances who have known each other for a few years and study together at the university, meet in the cafeteria during a break. While drinking coffee, they talk about their activities during the preceding week and their plans for the weekend. Then, they have the conversation transcribed below:

In the following scenario Avi is the target who initiated the topic, and the question condition is indicated in parentheses.

Avi: “Tell me, how are you managing with your work?”

Danny: “I barely manage; I don’t have time to breathe. You work a lot too, don’t you?”

Avi: “Yes, I am exhausted. I saw that the statistics grades are already on the bulletin board.”

Danny: “Yes, I saw, I passed by it before.” (What did you get?)

Avi: “Actually, I’m doing great in school lately. I got an A+.”

Danny: “Really? Well, OK, we have to go. Class is starting.”

Avi: “Let's go.”

In the following scenario Danny, the interactant, initiated the topic and the question condition is indicated in parentheses.

Avi: “Tell me, how are you managing with your work?”

Danny: “I barely manage; I don’t have time to breathe. You work a lot too, don’t you?”

Avi: “Yes, I am exhausted.”

Danny: “I saw that the statistics grades are already on the bulletin board.” (What did you get?)

Avi: (“Yes.”–only when there is no question) “Actually, I’m doing great in school lately. I got an A+.”

Danny: “Really? Well, OK, we have to go. Class is starting.”

Avi: “Let's go.”

In half of the experimental conditions Avi, the target person, raised the grades issue, and in the other half it was Danny, his interactant, who raised it. In half of the experimental conditions Danny asked Avi what grade he received, and in the other half there was no question.

After reading the scenario, participants were asked to complete a questionnaire that measured the dependent variables. Following Holtgraves and Srull (1989), the participants were asked to indicate on a 5-point scale their impression of Avi with regard to a series of attributes relating to sociability and boastfulness. These attributes included: sociable, arrogant, manipulative, considerate, boastful, egotistical, and maintains good social relationships. Then the participants were asked to indicate on a 5-point scale how much they thought Avi liked Danny, Danny liked Avi, and the degree of closeness between Avi and Danny. They were also asked to indicate how much they liked Avi and to what degree they would have liked to be friends with a person like Avi.

Results

Computation of the dependent variables

Based on an exploratory factor analysis made on all of Avi's attributes, two indices were created: (1) a boastfulness index (M = 3.08, SD = 0.92, Cronbach's α = 0.85), based on the mean of the traits: arrogant, manipulative, boastful, and egotistical; and (2) a sociability index (M = 3.40, SD = 0.52, Cronbach's α = 0.61), based on the mean of the traits: sociable, considerate, and maintains good social relationships. For each index, values ranged from 1 to 5, with larger numbers representing greater boastfulness and sociability.

Based on another exploratory factor analysis, two additional indices were created: (1) Danny's and Avi's relationships (M = 3.22, SD = 0.57, Cronbach's α = 0.72), based on the mean of the questions: (a) How much does Avi like Danny?, (b) How much does Danny like Avi?, and (c) What is the degree of closeness between Avi and Danny? (2) The participants’ liking for Avi (M = 2.98, SD = 0.68, Cronbach's α = 0.84), based on the mean of the questions: (a) How much do you like Avi?, and (b) How much would you like to have a friend like Avi? For each index, values ranged from 1 to 5, with larger numbers representing a closer relationship between Avi and Danny and a stronger liking on the part of the participants for Avi.

The primary analysis

Two-way ANOVAs were conducted, with issue raising and question asking as independent variables, and each of the four indices as dependent variables. This analysis revealed a main effect for asking a question on boastfulness, F(1, 101) = 8.03, p < 0.01, η ² = 0.07. Avi was perceived as less boastful when he answered a question than when he did not (see Table 1). Although there was no significant main effect for issue raising, F(1, 101) = 0.33, p > 0.1, η ² = 0.003, as expected there was a significant interaction between asking a question and issue raising, F(1, 101) = 4.24, p < 0.05, η ² = 0.04. In order to better understand the nature of the interaction, two simple ANOVAs were conducted separately for the conditions in which the interactant asked a question and for when he did not. When the interactant asked a question, there was no difference in the perception of boastfulness between the condition in which the target person raised the issue and the condition in which the interactant raised the issue, F(1, 51) = 1.04, p > 0.1, η ² = 0.02. However, as expected, when there was no question, raising the issue had a marginally significant impact. The target person was perceived as more boastful when he himself raised the issue than when his interactant raised the issue, F(1, 50) = 3.7, p = 0.06, η ² = 0.07.

Table 1. Impact of issue raising and question asking on perceptions of boastfulness and liking for the target person

	Boastfulness	Liking
Target person raised the issue
Question	2.70 (0.88)	3.13 (0.48)
No question	3.54 (0.79)	2.70 (0.54)
Total	3.13 (0.93)	2.92 (0.55)
Interactant raised the issue
Question	2.96 (0.94)	3.06 (0.96)
No question	3.09 (0.91)	3.04 (0.56)
Total	3.02 (0.92)	3.05 (0.78)
Total
Question	2.83 (0.91)	3.10 (0.75)
No question	3.33 (0.87)	2.87 (0.57)
Total	3.08 (0.92)	2.98 (0.68)
Table contains means on a scale of 1–5 and standard deviation (in parentheses).

Looking at the same interaction from a different perspective reveals that when the interactant raised the issue there was no difference in the perception of boastfulness between the condition in which the interactant asked a question and when he did not, F(1, 50) = 0.27, p > 0.1, η ² = 0.01. However, when the target person raised the issue, asking a question had a significant impact. The target person was perceived as more boastful when he was not asked a question than when he was, F(1, 51) = 13.40, p < 0.001, η ² = 0.21.

A similar pattern of results emerged with regard to the participants’ liking for Avi. This analysis revealed a marginal main effect for asking a question, F(1, 100) = 2.97, p = 0.09, η ² = 0.03. Avi was liked more by the participants when he answered a question than when he did not. There was no significant main effect for issue raising, F(1, 100) = 0.99, p > 0.1, η ² = 0.01 and there was no significant interaction between asking a question and issue raising, F(1, 100) = 2.52, p = 0.12, η ² = 0.03.

Although the interaction was not significant, the means showed the same pattern as in the boastfulness index. When the interactant asked a question, attitudes toward the target person were not affected by whether the target person raised the issue or the interactant raised the issue, F(1, 50) = 1.3, p > 0.1, η ² = 0.003. However, as expected, when there was no question, raising the issue had a significant impact. The target person was liked less when he himself raised the issue than when his interactant did, F(1, 50) = 4.86, p < 0.05, η ² = 0.09. Looking at the same interaction from a different perspective reveals that when the interactant raised the issue, there was no difference in liking for the target person between the condition in which the interactant asked a question and when he did not, F(1, 49) = 0.01, p > 0.1, η ² = 0.001. However, when the target person raised the issue, asking a question had a significant impact. The target person was liked less when he was not asked a question than when he was asked, F(1, 51) = 9.36, p < 0.01, η ² = 0.16. No significant results were obtained for the other two dependent variables.1

Discussion

Overall, the results of the current study provide partial support for the research hypotheses. In line with Holtgraves and Srull (1989), the self-promoter was perceived as less boastful and more likable when he answered a question than when he was not responding to a question. More importantly, asking a question had an effect only when the target person raised the issue. When the interactant raised the issue, the target person was perceived relatively positively, regardless of being asked a question or not. Boasting after the raising of the issue by the interactant probably conforms to the modesty norm. The raising of the issue might be perceived as equivalent to asking a question, as both of them intend to encourage a response from the fellow interactant.

Looking at the interaction from a different perspective revealed that, as expected, raising the issue had an effect on the perception of the self-promoter only when he was not asked a question. Raising the issue did not harm the perception of the self-promoters who succeeded in encouraging their interactants to ask them a question regarding their accomplishments. When that question was asked, the self-promoter was perceived similarly to when he raised the issue himself and when the issue was raised by his interactant.

Self-aggrandizement in response to a question presumably conforms to the norm of modesty. According to the mindlessness hypothesis (Langer, 1992), because this behavior does not violate the perceivers’ expectations, the perceivers process it mindlessly. Thus the perceivers do not become engrossed in the details of the conversation, and they disregard the context that preceded the question. The question that is asked activates a scheme for a polite conversation in which the description of success is initiated by the fellow interactant. Presumably the memory of the conversation is reconstructed in a way that erases the role of the self-promoter in promoting the question, and the initiator seems to be his/her interactant.

However, when there is no question, the self-promotion is presumably perceived as violating the modesty norm. According to the mindlessness hypothesis, people begin to pay attention to the details of a message only when they detect a mismatch between that message and a common script. In this case, then, they will focus on who raised the issue and presumably recognize the raising of the issue by the self-promoter as a manipulative attempt to create the right context for the discloser's self-praise. Thus they attribute more negative characteristics to the self-promoter who raised the issue than to the self-promoter who boasted after the issue was raised by his/her interactant.

While these explanations might seem logical, Study 1 does not provide any evidence attesting that this is indeed the process underlying the results. The same results might also be explained differently. For example, a person who answers a question might be perceived as polite and thus might be forgiven for raising the issue. However, a person who raises the issue, is ignored by his/her interactant (apparently because he/she does not want to talk about it), and boasts without being asked might be perceived as particularly rude. If the explanation based on the mindlessness hypothesis and on retrospective memory is correct, people might tend to forget that the self-promoter raised the issue when the boasting was a response to a question. This hypothesis was examined in Study 2.

Study 2

The aim of this study was to examine whether the impressions formed of the self-promoter indeed matched the recall pattern of the conversation. Study 1 showed that the self-promoter was perceived as boastful and was liked less when he raised the topic of the conversation and mentioned his high grade without being asked about it. However, raising the topic did not affect the perception negatively if the person was later asked about his grade. It is thus expected that, in the first condition, people will tend to remember that the self-promoter initiated the topic of the conversation. However, in the latter condition they will tend to think that the interactant raised the topic.

In this study the participants were asked to read the same conversation as in Study 1, only this time they were later asked to recall it. The participants were not asked to report impressions that they had in order to prevent this reported impression from influencing the recall.

Method

Participants

The participants were 61 undergraduate students of a university in Israel, of whom 68% were female, 30% were male, and 1 did not indicate his or her gender. Their mean age was 22.78 (SD = 2.05). The participants volunteered to participate in the experiment.

Materials and procedure

Participants were randomly assigned to the same four experimental conditions as in Study 1. In all conditions participants received a paper that included an introduction, demographic questions, and one version of a conversation from Study 1. The participants were told that this was a research project about interpersonal communication and were asked to fill in their demographic details and then read the conversation.2 In order to create a situation comparable to that of Study 1, they were not told that they would be asked to recall the conversation later on. After the participants had finished reading, five research assistants collected their papers and handed them new papers on which the first line of the conversation appeared, and they were asked to recall the rest of it and write it down. The condition to which each participant belonged was indicated by a small letter at the bottom of the paper, and the research assistants gave each participant a new paper with this same letter.

Each recalled conversation was encoded by two independent judges. The judges had to answer the following questions: (1) Was the grade (A+) mentioned?, (2) Was there a question about the grade before it was mentioned?, and (3) Who raised the topic of the conversation; that is, who started talking about the exam's grades? The Cohen's Kappas that were computed were all greater than 0.8.

Results

Three chi-square analyses were conducted with the four conditions as the independent variable and the three questions described above as the dependent variables. There was no significant difference between the four conditions about recalling the grade (A+). In fact, all participants included this information in their recalled conversation. The conditions did differ significantly in terms of remembering a question about the grade before it was mentioned, χ²(3, N = 61) = 8.54, p < 0.05. A series of chi-square analyses showed that the participants tended to recall a question more in the two conditions in which there was indeed a question (p < 0.05). However, there was no significant difference between the condition in which the self-promoter raised the issue and the friend asked the question, and the condition in which the friend raised the issue and did not ask a question (p < 0.05). Lastly, the conditions differed significantly in recalling who raised the topic of the conversation, χ²(3, N = 61) = 9.39, p < 0.05. A series of chi-square analyses showed that the only condition that differed from the rest was that in which the self-promoter raised the topic and was not asked a question (p < 0.05). This was the only condition in which most participants recalled that the topic was raised by the self-promoter. In all other conditions the participants tended to recall that the topic was raised by the interactant (see Table 2).

Table 2. The differences in recall between the four conditions

	Who raised the issue?		Was there a question?
The condition of the questionnaire	The self-promoter	The friend	Yes	No
The friend raised the issue and asked a question	4	11	14	1
The self-promoter raised the issue and the friend asked a question	6	9	13	2
The friend raised the issue and did not ask a question	7	9	10	6
The self-promoter raised the issue and the friend did not ask a question	12	3	8	7

Discussion

The recall pattern conforms to the hypothesis regarding the process underlying the results of Study 1. The participants in the various conditions did not differ in their recall of the issue that repeated in all types of conversations, namely the mentioning of the specific grade. Moreover, the participants remembered quite well if a question was asked pertaining to the grade. Interestingly, however, there was no significant difference between the condition in which the self-promoter raised the issue and the friend asked the question and the condition in which the friend raised the issue and did not ask a question. As argued in Study 1, the raising of the issue might be perceived as equivalent to asking a question, as both of them intend to encourage a response from the fellow interactant.

Most interesting, however, is the last result. As expected, when the self-promoter raised the topic and was asked about his grade, the participants tended to recall that the interactant raised the topic. Results here were the same as when the issue was indeed raised by the interactant. Only when the self-promoter raised the topic and was not asked about it did the participants tend to recall that the issue was raised by the self-promoter himself.

Thus, in accordance with the mindlessness hypothesis (Langer et al., 1978) and models of retrospective memory (e.g., Hirt, 1990), the participants focused their attention on the details of the conversation only when the message did not conform to their expectation based on the norm of politeness; that is, when the boasting did not follow a question. As hypothesized, when a question was asked it erased the context that preceded it, and the recalled context matched the polite scheme in which successful people reveal their success only after being asked about it.

General Discussion

Abundant research has documented the various impression management tactics (e.g., Leary, 1995), while much less attention has been devoted to their perception (e.g., Vonk, 1999). Research that did examine perceptions of impression management showed that perceivers pay attention to the social context in an attempt to understand the actors’ motives. For example, self-promotion was seen as less boastful when it was a response to a question (Holtgraves & Srull, 1989).

According to the above-mentioned research, perceivers are knowledgeable about impression management and use their knowledge when making judgments about impression managers. However, it is argued here that impression managers are also knowledgeable about the way they are perceived. They know that the social context can help them self-promote without being perceived as braggers. Thus they work actively to create the right context for their self-promotion (Holtgraves, 1990). The present research examined the perception formed of these self-promoters who actively work on creating the right conversational context for their self-praise.

While this active manipulation might be seen as a violation of the norm of politeness, this deviation seems to be forgiven when the speakers succeed in making their interactant ask them the awaited question. This finding is consistent with and provides support for the mindlessness hypothesis (Langer et al., 1978) and models of retrospective memory (e.g., Hirt, 1990). According to the former, people focus their attention on the details of the conversation only when the message does not conform to their expectation. A question that is asked is consistent with the norm of politeness and thus does not initiate intense processing. This same question activates a scheme for polite conversation in which people talk about their success only after being asked about it.3 This scheme prompts a reconstructed memory of the conversation in which the issue was not raised by the target person. This hypothesized process was indeed documented in the second study, which confirmed that the question that was asked tended to erase the memory of the fact that the self-promoter was the person who raised the topic of the conversation.

These findings contribute to the growing research on the effectiveness of impression management tactics generally and self-promotion specifically. Earlier research has focused mainly on the actual wording that accompanies self-promotion that impacts the resultant impression. Self-promotion was perceived more positively when combined with ingratiation (Godfrey, Jones, & Lord, 1986), with more negative portrayals of one's self (Robinson, Johnson, & Shields, 1995) or with apologizing (Levine & West, 1976). Unlike these studies, the current research focused on an impression management tactic that allows pure bragging, unaccompanied by moderating phrases, which does not lead to a negative impression but the success of which is dependent on the responsiveness of the interactant who creates the right context for the boast.

The current research may also contribute to attribution theories. It is widely documented that people tend to attribute the behavior of others to their internal attributes more than to the external context that was imposed on them, i.e., fall prey to the correspondence bias (e.g., Jones, 1979). However, previous research has shown that receiving information that suggests an ulterior motive leads to a suspension of judgment and avoidance of the correspondence bias (e.g., Fein, 1996). In that case people who ingratiate are not perceived as kind and helpful but as engaging in impression management. The current research showed that there are specific contextual cues, such as questions, that also reduce the tendency to make the correspondence inference, but by the same token affect perception in the opposite way. In other words, they allow self-promoters to be perceived more positively and not as impression managers.

The speakers, whose internal attributes make them brag and praise themselves, are perceived relatively positively if they succeed in creating the right context for their boasts. This context is so perfect that it seems to erase the manipulation that preceded it and thus is perceived as a legitimate, external context imposed on the self-promoters that excuses their boasting.

The current research suggests several new avenues for future research. First, additional support for the argument of the current research might be obtained by measuring the response time in the various conditions of the impression formation and the recall tasks. Such a study might provide additional indications that people allocate different amounts of cognitive processing for the norm-consistent versus the norm-violation conditions. Similarly, future studies might use a more complex stimulus that will enable the measurement of free recall level, as was done in previous research (e.g., Hastie, 1980; Hastie & Park, 1986). This measurement will be more precise than the classification used in Study 2 of the current research. Moreover, while the current research examined judgment and recall separately, future research might investigate them both in the same experiment and seek to determine whether differences in judgment are indeed mediated by differences in recall.

Future research might also examine how the manner in which self-promoters answer their interactant's question affects the way they are perceived. For example, a very quick, a lengthy, or an exaggerated response might hint that the self-promoters were just waiting for the question. Behaving in such ways might increase the chance of the tactic being detected and the self-promoters being perceived negatively. It might also be interesting to examine how the other interactant is perceived in the various conditions of the current research. For example, in the condition in which the self-promoter raised the issue about the grades and his interactant did not ask a question, the interactant might be perceived as trying to avoid revealing his poor grade.

Future research might also explore how the self-promoters who use the tactic presented in the current research are perceived by their interactants. Previous research has shown that the perception of the impression managers can differ depending on whether one is an outside observer or a participant in the interaction (e.g., Vonk, 2002). In our case, in line with the actor–observer effect (Jones & Nisbett, 1971), the interactant who asked the self-promoter a question might be more aware of the circumstances that made him ask this question. Thus, it might be hypothesized that these tactics would be less successful in making a good impression on the interactant than they would be on an outside observer. On the other hand, as is suggested by the perceiver-induced constraint (Gilbert & Jones, 1986), interactants tend to ignore the conversational context even if they created it themselves. Moreover, interactants tend to be more cognitively busy than outside observers (e.g., Vonk, 2002). Thus interactants might not be aware of the context that stimulated them to ask the question. Moreover, as people commonly want to believe that they are not easily manipulated (e.g., Perloff, 1999), interactants might refuse to admit that they were led by their conversation partner to ask him or her a question. Lastly, past research has documented a “positivity bias” in participants’ evaluation of their fellow interactants compared with the evaluation of outside observers (e.g., Street, Mulac, & Wiemann, 1988). Thus the tactics we are referring to might be even more successful in making a good impression on the interactant than on outside observers.

It would also be interesting to explore the perception of equal disclosure by the two interactants. Previous research has shown that people tend to match their self-presentation to their interactant. If one of the interactants is engaging in self-promotion, the other interactant also tends to self-promote (Baumeister, Hutton, & Tice, 1989). According to Holtgraves and Srull (1989), when the other interactant responds with an equally positive self-disclosure, the target person who self-promoted is perceived more positively. It would be interesting to examine the effect of the order of self-promotion. Who will be perceived more positively, the person who self-promoted first or the one who responded to him or her? Similarly, it would be interesting to examine how a self-promoter is perceived in a case in which, following his/her boasting, the self-promoter asks his or her interactant for their grade. While this question might be perceived as polite and conforming to the reciprocity norm, it might also be perceived as an attempt to make a downward social comparison.4

Besides its theoretical contribution, the current study has clear practical implications for daily interpersonal communications. There are many circumstances in which the only way people can make others aware of their accomplishments is to mention them. In situations ranging from a first date to a job interview, people commonly face the dilemma of how to make their listeners aware of their success without being perceived as braggers. The present research provides a possible solution to this dilemma. According to the findings, our best course of action is to encourage our interactants to ask us a question pertaining to our success. This question, it seems, is so powerful that it erases the manipulation that preceded it and creates the perfect context for bragging—a context that enables the impression manager to self-promote without being perceived as boasting.

Notes

¹ No significant main effect or interactions were revealed for gender (p > 0.05). However, an interaction approaching significance was revealed between gender and asking a question on the boastfulness index, F(1, 95) = 2.17, p = 0.15, η ² = 0.02. In line with previous research, which showed a preference of females for modest self-presentation (e.g., Brown, Uebelacker, & Heatherington, 1998) and polite language (e.g., Lakoff, 1975), women perceived the self-promoters who answered a question as less boastful than the self-promoters who did not answer a question, F(1, 74) = 10.85, p < 0.01, η ² = 0.13, while men did not show any difference in their evaluation of these two groups, F(1, 25) = 0.16, p > 0.1, η ² = 0.006.

² The demographic data were obtained from the first paper and relate to 63 participants who filled out this page. Two participants were dropped in the second phase of the experiment.

³ In order to support the argument regarding the violation of the norm of politeness, a pilot test was conducted with a sample from a population similar to that of the two main studies (n = 48). Participants were randomly assigned to the same four experimental conditions as in Studies 1 and 2 but the dependent variable this time concerned norm violation. The participants were asked to answer two questions on a 1–7 scale following the reading of the scenario: To what degree is what Avi is saying considered polite? To what extent does what Avi is saying fit or violate your expectation of this kind of conversation? These two questions formed an index with the larger numbers representing greater politeness (M = 4.84, SD = 1.28, Cronbach's α = 0.75). As expected, there was a significant difference in this politeness index between the four conditions, F(3, 44) = 2.90, p < 0.05, η ² = 0.17. A contrast analysis showed that the condition in which the self-promoter raised the issue and was not asked a question (M = 4.13, SD = 1.46) was rated as significantly less polite than the other conditions (friend raises and asks M = 5.40, SD = 0.91; friend raises no question M = 4.86, SD = 0.87; the self-promoter raises and the friend asks M = 5.25, SD = 1.32), t(44) = 2.76, p < 0.01.

⁴ The author would like to thank the two anonymous reviewers for contributing some of the ideas for future research.