ATTITUDES AND SOCIAL COGNITION

Automatic for the People: How Representations of Significant Others Implicitly Affect Goal Pursuit James Shah University of Wisconsin—Madison Five studies are presented that explore how representations of significant others may automatically affect goal pursuit. Specifically, evidence is presented that suggests goals may be primed by one’s representation of a significant other and that this priming may be moderated by one’s closeness to this other individual. It is also shown to be affected by the number of different goals associated with this person. The greater the number of goals associated with a significant other, the less likely this individual will invoke any 1 goal very strongly. Such goal priming is shown to have implications for the extent to which goals are pursued (as seen through task persistence and performance) as well as the extent to which they are inhibited or ignored (especially when an individual is associated with a goal unrelated to a current pursuit).

they are to be pursued in the moment or over a lifetime. Although the external prescriptions of others may or may not correspond at any given moment with our own internally derived pursuits, they may nevertheless come to influence our everyday behavior and subjective experiences (e.g., Moretti & Higgins, 1999b). Our mental construal of others, then, may impact our everyday behavior through their close association to various goals. Moreover, as with other cognitive associations, such associative effects may occur quite spontaneously. The mere activation of one’s internal representation of a close other may be enough to invoke the goals with which this individual is associated. The present article seeks to address the scope of such interpersonal priming effects and the degree to which they may unfold automatically by considering how significant others may implicitly influence goal accessibility, goal commitment, and goal pursuit.

Since Triplett (1898) first examined “dynamogenic factors” in competitive cycling, social psychology has been interested in how others may affect our own pursuits and strivings. It has often been assumed that the scope of such social influence extends to the “psychological presence” of significant others in that we are influenced not only by the actual company of these individuals, but by how we mentally represent them, whether they be parents, siblings, friends, colleagues, or even general “reference groups” (see, e.g., Bowlby, 1969; Kelley, 1952; Sherif, 1948). Indeed, there has been a longstanding focus in psychology on how such internal representations may come to influence our sense of self and efforts at self-regulation, whether it be, for example, through role taking (Mead, 1934), reflected self-appraisals (Cooley, 1902/1964), identification (A. Freud, 1937), or the process of internalization (Schaffer, 1968). Yet, surprisingly, there has been relatively little empirical exploration of how such “inner audiences” (see Horney, 1946; Moretti & Higgins, 1999b) are invoked and the cognitive mechanisms that may underlie their self-regulatory influence (but see Andersen & Glassman, 1996). Such influence may be particularly noteworthy in light of evidence suggesting that our self-representations often incorporate close significant others as well as the goals, values, and expectations that these others may have for us, whether

Goals as Knowledge Structures Recent research on the role of goals in self-regulation has increasingly recognized that goals, like attitudes, have many of the same “cognitive” features that characterize knowledge structures more generally (Bargh, 1990; Kruglanski, 1996; Shah & Kruglanski, 2000). Of particular interest has been goals’ accessibility, or how readily they come to mind in different situations or with different individuals (Chartrand & Bargh, 1996; Kruglanski, 1996; Shah, Kruglanski, & Friedman, 2002). Indeed, recent work has suggested that this cognitive property of goals significantly moderates the manner in which they direct behavior and the extent to which they influence emotional experience (see Austin & Vancouver, 1996; Carver & Scheier, 1998; Gollwitzer & Moskovitz, 1996; Higgins, 1997; Higgins, Shah, & Friedman, 1997; Shah, Higgins, & Friedman, 1998). Consistent with their status as mental representations whose accessibility can vary chronically and situationally (see Higgins,

The research reported in this article was supported, in part, by National Institute of Mental Health Grant MH63280-02 and by a grant from the University of Wisconsin—Madison Graduate School Research Committee. I thank my dedicated laboratory assistants for their invaluable aid in conducting this research, especially Alexa Daniels and Paige Brazy. Correspondence concerning this article should be addressed to James Shah, Department of Psychology, University of Wisconsin—Madison, 1202 West Johnson Street, Madison, Wisconsin 53706. E-mail: [email protected]

Journal of Personality and Social Psychology, 2003, Vol. 84, No. 4, 661– 681 Copyright 2003 by the American Psychological Association, Inc. 0022-3514/03/$12.00 DOI: 10.1037/0022-3514.84.4.661

661

SHAH

662

1996), a growing body of research has found compelling evidence that goals can be unconsciously activated by the environmental context in which they are pursued and that such activation can have significant self-regulatory implications. Aarts and Dijksterhuis (2000), for example, have recently argued that habits may be mentally represented as goal-action links whose connections are so strong that activation of the goal automatically invokes the related action. Such a conceptualization is consistent with recent work suggesting that goals can be primed by environmental cues and that their activation has significant consequences for both evaluation and behavior. Chartrand and Bargh (1996) have found that participants’ processing and recall of information about another participant was affected by the presentation of words intended to invoke an impression formation goal (e.g., evaluate or judgment). Similarly, Bargh and Barndollar (1996) reported finding that when an achievement or affiliation goal was primed through semantically related words such as strive or sociable, such priming influenced the social behavior participants demonstrated toward a teammate. Priming an affiliation goal lessened participants’ tendency to humiliate an underachieving teammate by far exceeding his or her performance. Priming an achievement goal, however, increased participants’ tendency to strive for the highest possible score regardless of their teammate’s performance (see also, Bargh, 1996, 1997). Goal priming has also been shown to influence how one experiences specific goal pursuits. Chartrand (2001), for instance, has recently demonstrated that unconsciously activated achievement goals may affect how one feels emotionally about easy and difficult anagram tasks. Participants primed with an achievement goal reported feeling relatively more positive when given an easy anagram task that would allow them to succeed and relatively more negative when given a difficult anagram task in comparison with participants in a control condition. The existence of such priming effects and their demonstrated consequences for self-regulation only increases the need to understand how such goal priming may arise in real-world settings. As mentioned earlier, one possibility of particular relevance to social psychology would be that such priming occurs through our representations of significant others, especially because these representations have been so closely linked to the concept of the self and the process of self-regulation (see, e.g., Andersen, Glassman, & Gold, 1998; Markus & Cross, 1990; Markus & Kitayama, 1991).

Implicit Effects of Significant Others Psychology has certainly long noted the psychological relevance of our interpersonal construals. Building on earlier psychodynamic theorizing (S. Freud, 1912/1958), Sullivan (1940, 1953), for example, posited that former relationships may “parataxically distort” one’s present social interactions when the feelings and experiences associated with these relationships are displaced onto one’s present interactions and social partners. More recent research on the phenomena of transference has offered ample evidence that the qualities, emotions, and motivations associated with close significant others may frequently be conferred on ourselves and on those around us (see Andersen & Berk, 1998; Andersen, Reznik, & Chen, 1997). Moreover, as originally noted by S. Freud (1912/ 1958), such transference may often occur quite automatically, requiring little, if any, conscious intent or awareness. Indeed, Glassman and Andersen (1999) recently found that individuals

subliminally primed with a description of a significant other were more likely to ascribe significant-other characteristics to a “game partner,” even when these characteristics went beyond what was actually primed (see also Andersen, Glassman, Chen, & Cole, 1995). In addition to influencing how we perceive others, activating representations of significant others may also affect how we act toward (or against) them. Andersen, Reznik, and Manzella (1996), for example, found that participants were more motivated to emotionally approach a target person in an experimental setting who most resembled a positive significant other and were more motivated to avoid a target person that resembled a negative other (in comparison with a yoked participants’ positive and negative significant other). Finally, significant others may also quite automatically affect how we come to see ourselves, as famously illustrated by Baldwin, Carrell, and Lopez (1990). These researchers found that priming psychology graduate students with a scowling picture of their department chair adversely affected the students’ assessment of their own research ideas (in comparison with students primed with the approving face of an unknown other). Similarly, practicing Catholic students were found to rate themselves more harshly when first subliminally presented with a scowling picture of the Pope than when presented with the scowling picture of an unknown other. Of interest, this effect was not found for nonpracticing Catholics whose identification with the Pope was presumably weaker (see also Baldwin, 1992, 1994). Moreover, significant others may not only affect how positive or negative we feel about ourselves, but the specific contents of our working self-concept. Hinkley and Andersen (1996), for example, found that after being exposed to a target person who resembled a significant other, participants’ working self-concept would more likely resemble their typical sense of self when with this significant other.

Goal Priming and Inhibition Through Significant Others The research described above offers compelling evidence that our complex representations of significant others may come to influence not only how we perceive others, and interact with them, but how we come to perceive and evaluate ourselves. These representations may also come to influence what we choose to desire and to pursue (see Kruglanski, 1996). Indeed, models of intrinsic motivation have long recognized that although everyday goals, such as those involving one’s grades or athletic pursuits, may often arise from “internal” sources (e.g., involving needs for autonomy, competence, and relatedness), they may also result from external pressures, such as those originating from parents, friends, or authority figures (see, e.g., Deci & Ryan, 2000; Ryan & Deci, 2000). Similarly, self-discrepancy theory (see Higgins, 1999) assumes that in addition to pursuing our own ideals and obligations, we often pursue the ideals and obligations that significant others have for us. Moretti and Higgins (1999a), for example, demonstrated that the pursuit of these “external” goals may significantly influence psychological well-being and interpersonal functioning when these goals match one’s own “internal” ideals and obligations. Although these results suggest that significant others may influence our goal pursuits as we gradually come to internalize the goals that others have for us, the present work seeks to examine whether the effects of significant others may be more spontaneous: specifically, whether representations of important

GOAL PRIMING THROUGH OTHERS

individuals may automatically trigger goal pursuit by increasing the salience of the goals that these individuals have for us. Moreover, just as significant others may automatically direct us toward goals, so might they also automatically direct us away from certain goals, either because they view them as relatively unworthy of pursuit (in comparison with what we could generally be pursuing instead) or because they have other goals in mind for us. Indeed, because individuals can potentially be pursuing any number of different goals (for a discussion on goal systems, see Shah & Kruglanski, 2000), and because their time and resources are limited (see, e.g., Baumeister, Bratslavsky, Muraven, & Tice, 1998), an important challenge of effective self-regulation may be to disregard the “unworthy” goals that could potentially be pursued (see Gollwitzer & Bayer, 1999; Kruglanski, 1996; Kuhl & Beckman, 1994). Another way that significant others may affect goal pursuit, then, is by automatically inhibiting those goals that may safely be ignored so that one may conserve their time and effort for the pursuit of relatively more worthy aims. Indeed, Shah, Friedman, and Kruglanski (2002) have recently found that the pursuit of a goal, whether it be a long-standing ambition or situational concern, is independently aided by both its own activation and the inhibition of alternative goals.

Moderating Factors Certainly not all our representations, however, have quite the same impact on our motivations and goal pursuits. Different individuals may bring to mind different types of goals and may do so with differing levels of effectiveness. Thus, although our mothers may generally invoke goals relating to our careers or family, some mothers may not invoke such goals so readily. A number of factors may moderate the type of goal priming effects described above. First, although one may generally incorporate the goals, beliefs, and expectancies of significant others, one’s sense of self may be more readily influenced by those individuals with whom one is particularly close (see Aron, Aron, Tudor, & Nelson, 1991; Brewer & Gardner, 1996; Read & Miller, 1989). We may, of course, place greater value in goals that close significant others want us to pursue, but interpersonal closeness may also affect how readily these goals are brought to mind, either by increasing how strongly we associate ourselves to this other individual (i.e., the self– other association) or how strongly we associate our significant other to the goals he or she has for us (other– goal association). Aron et al. (1991), for example, illustrated the cognitive significance of being in a close relationship in terms of how our representation of this other individual may come to “overlap” with our own sense of self, and how such overlap may signify stronger self– other associations. Closeness to a significant other is also often characterized by a stronger emotional bond, and more positive and frequent experiences and interactions with this individual that may also foster a greater awareness, and acceptance, of the goals that this individual has for us, increasing how strongly one associates this significant other with specific goals (i.e., other– goal associations). Although the present studies focus on the bond one feels toward a close significant other (rather than one’s perceived similarity to this individual), increasing either type of association (self– other or other– goal) may facilitate the cognitive activation of goals through significant others.

663

A second moderating factor may be the motivational complexity of our significant-other representations, as signified by the number of different goals with which each is associated. Certainly, for instance, parents may differ in the range of goals they have for their child. And although a greater number of goals may suggest a wider range of parental influence, such breadth may actually limit the degree to which a parent invokes any particular goal very strongly. Such a possibility is reminiscent of the classic “fan effect” (Anderson, 1974, 1983; Anderson & Reder, 1999; Radvansky, 1999) in which the greater the number of specific facts linked to a general mental construct (i.e., the greater the general constructs’ “fan size”), the less likely any particular fact will be retrieved or recalled. Thomsen, Lavine, and Kounios (1996), for instance, have recently demonstrated that individuals’ general values and more specific attitudes more readily prime each other when they each had relatively few other cognitive associates. Moreover, even when parents strongly invoke multiple goals in their child, these goals may end up competing with each other for “regulatory resources” (i.e., attention and effort). These competing goals may inhibit each other, limiting the degree to which any would be acted upon. Thus, whether through response competition or through spreading activation, the motivational complexity of our significant-other representations may limit the degree to which they invoke any particular goal, all other factors being equal.

Theoretical Summary Three hypotheses are proposed, then, regarding how goals may be automatically invoked by others: Because significant others are so intimately tied to our own self-concept and efforts at selfregulation, invoking our mental representations of these individuals should increase the salience of the goals to which they are closely associated. Moreover, such goal priming should be more likely in close relationships and less likely when representations of significant others are “motivationally complex” (i.e., linked to numerous and potentially competing goals). Finally, if such priming occurs, it should have implications for the activation and maintenance of goal pursuits and, in some instances, for their inhibition and deactivation. Five studies are presented to explore these hypotheses. These studies examine both self-set and experimentally defined goals, use a variety of priming techniques, and examine a range of self-regulatory consequences.

Study 1 Study 1 sought to examine whether representations of significant others may prime intentions in participants and whether this priming may vary as a function of how close participants feel to the significant others in question. More precisely, participants were asked to provide the name of their mother and a friend, and a goal that each of these significant others had for them. Participants were also asked to indicate how close they were to these individuals. Later, participants were asked to indicate how committed they were to pursuing these goals in the upcoming week while being subliminally primed with either the name of their friend, their mother, or a control word.

SHAH

664 Method Participants

Fifty-seven University of Wisconsin—Madison undergraduate psychology students (36 males and 21 females) participated in return for course credits. Participants were between the ages of 18 and 25. The gender of the participants did not significantly influence the presented results either singularly or interactively and thus will not be discussed further.

Procedure Participants completed the experimental procedure on personal computers. Each participant was first asked to provide the first name of their mother and the first (and different) name of a friend. Participants were also asked to provide an attribute that represented a goal their mother had for them (e.g., educated, social, and attractive), and a different attribute that represented a goal their friend had for them. Participants then completed a series of ratings that included their separate estimations of how close they were to their mother and to their friend. Specifically, participants were asked to indicate how close they were to their mother and to their friend, and how close they thought their mother and friend were to them. Each of these ratings was provided on a 10-point scale ranging from 1 (not at all) to 10 (extremely). Participants then completed a separate filler task designed to lessen the salience of the previously described questions.1 Finally, after completing this filler task (which required approximately 15 min), participants were asked to indicate how committed they were in the upcoming week to attaining or possessing the attributes they had listed previously. Specifically, participants rated how focused they currently were on attaining each attribute listed previously (i.e., their “mother goal” and “friend goal”) on a 7-point scale ranging from 1 (not at all) to 7 (extremely). They also rated how frequently they would try to pursue each of these attribute goals in the next week from 1 (not at all) to 7 (all the time). Before answering each of these questions, the computer program primed the participants with either the name of their mother, the first name of their friend, or a control word (space). The type of prime was randomly varied between participants. Each prime word was presented for 50 ms in the center of the computer screen and was backward masked by the presentation of the subsequent question.2 Closeness to friend and mother. Not surprisingly, participants’ ratings of their closeness to their friend were highly correlated with each other (r ⫽ .96). These ratings were summed to form a closeness-to-friend score. Similarly, participants’ closeness ratings regarding their mother were also highly correlated (r ⫽ .94), and they were also summed to form a closeness-to-mother score. Commitment to mother’s goal and friend’s goal. Participants’ overall commitment to the goals that their mother and friend were perceived to have for them was calculated by summing the focus and frequency-ofpursuing rating for each goal. These two ratings were also found to be highly correlated with each other (r ⫽ .71 for friend’s goal; r ⫽ .67 for mother’s goal).

Results Regression analyses examined the effect of priming and participants’ closeness to significant others on their self-reported commitment ratings for the goals that their significant others wanted them to pursue. Separate analyses compared participants’ intended commitment to the goal that their mother wanted them to pursue when they were primed with the name of their mother versus when they were primed with a control word or when primed with the name of their friend. These analyses included the priming manipulation, participants’ closeness to their mother, and the interaction of this closeness variable with the priming manipulation. These analyses found that participants’ commitment was significantly

greater when the name of participants’ mother was primed versus a control word or the name of a friend—B ⫽ 1.19, F(1, 34) ⫽ 9.31, p ⬍ .01; B ⫽ 0.93, F(1, 34) ⫽ 9.03, p ⬍ .01, respectively. Moreover, these analyses also revealed that participants’ closeness to their mother had a significant effect on their commitment to their mother’s goal—B ⫽ 1.08, F(1, 34) ⫽ 4.92, p ⬍ .05 when compared with the control prime condition; B ⫽ 0.73, F(1, 34) ⫽ 4.28, p ⬍ .05 when compared to the friend prime condition—and, perhaps more importantly, this closeness significantly moderated the previously mentioned priming effect—B ⫽ 1.17, F(1, 34) ⫽ 5.12, p ⫽ .01 when compared to the control prime condition; B ⫽ 0.83, F(1, 34) ⫽ 5.28, p ⬍ .01 when compared with the friend prime condition. The direction of these interactions suggested that the greater participants’ closeness to their mother, the more likely that priming her name will increase participants’ commitment to their mother’s goal. A separate set of regression analyses examined participants’ intended commitment to the goal that their friend had for them in the upcoming week when participants were primed with the name of their friend versus when they were primed with a control word or the name of their mother. These analyses included the priming 1 Because participants identified their significant other in some way in most of the presented studies, a delay was included in the design of these studies to lessen any memory enhancement in advance of the primes (for a discussion of this issue, see Greenwald & Banaji, 1989). It should also be noted that participants did not identify their father in advance of the priming procedure in Study 2 and that Fitzsimmons and Bargh (2003) have reported similar goal priming effects with experimental designs that gather significant-other information weeks in advance of the priming procedure. 2 Working under the assumption that relatively longer exposure times might increase the strength of priming effects (see Greenwald, 1992), the priming procedure used in Studies 1, 3, 4, and 5 presented primes for 50 ms. Given the length of these exposure times, the astute reader may wonder whether the primes in these studies were truly subliminal. Indeed, although Perdue, Dovidio, Gurtman, and Tyler (1990) have used a similar prime exposure time (55 ms), Bargh and Chartrand (2000) have noted that care needs to be taken in assessing whether relatively longer foveal prime exposures are indeed subliminal. As they note, whether a prime is subliminal may depend on specific characteristics of the priming procedure, such as the relative brightness of the primes, the distance of the computer display from the participant, and the manner in which the primes are masked. With this in mind, two steps recommended by Bargh and Chartrand were taken to ensure that participants could not consciously recognize the presented primes in these specific studies. A funneled debriefing procedure was used in each of the presented studies that (a) first probed for general awareness of the priming manipulation and the primes and (b) then told participants of the existence of this priming manipulation in order to assess whether they could identify any of the prime words used in the procedure (see, e.g., Chartrand & Bargh, 1996). None of the participants in these studies reported seeing the presented primes when initially probed, nor could they identify any of the prime words even after being told of their existence. Moreover, an additional set of 39 participants were given the identical priming procedures used in these studies after first being told of the existence of the primes and instructed to name the prime word after each trial. None of these participants were able to correctly identify any of the prime words presented to them. Finally, it should also be noted that Study 2 found consistent priming effects while using a priming procedure that exposed participants to primes for only 10 ms. Moreover, many of the priming effects described in the present article have been replicated elsewhere using this quicker (10 ms) prime exposure time (see Shah, 2002).

GOAL PRIMING THROUGH OTHERS

665

Figure 1. Goal commitment as a function of priming condition, the importance of the task goal to participants’ significant other, and participants’ closeness to their significant other.

manipulation, participants’ closeness to their friend, and the interaction of this closeness variable with the priming manipulation. These analyses revealed that participants’ commitment to their friends’ goal significantly increased when participants were primed with the name of their friend—B ⫽ 1.24, F(1, 34) ⫽ 12.10, p ⬍ .01 when compared with the control prime condition; B ⫽ 1.43, F(1, 34) ⫽ 12.48, p ⬍ .01 when compared with the mother prime condition. Moreover, these analyses also revealed that participants’ closeness to their friend had at least a marginally significant effect on their commitment to their friend’s goal— B ⫽ 0.76, F(1, 34) ⫽ 4.07, p ⬍ .05 when compared with the control prime condition; B ⫽ 0.97, F(1, 34) ⫽ 3.26, p ⬍ .10 when compared with the mother prime condition—and also significantly moderated the priming effects described above—B ⫽ 1.02, F(1, 34) ⫽ 6.26, p ⬍ .05 when compared to the control prime condition; B ⫽ 1.18, F(1, 34) ⫽ 4.38, p ⬍ .05 when compared to the mother prime condition. The direction of these interactions sug-

gested that the greater participants’ closeness to their friend, the more likely that priming his or her name will increase participants’ commitment to their friend’s goal. Figure 1 illustrates the significant singular and interactive effects described above.3 The results of Study 1 provide compelling initial evidence that our representations of significant others may have an implicit motivational relevance insomuch as these representations may automatically increase one’s commitment to the goals that these individuals have for us. Indeed, subliminally priming the name of participants’ mother increased participants’ commitment to the goal that their mother had for them, whereas priming the name of 3

In all of the presented studies, predicted values were calculated by using values that were one standard deviation above and below the means of the relevant variables in the regression equation (for a more elaborate discussion of simple slope analysis, see Aiken & West, 1991).

SHAH

666

participants’ friend increased participants’ commitment to the goal that their friend had for them. Moreover, Study 1 provided initial evidence that this interpersonal form of motivational priming is itself moderated by how close we perceive ourselves to the significant other in question. Priming the name of participants’ mother or friend had a significantly stronger effect when participants reported being close to the primed person than when they were relatively distant, suggesting that the nature of one’s relationship to a significant other may dictate how they influence one’s goal pursuits and goal intentions. Although Study 1 examined goals of obvious importance to participants’ significant others, it did not assess actual goal pursuit. Moreover, although this study compellingly demonstrated that representations of significant others may affect the strength of one’s pursuit intentions, it failed to examine the process underlying this motivational priming effect. Specifically, does the priming of a significant other actually increase the accessibility of associated goals? Study 2 sought to examine this possibility by using a technique similar to that described by Mikulincer (1998), in which goal accessibility was assessed through the speed of participants’ responses to goal-related words on a lexical decision task. The present study also sought to demonstrate that, in addition to affecting goal accessibility and goal commitment, such priming may also influence actual goal pursuit.

Study 2 Study 2 examined whether priming participants’ representation of their father would increase their pursuit of a task goal, especially when participants reported feeling close to their father and perceived their father as wanting them to do well on the task. Participants completed an anagram task that was described as an important measure of analytic reasoning and were instructed to find as many different solutions to the anagrams as possible. Immediately before they completed this task, participants were subliminally primed either with words representing the concept “father” or nonword controls. This priming manipulation was followed by an assessment of the accessibility of the task goal and an assessment of participants’ actual goal pursuit, as measured by their performance and persistence on the anagram task.

Method Participants Fifty-one University of Wisconsin—Madison undergraduate psychology students (19 males and 32 females) participated in return for course credits. Participants were between the ages of 18 and 25. The gender of the participants did not significantly influence the presented results either singularly or interactively and thus will not be discussed further.

Procedure The procedure was carried out on IBM-compatible computers. Participants were informed that they would be completing an anagram task that was portrayed as an accurate measure of analytic reasoning. Participants were instructed that each anagram would consist of a string of letters that they were to unscramble into as many different words as possible using all of the letters in the string. All participants were additionally informed that individuals strong in analytic reasoning could find 80% of the possible words in the set of anagrams. Participants were then instructed that before beginning the actual anagram task, it was important to assess how quickly

they recognized words in general. Participants were told that even though such word-recognition speed is not directly related to analytic reasoning, it does affect how they do the anagrams. Accordingly, they would be asked to determine whether a presented letter string was a word or not as quickly and as accurately as they could. Participants were given an example followed by 12 warm-up presentations. Unbeknownst to the participants, each of the practice trials was preceded by a prime word (or letter string) that was presented for 10 ms and was backward masked by an “orientation marker” that consisted of a letter string (xxxxxxxxxxxxxx). After a brief delay varying randomly in duration, participants saw a letter string and had to determine as quickly as possible whether it was a word or not by pressing the y or n key. Participants in the “father prime” condition were primed with the words father and dad on alternate trials. Participants in the “control prime” condition were primed with two different nonword letter strings. These primes were presented for 10 ms (backward masked). Participants were then presented with an additional series of 34 trials that were designed to assess the accessibility of the previously presented task goal. These trials contained target words relating to the task goal as well as control words (and an equal number of nonwords). No primes were presented before these trials (participants were instead first presented with the same mask used on the initial trials). Six of these trials involved target words that represented the task goal (e.g., anagram, analytic). These words were chosen because they reflected the “contents” of the task goal in representing what they were doing (solving anagrams) and why they were doing it (to demonstrate analytic reasoning). Two other trials consisted of control words that the participants had also recently seen but that were unrelated to the goal (e.g., spacebar). After completing the lexical decision task, all participants were asked to rate how committed they were to doing well on the anagram task and how difficult they perceived this task to be. Each rating was provided on a 10-point scale ranging from 1 (not at all) to 10 (extremely). They then completed the eight anagrams comprising the critical “anagram test.” The computer recorded how long participants took to solve each anagram and the answers they provided. At that point, participants were asked to provide their general impressions of the task. Specifically, participants were asked to indicate how important it would be to their father that they possess analytic reasoning and how close, strong, and positive their relationship is to their father. Each of these three ratings was provided on a 10-point scale ranging from 1 (not at all) to 10 (extremely). After being asked additional “filler” questions, participants were thoroughly debriefed and probed for possible suspicion. The true nature of the study was fully revealed to participants, and they were encouraged to comment or pose any questions they might have had about the nature of the study. Participants’ closeness to their father. Participants’ closeness to their father was assessed by summing their scaled responses to the three separate questions about their relationship to their father (participants’ responses to these questions were highly interrelated, Cronbach’s alpha ⫽ .98). The higher this score, the closer participants’ relationship to their father. Response speed to goal-related items. The speed of participants’ responses to goal-related items on the lexical decision task was determined by summing participants’ reaction times on the goal-related trials (i.e., to words related to the task goal) after first submitting each of the latencies to a natural log transformation to lessen the impact of outliers. We assumed that the speed of participants’ responses on these trials was proportionate to the accessibility of the task goal. Because the latency of incorrect responses would be difficult to interpret in terms of accessibility, only correct responses were used in all subsequent analyses (see Bargh, Chaiken, Govender, & Pratto, 1992; Fazio, 1990).4

4

No systematic differences were found in participants’ error rate on the lexical decision trials for any of the reported studies. The average error rate for the lexical decision trials was 3.0% in Study 2, 2.3% in Study 3, 1.3% in Study 4, and 2.4% in Study 5.

GOAL PRIMING THROUGH OTHERS Response speed to control items. The speed of participants’ responses to control items was calculated in an identical manner. We summed their reactions to the control words on the lexical decision task after first log transforming each of the latencies to lessen the impact of outliers. Again, only correct responses were used.5 Task performance and persistence. The success of participants’ performance on the anagram task was operationally defined in terms of the sum of correct words they found across all of the eight trials. Similarly, participants’ persistence at the task was operationally defined by summing the times they spent on each of the eight trials after first log transforming each trial time to lessen the impact of outliers.

Results Effects on Goal Accessibility and Rated Goal Commitment An initial regression analysis examined whether the priming manipulation, participants’ closeness to their father, and the degree to which their father was perceived to value the task goal affected the accessibility of the task goal, as assessed through the lexical decision task. This regression analysis included the three variables mentioned above and their two-way and three-way interactions, as well as participants’ accessibility of nongoal words as a control. This analysis found that the father priming manipulation did not significantly affect the accessibility of the goal-related words (F ⬍ 1). This accessibility (the inverse of participants’ goal-related latency) was marginally positively related to participants’ closeness to their father, B ⫽ 0.038, F(1, 42) ⫽ 3.23, p ⬍ .10, and significantly related to the value that their father was perceived to place in pursuing the goal, B ⫽ 0.046, F(1, 42)⫽ 5.24, p ⬍ .05. This analysis also revealed that participants’ closeness to their father and the value that their father was perceived to place in pursuing the goal interacted with the priming manipulation, B ⫽ 0.056, F(1, 42) ⫽ 9.19, p ⬍ .01; B ⫽ 0.042, F(1, 42) ⫽ 7.91, p ⬍ .01, respectively, and with each other, B ⫽ 0.049, F(1, 42) ⫽ 6.64, p ⫽ .01. Perhaps most importantly, the three-way interaction of participants’ closeness to their father, the value that their father was perceived to place in pursuing the goal, and the priming manipulation was significantly positive, B ⫽ 0.046, F(1, 42) ⫽ 5.24, p ⬍ .05, suggesting that the effect of priming the father concept increased with the degree to which participants were close to their father and with the degree to which their father was perceived to value their pursuit of this goal (see Figure 2A). No significant differences were found in the speed of participants’ responses to the control items (Fs ⬍ 1). A separate regression analysis examined whether the same three independent variables described above singularly and interactively influenced participants’ commitment to the task goal. This regression analysis found that the father priming manipulation and participants’ closeness to their father did not significantly affect participants’ task goal commitment (Fs ⬍ 1). The value that participants perceived their father as having for the goal, however, was significantly related to their own commitment, B ⫽ 1.72, F(1, 43) ⫽ 24.65, p ⬍ .001. This significant main effect was additionally found to be significantly stronger when participants were primed with father-related words, B ⫽ 0.69, F(1, 43) ⫽ 4.81, p ⬍ .05. Although none of the other two-way interactions were significant, the three-way interaction of participants’ closeness to their father, the perceived value that their father placed in pursuing the

667

goal, and the priming manipulation was significantly positive, B ⫽ 0.68, F(1, 43) ⫽ 5.78, p ⬍ .05, suggesting that the effect of priming father-related words on their task goal commitment increased with the degree to which participants were close to their father and with the degree to which their father was perceived to value the pursuit of this goal, as illustrated in Figure 2B.

Effects on Goal Pursuit and Attainment But does goal priming through a significant other actually influence the extent to which participants pursue the presented task goal? To examine this possibility, separate regressions examined participants’ anagram persistence and performance as a function of the priming manipulation, participants’ closeness to their father, the degree to which their father was perceived to value the task goal, and the two-way and three-way interactions of these variables. Similar to the accessibility and commitment effects described above, these analyses found that the father priming manipulation and participants’ closeness to their father did not significantly affect participants’ task goal persistence or performance. The value that participants perceived their father as having for the goal, however, was significantly related to participants’ persistence, B ⫽ 2.17, F(1, 43) ⫽ 8.27, p ⬍ .01, and to their performance, B ⫽ 1.67, F(1, 43) ⫽ 4.74, p ⬍ .05. The above significant main effects were additionally found to be significantly stronger when participants were primed with father-related words—B ⫽ 1.48, F(1, 43) ⫽ 4.63, p ⬍ .05 for participants’ task persistence; B ⫽ 1.60, F(1, 43) ⫽ 4.21, p ⬍ .05 for their task performance—and when participants were close to their father— B ⫽ 1.52, F(1, 43) ⫽ 3.93, p ⫽ .05 for participants’ persistence; B ⫽ 1.69, F(1, 43) ⫽ 4.73, p ⬍ .05 for participants’ performance. Although the priming manipulation did not significantly interact with participants’ closeness to their father to predict task persistence or performance, the three-way interaction of participants’ closeness to their father, the perceived value that their father places in the task goal, and the priming manipulation significantly predicted participants’ task persistence, B ⫽ 1.57, F(1, 43) ⫽ 4.31, p ⬍ .05, and their task performance, B ⫽ 1.58, F(1, 43) ⫽ 4.18, p ⬍ .05, suggesting that the effect of priming father-related words on participants’ task goal persistence and performance increased 5 To keep the total number of trials to a minimum and to limit the possibility of suspicion regarding the purpose of the lexical decision task (because participants had recently been exposed to these words), only two trials involving “previously introduced” (and positively valenced) target words were included. This does potentially raise issues regarding the reliability of participants’ control accessibility time. However, it should be noted that the participants’ control accessibility are included in all of the regression analyses that involve their goal-related accessibility and were not found to have significant effects. Moreover, the lexical decision task included six additional trials with positively valenced target words that participants had not recently seen (e.g., handle, simple). Although these six trials were not used to calculate participants’ control accessibility because they used target concepts not previously introduced in the experimental procedure, participants’ lexical decision times on these trials did not systematically differ from their times on the original control trials in any of the presented studies. If these additional trials are included in calculating participants’ control accessibility, the significant effects of participants’ goal accessibility remain unchanged, as do the nonsignificant effects of participants’ control accessibility.

SHAH

668

Figure 2. Goal commitment and goal accessibility by father prime manipulation, value of goal to father, and closeness to father. RT ⫽ reaction time.

with the degree to which participants were close to their father and with the degree to which their father was perceived to value their pursuit of this goal. These three-way interactions are illustrated in Figures 3A and 3B.

Mediation By Goal Accessibility To examine whether the previously mentioned priming effects on anagram persistence and performance were mediated by changes in goal accessibility, it is necessary to show that persis-

tence and performance are directly affected by participants’ goal accessibility and that when these direct effects are accounted for, the main and interactive effects involving the priming manipulation are rendered nonsignificant (see Baron & Kenny, 1986). Thus, two final regression analyses examined the effect of priming on anagram performance and persistence, while also including the level of participants’ task goal accessibility and their control accessibility (as measured on the lexical decision task). When goal accessibility is included in these analyses, these differences are found to have a significant effect on performance and

GOAL PRIMING THROUGH OTHERS

669

Figure 3. Anagram persistence and performance by father prime manipulation, value of goal to father, and closeness to father.

persistence, B ⫽ 11.07, F(1, 41) ⫽ 6.49, p ⫽ .01; B ⫽ 8.29, F(1, 41) ⫽ 4.71, p ⬍ .05, respectively, and to render the previously significant main and interactive effects of priming on anagram performance and persistence nonsignificant (all Fs ⬍ 1). These analyses also revealed that participants’ control accessibility was not significantly related to their performance and persistence.6 The results of Study 2 provide additional evidence that participants’ representation of significant others may have motivational relevance in automatically affecting goal accessibility and goal commitment, as well as the degree to which the goal is pursued and

attained (as represented by anagram persistence and performance respectively). Moreover, as suggested in Study 1, these priming effects were found to be moderated by participants’ relationship to

6 Although Studies 2–5 find evidence that changes in goal accessibility mediate significant-other priming effects on task performance and task persistence, the more direct link between goal accessibility and goal-related performance and persistence may itself be mediated (or partially mediated) by changes in how goals are consciously construed. Indeed, Shah (2002)

SHAH

670

their father and the value that their father is perceived to have for the task goal. The closer participants’ relationship to their father and the greater the father is perceived to value the task goal, the more readily the psychological presence of this individual will increase goal accessibility and goal commitment. Up to this point, the psychological presence of individuals’ significant others has been considered in terms of enhancing goal accessibility, goal commitment, and goal pursuit. Yet, if the motivational effects of significant others depend on the goals these individuals have for us, then these individuals may also lower baseline levels of goal accessibility, commitment, and pursuit when they place little or no value in a given goal or when these individuals are very much associated with the pursuit of a different goal. Just as individuals might serve to direct attention toward certain pursuits, so might they serve to direct attention away from “trivial” pursuits that presumably take time and attention away from more worthy pursuits. This possibility is explored in Study 3, which also examines whether such inhibitory effects may themselves be moderated by the strength of one’s relationship to a significant other. Study 3 also used a different priming procedure than that used in Study 2. Rather than priming a generally relevant significant other, such as one’s father, Study 3 primed participants with the particular significant other that they perceived to be most relevant (in a positive or negative sense) to the task at hand.

Study 3 Study 3 asked participants to complete an anagram task, which was now described as a measure of verbal fluency. Before completing this task, participants were primed with either the name of a significant other who would want them to possess verbal fluency, the name of a significant other who would least want them (or not want them) to possess verbal fluency, or a control nonword. This priming manipulation was again followed by an assessment of the accessibility of the task goal and an assessment of participants’ actual goal pursuit.

Method Participants One hundred fifty-three University of Wisconsin—Madison undergraduate psychology students (83 males and 70 females) participated in return for course credits. Participants were between the ages of 18 and 25. The gender of the participants did not significantly influence the presented results either singularly or interactively and thus will not be discussed further.

Procedure The procedure was carried out on personal computers. Participants were informed that they would be completing an anagram task identical to the task described in Study 2, except that this time the task was portrayed as a measure of verbal fluency—a “valuable skill for reading and writing.” Participants were told that individuals with a high degree of verbal fluency

has recently found evidence that, in priming goals, significant others may also affect how these goals are consciously appraised (e.g., in terms of their difficulty or regulatory focus), and how they are subsequently pursued.

could find 80% of all the possible solutions. At that point, participants were asked to indicate the name of a significant other who would most want them to possess verbal fluency and the name of the significant other who would least want them (or not want them) to possess verbal fluency. To disguise the true nature of the study and to lessen the accessibility of the previously listed names, participants were asked to list the names of additional individuals who were significant in various unrelated ways (e.g., the person they knew who most wanted them to be strong). Participants then completed a questionnaire that included questions concerning how close the participant was to each of the individuals they had listed previously (which included the significant other who most wanted them to possess verbal fluency and the individual who least wanted them to possess this fluency). Specifically, participants were asked to indicate how close they felt toward the significant other and how close that individual felt toward them. Each answer was provided on a 7-point scale ranging from 1 (not at all) to 7 (extremely). After answering these questions, participants went on to complete a battery of “filler” questionnaires that were comprised of questions unrelated to the current study and designed to lessen the general accessibility of the individuals that they had previously listed and rated (see Baldwin, 1994, for a similar procedure). After completing these filler questionnaires (which required approximately 10 minutes), participants were told that they would now complete the anagram task that had been described previously. Participants first performed three “practice” anagrams before proceeding with the actual anagram test. Before each of these practice anagrams, the computer flashed a prime word for 50 ms (backward masked). Participants were randomly assigned to receive one of three different prime words. One group of participants was primed with the name of the significant other who most wanted them to possess verbal fluency. A second group was primed with the name of the person who least cared if they possessed verbal fluency. Finally, the remaining participants were primed with a nonword letter string (xzsdfgjk). Before completing the anagrams, participants were then given a lexical decision task very similar to that described in Study 2.7 The same justification for this task was also provided (that it was needed to account for differences in word-recognition speed) and the same procedure was followed to subliminally present the prime word that participants had been exposed to while completing the practice anagrams. Once again, participants were primed with this word before each of the 12 warm-up sets of lexical decision trials. The primes were flashed for 50 ms and backward masked. Participants were then presented with an additional series of 34 trials that were designed to assess the accessibility of the previously presented task goal. These trials contained target words relating to the task goal as well as control words (and an equal number of nonwords). As in Study 2, no primes were presented before these trials. Six of these trials involved target words that represented the task goal (e.g., anagram, reading, and writing). These words were chosen because they reflected the “contents” of the task goal in representing what they were doing (solving anagrams) and why they were doing it (to indicate writing and reading skill). Two other trials consisted of control words that the participants had also recently seen but that were unrelated to the goal (e.g., spacebar). After completing the lexical decision task, all participants were asked to rate how committed they were to doing well on the anagram task and the difficulty of this task on a 10-point scale ranging from 1 (not at all) to 10 (extremely). Participants then completed the eight anagrams comprising the critical “anagram test.” The computer recorded how long participants took to solve each anagram and the answers they provided. At that point, participants were thoroughly debriefed and probed for possible suspicion. The true nature of the study was fully revealed to participants, and they

7 Participants in Studies 3–5, however, were given additional opportunities to rest between lexical decision trials, in an effort to increase the overall stability of reaction times.

GOAL PRIMING THROUGH OTHERS were encouraged to ask any questions they still had about the nature of the study. Participants’ closeness to their significant others. The degree to which participants felt close to the two significant others they had listed initially was determined by separately summing their closeness ratings for each of these individuals. Thus, two separate scores were created, one for the significant other who most wanted them to possess verbal fluency and one for the significant other who least wanted them to possess this fluency. Response speed to goal-related items and control items. The speed of participants’ responses to goal-related items on the lexical decision task was determined by summing participants’ reaction times on the goalrelated trials (i.e., to words related to the task goal). The speed of participants’ responses to control items was summed in an identical manner. As in Study 2, each of these latencies was log transformed and only correct responses were used. Task performance and persistence. Participants’ anagram task performance was again calculated by summing the number of correct solutions found on the eight anagram trials. Anagram persistence was calculated by summing the time spent on the eight trials after first log transforming each trial to lessen the impact of outliers.

Results Effects on Goal Accessibility Two separate initial regression analyses examined whether the priming manipulation and participants’ closeness to their listed significant others affected the accessibility of the task goal, as assessed through the lexical decision task. These analyses included the priming manipulation, participants’ closeness rating, and the interaction of these two variables as well as participants’ control accessibility. The first analysis compared participants primed with the control word with those primed with their significant other who most wanted them to possess verbal fluency. Priming this significant other was found to significantly increase the accessibility of the task goal, as seen in decreased reaction times to goal-related items on the lexical decision task (accounting for reaction times to control items), B ⫽ 0.033, F(1, 83) ⫽ 7.13, p ⬍ .05. Moreover, this analysis also found that this priming effect was significantly affected by participants’ closeness to the primed significant other, B ⫽ 0.033, F(1, 83) ⫽ 3.87, p ⬍ .05, and that this closeness significantly moderated the previously described priming effect, B ⫽ 0.031, F(1, 83) ⫽ 4.04, p ⬍ .05, in such a way as to suggest that the positive effect of priming this significant other on the accessibility of the task goal became significantly more positive as participants’ closeness to this individual increased. No significant differences were found in the speed of participants’ responses to the control items (Fs ⬍ 1). A second analysis compared participants primed with the control word with those primed with their significant other who least wanted them to possess verbal fluency. Priming this significant other was found to significantly decrease the accessibility of the task goal, as seen in increased reaction times to goal-related items on the lexical decision task (accounting for reaction times to control items), B ⫽ ⫺0.031, F(1, 107) ⫽ 6.25, p ⬍ .01. Moreover, this analysis also found that this priming effect was marginally affected by participants’ closeness to this significant other, B ⫽ ⫺0.021, F(1, 107) ⫽ 2.82, p ⬍ .10, and that this closeness significantly moderated the previously described negative priming effect, B ⫽ ⫺0.028, F(1, 107) ⫽ 4.86, p ⬍ .05, in such a way as to suggest that the negative effect of priming this significant other

671

on the accessibility of the task goal became significantly more negative as participants’ closeness to this individual increased. No significant differences were found in the speed of participants’ responses to the control items (Fs ⬍ 1). The accessibility effects described above are illustrated in Figures 4A and 4B.

Effects on Goal Commitment Two other regression analyses examined whether the priming manipulation and participants’ closeness to their listed significant others affected their rated commitment to the task goal. These analyses included the priming manipulation, participants’ closeness rating, and the interaction of these two variables. The first analysis compared participants primed with the control word with those primed with their significant other who most wanted them to possess verbal fluency. Priming this significant other was found to significantly increase participants’ rated commitment to the task goal, B ⫽ 0.49, F(1, 84) ⫽ 5.77, p ⬍ .05. Moreover, this analysis also found that this commitment rating was affected by participants’ closeness to this significant other, B ⫽ 0.64, F(1, 84) ⫽ 8.04, p ⬍ .01, and that this closeness significantly moderated the previously described priming effect, B ⫽ 0.49, F(1, 84) ⫽ 4.63, p ⬍ .05, in such a way as to suggest that the positive effect of priming this significant other on commitment to the task goal became significantly more positive as participants’ closeness to this individual increased. A separate analysis compared participants primed with the control word with those primed with their significant other who least wanted them to possess verbal fluency. Priming this significant other was found to significantly decrease commitment to the task goal, B ⫽ ⫺0.50, F(1, 108) ⫽ 5.02, p ⬍ .05. Moreover, this analysis also found that this commitment rating was marginally affected by participants’ closeness to this significant other, B ⫽ ⫺0.43, F(1, 108) ⫽ 3.00, p ⬍ .10, and that this closeness significantly moderated the previously described negative priming effect, B ⫽ ⫺0.52, F(1, 108) ⫽ 5.23, p ⬍ .05, in such a way as to suggest that the negative effect of priming this significant other on commitment to the task goal became significantly more negative as participants’ closeness to this individual increased. The commitment effects described above are illustrated in Figures 4C and 4D.

Effects on Goal Pursuit and Attainment But did the priming of significant others actually influence the extent to which participants pursued the presented task goal? To examine this possibility, separate regressions examined participants’ anagram performance and persistence as a function of the priming condition and their closeness to their listed significant others, as well as the interaction of these two variables. Similar to the accessibility and commitment effects described above, participants who were primed with the name of the significant other who most wanted them to possess verbal fluency persisted longer on the anagrams and found significantly more solutions than participants in the control condition—B ⫽ 0.93, F(1, 84) ⫽ 8.17, p ⬍ .01 for persistence; B ⫽ 1.14, F(1, 84) ⫽ 6.48, p ⫽ .01 for performance. These analyses also revealed that participants’ closeness to this individual had at least a marginally significant positive effect on persistence and performance—B ⫽ 0.66, F(1, 84) ⫽ 3.85, p ⬍ .05

672

SHAH

Figure 4. Task goal accessibility and task goal commitment by importance to significant other, salience of this other, and closeness to this other.

for persistence; B ⫽ 0.86, F(1, 84) ⫽ 2.84, p ⬍ .10 for performance—and that this closeness significantly moderated the positive priming effects on anagram performance and persistence described above—B ⫽ 0.89, F(1, 84) ⫽ 6.90, p ⫽ .01 for persistence; B ⫽ 0.99, F(1, 84) ⫽ 3.93, p ⫽ .05 for performance—in such a way as to suggest that the positive effect of priming this significant other on anagram persistence and performance became significantly more positive as participants’ closeness to this individual increased. These effects are illustrated in Figures 5A and 5B. A separate set of regression analyses found that participants primed with the name of the significant other who least wanted them to do well on the anagrams persisted significantly less and performed significantly worse than participants in the control condition—B ⫽ ⫺0.76, F(1, 108) ⫽ 6.24, p ⫽ .01 for persistence; B ⫽ ⫺1.01, F(1, 108) ⫽ 5.36, p ⬍ .05 for performance. These analyses also revealed that participants’ closeness to this individ-

ual had a marginally significant negative effect on persistence and performance—B ⫽ ⫺0.89, F(1, 108) ⫽ 2.84, p ⬍ .10 for persistence; B ⫽ ⫺1.24, F(1, 108) ⫽ 2.77, p ⬍ .10 for performance— and that this closeness significantly moderated the negative priming effects on anagram persistence and performance described above—B ⫽ ⫺0.65, F(1, 108) ⫽ 4.24, p ⬍ .05 for persistence; B ⫽ ⫺0.87, F(1, 108) ⫽ 4.12, p ⬍ .05 for performance—in such a way as to suggest that the negative effect of priming this significant other on anagram persistence and performance became significantly more negative as participants’ closeness to this individual increased. These effects are illustrated in Figures 5C and 5D.

Mediation Effects To examine whether the previously mentioned priming effects on anagram persistence and performance were mediated by

GOAL PRIMING THROUGH OTHERS

673

Figure 5. Task persistence and task performance by importance to significant other, salience of this other, and closeness to this other.

changes in goal accessibility, we followed the same analytic strategy detailed in Study 2 (see Baron & Kenny, 1986). Thus, four final regression analyses examined the main and interactive effects of the priming manipulation and participants’ closeness to their significant other on anagram performance and persistence, while also including the level of participants’ task goal accessibility and their control accessibility (as measured on the lexical decision task). When differences in goal accessibility are included in the analyses examining the effect of priming the significant other who most wanted participants to do well, these differences were found to have a significant effect on performance and persistence— B ⫽ 8.54, F(1, 82) ⫽ 8.46, p ⬍ .01 for performance; B ⫽ 8.87, F(1, 82) ⫽ 8.93, p ⬍ .01 for persistence—and to render nonsignificant the positive priming and closeness effects described earlier—B ⫽ 0.58, F(1, 82) ⫽ 1.41, p ⬎ .20 for the main effect of priming on persistence (all other Fs ⬍ 1). When participants’ goal accessibility and their control accessibility are included in the analyses examining the effect of priming

the significant other who least wanted participants to do well, these differences were again found to have a significant effect on performance and persistence—B ⫽ 9.35, F(1, 106) ⫽ 8.30, p ⬍ .01 for performance; B ⫽ 7.51, F(1, 106) ⫽ 6.72, p ⫽ .01 for persistence—and to render nonsignificant the negative priming and closeness effects described earlier (all Fs ⬍ 1). The results of Study 3 provide strong evidence that participants’ representation of significant others may have motivational relevance in automatically increasing or decreasing goal accessibility and goal commitment, which subsequently influences the degree to which the goal is pursued and attained (as represented by anagram persistence and performance respectively). Moreover, as suggested in Study 1, these priming effects were found to be moderated by participants’ relationship to their listed significant others. The closer participants’ relationship, the more readily their significant other will increase (or decrease) task goal accessibility, task goal commitment, and participants’ task persistence and performance.

SHAH

674

Although the studies described so far have demonstrated that the implicit influence of significant others may vary as a function of the strength of our relationship to these individuals, this influence may also vary as a function of other factors of the situation or the individuals involved. As mentioned earlier, one such factor may be the number of additional goals that a significant other has for us. The greater number of different goals that a significant other has for us, the less likely that any one of these goals will be strongly invoked, and acted upon, when encountering or thinking about this individual. Study 4 sought to examine this possibility with respect to participants’ pursuit of a given task goal.

Study 4 The design of Study 4 was very similar to that of Study 3, except that participants were asked to provide the name of a close significant other who most wanted them to possess verbal fluency and the additional goals this individual wanted them to pursue in order to examine how the number of such goals would moderate the motivational effects of priming this significant other.

Method Participants Forty-eight University of Wisconsin—Madison undergraduate psychology students (17 males and 31 females) participated in return for course credits. Participants were between the ages of 18 and 25. The gender of the participants did not significantly influence the presented results either singularly or interactively and thus will not be discussed further.

suspicions in a manner identical to that described in Study 3. Participants’ response speed to goal-related items and to control items, and their anagram performance and persistence, were calculated in the same manner as described in Study 3. The number of additional goals that their significant other had for them was also calculated by summing the number of such goals listed by each participant.

Results Effects on Goal Accessibility An initial regression analysis examined whether the priming manipulation and the number of other goals that participants’ significant other had for them affected the accessibility of the task goal, as assessed through the lexical decision task. This regression analysis included the priming manipulation, the measure of their significant others’ additional goals for them, and the interaction of these two variables, as well as participants’ control accessibility. The analysis found that the priming of participants’ close significant other who most wanted them to possess verbal fluency significantly increased the accessibility of the task goal, as seen in the speed at which participants responded to goal-related items (in comparison with control items) on the lexical decision task, B ⫽ 0.045, F(1, 43) ⫽ 6.67, p ⫽ .01. This significant priming effect was also found to be significantly weakened by increases in the number of other goals that participants’ significant other had for them, B ⫽ ⫺0.035, F(1, 43) ⫽ 4.86, p ⬍ .05 (see Figure 6A). The number of other goals associated with a significant other did not directly affect the accessibility of the task goal (F ⬍ 1), and no significant differences were found in the speed of participants’ responses to the control items (Fs ⬍ 1).

Procedure The procedure was very similar to that used in Study 3. Once again, participants completed an anagram task on the computer that was described as a measure of verbal fluency—a “valuable skill for reading and writing”—and all participants were additionally informed that individuals with a high degree of verbal fluency could find 80% of the possible words in the set of anagrams. At that point, participants were asked to indicate the name of a close significant other who would most want them to possess verbal fluency and the names of additional individuals who were significant in various unrelated ways (e.g., the person they knew who most wanted them to be strong). Participants then completed a questionnaire that asked participants, among many other filler questions, to list all the other goals that the close significant other linked to verbal fluency would also want them to possess (e.g., “to be social,” “to get married,” “to be religious,” “to do well in school this semester”). After listing these additional goals, participants went on to complete a battery of “filler” questionnaires designed to lessen the accessibility of the names and responses they had provided to this point. After completing these filler questionnaires (which required approximately 10 min) participants were told that they would now complete the anagram task that had been described previously. Before completing the actual anagram task, participants were primed with either the name of the significant other who wanted them to do well on the anagram task or a nonword letter string while completing the three practice anagrams and the practice trials of the lexical decision task. This priming procedure was identical to that described in Study 3. After completing the lexical decision task, participants rated how committed they were to doing well on the anagram task and the difficulty of this task. Both ratings were provided on a 10-point scale ranging from 1 (not at all) to 10 (extremely). Participants then completed the actual anagram trials and afterwards were thoroughly debriefed about the true nature of the study and probed for possible

Effects on Goal Commitment A separate regression analysis examined whether the priming manipulation and the number of other goals that participants’ significant other had for them also affected participants’ commitment to the task goal. This regression analysis included the priming manipulation, the measure of their significant others’ additional goals for them, and the interaction of these two variables. The analysis found that the priming of participants’ close significant other who most wanted them to possess verbal fluency significantly increased commitment to the task goal, B ⫽ 0.54, F(1, 44) ⫽ 4.76, p ⬍ .05. This significant priming effect was also found to be significantly weakened by increases in the number of other goals that participants’ significant other had for them, B ⫽ ⫺0.67, F(1, 44) ⫽ 4.33, p ⬍ .05 (see Figure 6B). The number of other goals associated with a significant other did not directly affect the accessibility of the task goal (F ⬍ 1).

Effects on Goal Pursuit and Attainment Separate regressions again examined participants’ anagram performance and persistence as a function of the priming condition and the number of other goals that participants’ significant other had for them. Similar to the accessibility and commitment effects described above, participants who were primed with the name of the close significant other who wanted them to possess verbal fluency persisted longer on the anagrams and found significantly more solutions than participants in the control condition—

GOAL PRIMING THROUGH OTHERS

675

Figure 6. Task goal accessibility, task goal commitment, task persistence, and task performance by priming manipulation and number of additional goals associated with significant other.

B ⫽ 1.84, F(1, 44) ⫽ 9.89, p ⬍ .01 for persistence; B ⫽ 2.16, F(1, 44) ⫽ 11.15, p ⬍ .01 for performance. Although the number of other goals associated with a significant other did not directly affect anagram persistence or performance (Fs ⬍ 1), the previously described priming effects were also found to be significantly weakened by increases in the number of other goals that participants’ significant other had for them—B ⫽ ⫺1.46, F(1, 44) ⫽ 4.14, p ⬍ .05 for anagram persistence; B ⫽ ⫺1.21, F(1, 44) ⫽ 4.28, p ⬍ .05 for anagram performance—as illustrated in Figures 6C and 6D.

Mediation Effects To examine whether the previously mentioned priming effects on anagram persistence and performance were mediated by changes in goal accessibility, we followed the same analytic strategy detailed in Study 2 (see Baron & Kenny, 1986). Thus, two final regression analyses examined the effect of priming on anagram performance and persistence, while also including the level of participants’ task goal accessibility (as measured on the lexical decision task). When differences in goal accessibility are included

SHAH

676

in these analyses, these differences are found to have a significant effect on performance, B ⫽ 10.88, F(1, 42) ⫽ 6.66, p ⫽ .01, and persistence, B ⫽ 7.16, F(1, 42) ⫽ 4.92, p ⬍ .05, and to render the previously significant main effects of priming on anagram performance and persistence nonsignificant—B ⫽ 1.09, F(1, 42) ⫽ 2.27, p ⬎.10 for performance; B ⫽ 0.775, F(1, 42) ⫽ 1.75, p ⬎.15 for persistence. The inclusion of these variables also rendered nonsignificant the interactive effects of priming with the number of other goals that participants’ significant other had for them (Fs ⬍ 1). In providing further evidence that participants’ representation of significant others may automatically increase goal accessibility and, subsequently, the degree to which a goal is pursued and attained, Study 4 also demonstrates that these implicit goal priming effects may be moderated by yet another factor: the extent to which a significant other is uniquely associated with the goal in question. The results suggest that the greater the number of goals that a significant other may have for us, the less readily he or she may automatically invoke any one goal in particular. The studies described to this point have examined how significant others may implicitly induce or inhibit goal pursuit, without directly considering how these effects may depend on one’s current conscious intentions. Indeed, in daily life we may often be primed to pursue a particular goal in situations where we are deliberately pursuing something else. How might thoughts of a significant other influence a previously chosen pursuit? Shah and Kruglanski (2002) recently found that when participants were consciously pursuing a task goal, the effect of priming a different goal depended on how this second goal related to the first. Priming an unrelated goal was found to draw attention and effort away from a conscious pursuit, as indicated by decreased task performance and persistence. Attention and effort were pulled toward the conscious pursuit, however, when individuals were primed with an alternative they perceived as being related to their focal pursuit. Study 5 sought to examine whether priming participants’ representation of a significant other would show a similar pattern of influence, in that priming a significant other associated with a goal unrelated to the task at hand would pull attention and effort away from a focal pursuit whereas priming a significant other associated with a goal related to the presented task would pull resources toward this activity.

between the ages of 18 and 25 (the gender of the participants was not recorded in this study).

Procedure The procedure was again carried out on IBM-compatible computers. Participants were first informed they would be completing an anagram task and were given the same rationale and instructions for this task as described in Study 2. Participants were then told that after they completed this anagram task (portrayed here as a measure of analytic reasoning), they would also be completing the “object-use task,” which was described as a measure of functional creativity. The object-use task was described as requiring participants to come up with as many different uses for common objects as they can think of. Participants were told that one purpose of this study was to examine whether functional creativity was related to analytic reasoning. At that point, participants were asked to indicate the name of a close significant other who would most want them to possess functional creativity as well as additional individuals who were significant in various unrelated ways. Participants were also asked to indicate how much they thought that functional creativity facilitated analytic reasoning on a 10point scale ranging from 1 (not at all) to 10 (extremely). Participants then completed a battery of “filler” questionnaires designed to lessen the accessibility of the names that they had previously listed and the relevant facilitation rating. After completing these filler questionnaires (which required approximately 10 min), participants were told that they would now complete the anagram task that had been described previously. Before completing the actual anagram task, participants were primed with either the name of the significant other who wanted them to possess functional creativity or a nonword letter string while completing the practice trials of the lexical decision task. This priming procedure was identical to that described in Studies 3 and 4. The lexical decision task was very similar to the one presented in these studies, except that it contained items relating to the functional creativity task as well as to control items. After completing the lexical decision task, participants rated how committed they were to doing well on the anagram task and the difficulty of this task on the same 10-point scale described earlier. Participants then completed the actual anagram trials and were then thoroughly debriefed about the true nature of the study and probed for possible suspicions. Participants’ response speed to items relating to the “alternative” goal (the functional creativity task) and to control items were calculated in the same manner as described in the previous studies, as was their anagram performance and persistence.

Results Study 5 Study 5 used a procedure similar to that described in Study 2, except that participants were led to believe that they would be completing two tasks. One of these tasks was described as assessing “analytic reasoning” and the other was described as assessing “functional creativity.” Before completing the first task, participants were primed with either a control nonword or the name of a significant other associated with the second task in order to examine the effect of priming this significant other on participants’ pursuit of the first task and how this priming effect may depend on the perceived relation between the two task goals.

Method Participants Fifty-three University of Wisconsin—Madison undergraduate psychology students participated in return for course credits. Participants were

Effects on Alternative-Goal Accessibility An initial regression analysis found that participants in the condition that primed their close significant other who most wanted them to possess functional creativity were significantly quicker at responding to the alternative-goal items on the lexical decision task than participants in the control condition (M ⫽ 667 ms vs. M ⫽ 732 ms), F(1, 48) ⫽ 11.52, p ⬍ .01, suggesting that this priming manipulation significantly increased the accessibility of the alternative task goal. As expected (and consistent with the findings of Shah & Kruglanski, 2002), participants’ facilitation rating (i.e., the degree to which they saw functional creativity as facilitating analytic reasoning) did not directly affect the accessibility of the alternative goal and did not moderate the priming effect described above (all Fs ⬍ 1). No significant differences were found in the speed of participants’ responses to the control items (Fs ⬍ 1).

GOAL PRIMING THROUGH OTHERS

677

Effects on Focal Goal Commitment A separate regression analysis found that the priming of participants’ close significant other who most wanted them to possess functional creativity had no main effect on participants’ commitment to the anagram task (F ⬍ 1). Participants’ facilitation rating, however, was found to significantly affect participants’ commitment to the anagram task, B ⫽ 0.484, F(1, 49) ⫽ 5.68, p ⬍ .05, and to significantly moderate the effect of the priming manipulation such that the effect of the priming manipulation on focal goal commitment became more positive to the degree that participants perceived functional creativity as facilitating analytic reasoning, B ⫽ 0.48, F(1, 49) ⫽ 5.50, p ⬍ .05. This significant interaction is illustrated in Figure 7A.

Effects on Goal Pursuit and Attainment Separate regressions again examined participants’ anagram performance and persistence as a function of the priming condition and their facilitation rating. Consistent with the results of Shah and Kruglanski (2002), these analyses found no significant effects for the priming manipulation on either anagram performance or persistence (all Fs ⬍ 1). Participants’ facilitation rating, however, was found to significantly affect participants’ persistence, B ⫽ 1.49, F(1, 49) ⫽ 5.58, p ⬍ .05, and performance, B ⫽ 1.35, F(1, 49) ⫽ 4.03, p ⫽ .05, on the anagram task, and to significantly moderate the effect of the priming manipulation such that the effect of the priming manipulation on anagram persistence and performance became more positive to the degree with which participants perceived functional creativity as facilitating analytic reasoning—B ⫽ ⫺1.48, F(1, 49) ⫽ 5.52, p ⬍ .05 for persistence; B ⫽ ⫺1.63, F(1, 49) ⫽ 5.85, p ⬍ .05 for performance. These significant interactions are illustrated in Figures 7B and 7C.

Mediation Effects To examine whether the previously mentioned priming effects on anagram persistence and performance were mediated by changes in the accessibility of the alternative goal, we again performed two final regression analyses to examine the effects of the priming manipulation and participants’ facilitation rating on anagram performance and persistence, while also including the level of participants’ alternative goal accessibility and control accessibility, as well as the interaction of these accessibility variables with participants’ facilitation rating. When these additional variables are included in these analyses, the interactive effect of alternative goal accessibility and participants’ facilitation rating is found to have a significant effect on persistence, B ⫽ 1.86, F(1, 45) ⫽ 6.95, p ⫽ .01, and performance, B ⫽ 3.68, F(1, 45) ⫽ 7.29, p ⫽ .01, and to render the previously significant interactive effects of priming and participants’ facilitation rating on anagram performance and persistence nonsignificant (Fs ⬍ 1). The results of Study 5 demonstrate yet another consequence of implicit goal priming: that it may inhibit the pursuit of unrelated intentions. Consistent with the results of Shah and Kruglanski (2002), the goal pursuit of participants primed with the name of an individual associated with a different goal depended on the participants’ perception of how this other goal related to the task goal at hand (i.e., the anagrams). When functional creativity was seen

Figure 7. Goal commitment, task persistence, and task performance by significant-other priming manipulation and the perceived relation of alternative goal to focal pursuit.

as unrelated to analytic reasoning, then priming an individual associated with functional creativity pulled attention and commitment away from the task at hand. Alternatively, priming the individual associated with functional creativity pulled attention and commitment toward the analytic reasoning goal when the two goals were perceived to facilitate each other.

General Discussion The present set of studies has sought to illustrate the implicit motivational effects that significant others may have on us, and to

678

SHAH

additionally begin to map both the determinants of such influence and its consequences for specific goal pursuits and goal-related experiences. Despite variations in methodology, and in the types of goals and self-regulatory consequences examined, the results of these studies have been quite consistent. Study 1 primed the name of participants’ mother or friend and demonstrated an effect on participants’ self-rated goal commitment. Study 2 extended these results to show that participants’ representation of their father might influence the commitment, accessibility, and pursuit of a given task goal when participants were close to their father and perceived that he would want them to do well on the presented task. Study 3 replicated the results of Study 2 using a different significant-other prime and further demonstrated that significant others could automatically inhibit commitment, accessibility, and pursuit of a task goal. In replicating the general priming effects yet again, Study 4 found evidence for another moderating factor: the motivational complexity of one’s representation of a significant other in terms of the additional goals to which this significant other is associated. Finally, Study 5 found evidence suggesting that representations of significant others can interfere with a conscious goal pursuit when one strongly associates this significant other with the pursuit of an unrelated goal. The present results are supportive of a connectionist conception of implicit social influence in which significant others vary in their cognitive connection to specific goals and motivations (see Shah & Kruglanski, 2000). Activating representations of these individuals may “spread” to related goals commensurate with the strength of the relevant person– goal association, which itself may depend on both motivational and cognitive factors, such as one’s closeness or attraction to the significant other and the complexity with which one associates this individual to other goals (see also Shah & Kruglanski, 2000). The present results extend research on the general self-regulatory significance of others’ goals and standards for us, such as the extensive work on self-discrepancy theory (Higgins, 1987; Moretti & Higgins, 1999a, 1999b), by specifying how significant others’ goals and standards for us may be automatically activated “in the moment” and thus the conditions under which significant others’ goals exert the greatest, and perhaps most automatic, influence on our own self-regulatory efforts. This work also goes beyond previous examinations of the implicit effect of significant others in focusing on the motivational (rather than the strictly evaluative, emotional, or behavioral) impact of others. Thus, whereas previous research has focused on the ways in which significant others may affect the standards one adopts in evaluating the self (see Baldwin, 1994), or directly influence action through a “perception-behavior” link (see Bargh, Chen, & Burrows, 1996; Dijksterhuis & van Knippenberg, 1998), the present results have highlighted the motivational influences of significant others by demonstrating how they may implicitly affect goal focus and goal adoption (as evident, e.g., in changes in goal activation, goal commitment, and task persistence). Such implicit goal priming may have distinct effects on behavior, cognition, and affective experience. For instance, although goal priming through significant others may influence behavior, as seen in task-related persistence and performance, such “goal-related” effects may differ from “perception-behavior” effects in that the former may be extinguished when the goal is attained or may actually increase with time if the activated goal remains unfulfilled (see Bargh, Gollwitzer, Lee-Chai, Barndollar, & Troetschel, 2001). Indeed, the fact

that the significant-other priming manipulation in the present set of studies affected task persistence and performance even after a delay (which was required for the completion of the lexical decision trials) further suggests that this priming manipulation was automatically invoking a task-relevant goal.

Implications for Goal-Related Appraisals and Emotional Experience In addition to invoking behavior, the process of goal pursuit may also prompt appraisals, both of the goals themselves and of the outcome of one’s attainment efforts (see Carver & Scheier, 1998; Kruglanski et al., 2000). Even while implicitly affecting the extent to which we automatically pursue goals, significant others may also implicitly affect how we come to appraise them. The degree to which we perceive a particular goal to be difficult or enjoyable, for example, may be implicitly influenced by our thoughts of whether a significant other thinks it will be difficult (or boring) for us to pursue (see Shah, 2002). Moreover, significant others may also affect how we come to appraise, and feel about, the outcome of our goal pursuits. Indeed, Shah (2002) recently found that priming the name of a significant other could influence the degree to which participants viewed a given task goal as an ideal or as an obligation and thus how they emotionally experienced success or failure on the task goal. When participants were primed with the name of a significant other who was perceived to ideally want them to do well, participants felt more cheerful when given success feedback and more dejected when given failure feedback. Alternatively, when participants were primed with the name of a significant other who was perceived to feel that they had a responsibility to do well on the task goal, they tended to feel more relaxed after receiving success feedback and more anxious when receiving negative feedback. Both findings are in line with the emotional distinctions outlined by regulatory focus theory (Higgins, 1998), and suggest that the implicit effects of significant others may extend the emotional consequences of goal pursuit (see also Chartrand, 2001).

Implications for Self-Evaluations The present results may also offer another mechanism for understanding the implicit ways in which significant others influence more general self-evaluations. Consider, for example, Baldwin’s (1994) investigation of how significant others implicitly affect the standards one uses for self-evaluation. Using a similar “significantother” priming procedure to that used in the present studies, this research found that participants primed with a “critical” significant other reported lower self-esteem and greater concern with evaluations from others after completing a difficult memory task than participants primed with an “accepting” significant other, suggesting that significant others are implicitly affecting the standards used for self-evaluation. The present results suggest that in addition to affecting the severity of one’s self-evaluative standards, priming a “critical” significant other (rather than an “accepting” other) may have invoked a goal to perform well on the task. The inherent difficulty of the memory task, however, may have prevented participants from fulfilling this goal, and this goal-related failure may have independently decreased their sense of selfefficacy and self-esteem. Thus, priming a relatively “critical”

GOAL PRIMING THROUGH OTHERS

significant other may not only raise one’s standard for defining success, it may also strengthen one’s motivation (or goal) to be successful, and thus one’s discouragement or disappointment over failure. Both of these priming effects may independently come to affect one’s self-evaluations.

Implications for Interpersonal Interactions and Long-Term Relations The implicit effects of others may also extend to our social interactions. Rarely are goals pursued in a social vacuum and to the extent that other individuals influence whether and how we pursue goals; they may also influence how we interact with them and with others, especially when these interactions are instrumental in pursuing the goal at hand. If thoughts of one’s mother, for example, bring to mind the goal to socialize, such a goal may affect how one interacts with his or her mother and perhaps with the rest of the family as well. The notion that individuals may implicitly prime our goal pursuit also raises the issue of whether our chronic social interactions may sometimes represent attempts to regulate the salience and pursuit of specific goals. We may seek out individuals strongly associated with the goals we want to pursue and avoid others strongly associated with goals we wish to drop as when, for instance, our focus on work leads us to forge closer relations with our colleagues and to avoid old friends (at least temporarily). If significant others are strongly associated with specific goals, these associations may also come to influence how we construe these individuals. Significant others linked to the same goal may come to be viewed as more interchangeable, or substitutable, with each other than with those linked to very different goals. Moreover, one’s situational attraction to a particular person may depend, in part, on whether the goals associated with this person are compatible with one’s current pursuits. Person– goal associations may ultimately come to influence the nature of our long-term relationships, and how such relationships change over time. A spouse or partner, for example, may be strongly, and perhaps exclusively, associated with goals of “love and romance” early in a relationship. With each passing year, however, this person may become increasingly associated with other, more mundane goals, such as those involving bills, taxes, and carpooling. The development of such associations may offer a mechanism for explaining the manner in which romantic relationships change, and often wane, over time (see, e.g., Berscheid, Snyder, & Omoto, 1989; Huston & Chorost, 1994). Thus, our partner’s propensity to spontaneously invoke our love and desire may be linked to the increasingly complex set of goals and motivations that we come to connect to our partner.

Considering Different Types of Person–Goal Associations Given the various ways in which significant others might come to automatically influence goal pursuits and social relations, it will be particularly important to more closely examine how individuals come to be associated with specific goals. Although the present research has explored how representations of others may serve as primes for the goals that these individuals have for us, this is likely not the only way in which significant others may come to automatically activate goal pursuits. For one, individuals may become

679

closely associated with goals because these individuals are the goals themselves—that is, the social ends of one’s strivings. One may desire, for instance, to attain the affection, love, or respect of a particular other (e.g., one’s parent, a potential romantic partner, or one’s boss), or one may want to possess, control, or dominate a particular other. In such instances, a significant other may become closely tied to the goals one has with regard to this individual. Inducing thoughts of such significant others, then, may be enough to invoke these goals, even when these individuals are not actually present. In addition to representing the object of one’s pursuit, significant others may become associated with a specific goal because they constitute a means for attaining a particular goal. Thus, one may come to see another as vital for attaining one’s objective of feeling “powerful,” “smart,” or “capable.” Such significant others may actively help one attain goals (such as when one helps another get elected president) or may simply constitute social mirrors that reflect one’s various desired attributes, and thus serve as comparison standards to assess progress. Indeed, Shah and Kruglanski (in press) have recently found evidence suggesting that goals may be primed by attainment means in a bottom-up fashion, whether these means constitute situations, behaviors, or other individuals.

Considering Other Factors Affecting Person–Goal Associations Finally, in examining how individuals become closely associated with goals, it will be vital to more fully consider qualities of the goals and individuals themselves. Although the present research has begun to examine the types of individuals that are likely to become associated with goals (e.g., close individuals not also associated with other goals), one might also consider the qualities of goals that might encourage their association to specific individuals. Person– goal associations that arise because individuals are seen as means for attaining goals, for instance, may be affected not only by other goals linked to the same person (as examined in Study 4), but also by other individuals who are means to the same goal. The strength of this type of person– goal association, then, may be affected by the abstractness of the goal involved. Relatively abstract goals such as being “friendly,” which can be gratified through virtually any person one might encounter, may be less strongly invoked by any particular individual than relatively more specific (or concrete) goals such as “developing a friendship,” which would presumably be gratified by a much more restricted set of individuals. Although the above possibilities await future study, they illustrate the potential of examining the implicit motivational effects of significant others and the utility of considering such effects in terms of the strength of person– goal associations and the motivational and cognitive factors underlying these connections. Such an approach may provide a framework to better understand the process through which goals are implicitly invoked in everyday social settings and the various self-regulatory and interpersonal consequences of such goal priming.

References Aarts, H., & Dijksterhuis, A. (2000). Habits as knowledge structures: Automaticity in goal-directed behavior. Journal of Personality and Social Psychology, 78, 53– 63.

680

SHAH

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. Andersen, S. M., & Berk, M. S. (1998). Transference in everyday experience: Implications of experimental research for relevant clinical phenomena. Review of General Psychology, 2, 81–120. Andersen, S. M., & Glassman, N. S. (1996). Responding to significant others when they are not there. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition (Vol. 3, pp. 262–321). New York: Guilford Press. Andersen, S. M., Glassman, N. S., Chen, S., & Cole, S. W. (1995). Transference in social perception: The role of chronic accessibility in significant-other representations. Journal of Personality and Social Psychology, 69, 41–57. Andersen, S. M., Glassman, N. S., & Gold, D. A. (1998). Mental representations of the self, significant others, and nonsignificant others: Structure and processing of private and public aspects. Journal of Personality and Social Psychology, 75, 845– 861. Andersen, S. M., Reznik, I., & Chen, S. (1997). The self in relation to others: Motivational and cognitive underpinnings. Annals of the New York Academy of Sciences, 818, 233–275. Andersen, S. M., Reznik, I., & Manzella, L. M. (1996). Eliciting transient affect, motivation, and expectancies in transference: Significant-other representations and the self in social relations. Journal of Personality and Social Psychology, 71, 1108 –1129. Anderson, J. R. (1974). Retrieval of propositional information from longterm memory. Cognitive Psychology, 6, 451– 474. Anderson, J. R. (1983). The architecture of cognition. Cambridge, MA: Harvard University Press. Anderson, J. R., & Reder, L. M. (1999). The fan effect: New results and new theories. Journal of Experimental Psychology: General, 128, 186 – 197. Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relationships as including other in the self. Journal of Personality and Social Psychology, 60, 241–253. Austin, J. T., & Vancouver, J. B. (1996). Goal constructs in psychology: Structure, process, and content. Psychological Bulletin, 120, 338 –375. Baldwin, M. W. (1992). Relational schemas and the processing of social information. Psychological Bulletin, 112, 461– 484. Baldwin, M. W. (1994). Primed relational schemas as a source of selfevaluative reactions. Journal of Social and Clinical Psychology, 13, 380 – 403. Baldwin, M. W., Carrell, S. E., & Lopez, D. F. (1990). Priming relationship schemas: My advisor and the Pope are watching me from the back of my mind. Journal of Experimental Social Psychology, 26, 435– 454. Bargh, J. A. (1990). Auto-motives: Preconscious determinants of thought and behavior. In E. T. Higgins & R. M. Sorrentino (Eds.), Handbook of motivation and cognition: Foundations of social behavior (Vol. 2, pp. 93–130). New York: Guilford Press. Bargh, J. A. (1996). Automaticity in social psychology. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 169 –183). New York: Guilford Press. Bargh, J. A. (1997). The automaticity of everyday life. In R. S. Wyer (Ed.), The automaticity of everyday life: Advances in social cognition (Vol. 10, pp. 1– 61). Mahwah, NJ: Erlbaum. Bargh, J. A., & Barndollar, K. (1996). Automaticity in action: The unconscious as a repository of chronic goals and motives. In P. M. Gollwitzer & J. A. Bargh (Eds.), The psychology of action: Linking cognition and motivation to behavior (pp. 457– 481). New York: Guilford Press. Bargh, J. A., Chaiken, S., Govender, R., & Pratto, F. (1992). The generality of the automatic attitude activation effect. Journal of Personality and Social Psychology, 62, 893–912. Bargh, J. A., & Chartrand, T. (2000). Studying the mind in the middle: A practical guide to priming and automaticity research. In H. Reis & C.

Judd (Eds.), Handbook of research methods in social psychology (pp. 253–285). New York: Cambridge University Press. Bargh, J. A., Chen, M., & Burrows, L. (1996). The automaticity of social behavior: Direct effects of trait concept and stereotype activation on action. Journal of Personality and Social Psychology, 71, 230 –244. Bargh, J. A., Gollwitzer, P. M., Lee-Chai, A., Barndollar, K., & Troetschel, R. (2001). The automated will: Nonconscious activation and pursuit of behavioral goals. Journal of Personality and Social Psychology, 81, 1014 –1027. Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182. Baumeister, R. F., Bratslavsky, E., Muraven, M., & Tice, D. M. (1998). Ego depletion: Is the active self a limited resource? Journal of Personality and Social Psychology, 74, 1252–1265. Berscheid, E., Snyder, M., & Omoto, A. M. (1989). Issues in studying close relationships: Conceptualizing and measuring closeness. In C. Hendrick (Ed.), Review of personality and social psychology (Vol. 10, pp. 63–91). Newbury Park, CA: Sage. Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York: Basic Books. Brewer, M. B., & Gardner, W. (1996). Who is this “we”? Levels of collective identity and self representations. Journal of Personality and Social Psychology, 71, 83–93. Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior. Cambridge, England: Cambridge University Press. Chartrand, T. L. (2001, June). Consequences of succeeding and failing at nonconscious goals. Paper presented at the meeting of the American Psychological Society, Toronto, Ontario, Canada. Chartrand, T. L., & Bargh, J. A. (1996). Automatic activation of impression formation and memorization goals: Nonconscious goal priming reproduces effects of explicit task instructions. Journal of Personality and Social Psychology, 71, 464 – 478. Cooley, C. H. (1964). Human nature and the social order. New York: Schocken Books. (Original work published 1902) Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry, 11, 227–268. Dijksterhuis, A., & van Knippenberg, A. (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. Fazio, R. H. (1990). A practical guide to the use of response latency in social psychological research. In C. Hendrick & M. S. Clark (Eds.), Research methods in personality and social psychology: Review of personality and social psychology (Vol. 11, pp. 74 –97). Newbury Park, CA: Sage. Fitzsimmons, G. M., & Bargh, J. A. (2003). Thinking of you: Nonconscious pursuit of interpersonal goals associated with relationship partners. Journal of Personality and Social Psychology, 84, 148 –163. Freud, A. (1937). The ego and the mechanisms of defense. New York: International Universities Press. Freud, S. (1958). The dynamics of transference. In J. Strachey (Ed. and Trans.), The standard edition of the complete psychological works of Sigmund Freud (Vol. 12, pp. 97–108). London: Hogarth Press. (Original work published 1912) Glassman, N. S., & Andersen, S. M. (1999). Activating transference without consciousness: Using significant-other representations to go beyond what is subliminally given. Journal of Personality and Social Psychology, 77, 1146 –1162. Gollwitzer, P. M., & Bayer, U. (1999). Deliberative versus implemental mindsets in the control of action. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 403– 422). New York: Guilford Press.

GOAL PRIMING THROUGH OTHERS Gollwitzer, P. M., & Moskovitz, G. B. (1996). Goal effects on action and cognition. In E. T. Higgins & A . W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 361–399). New York: Guilford Press. Greenwald, A. G. (1992). New look 3: Unconscious cognition reclaimed. American Psychologist, 47, 766 –779. Greenwald, A. G., & Banaji, M. R. (1989). The self as a memory system: Powerful, but ordinary. Journal of Personality and Social Psychology, 57, 41–54. Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319 –340. Higgins, E. T. (1996). Knowledge activation, accessibility, applicability, and salience. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 133–168). New York: Guilford Press. Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52, 1280 –1300. Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational principle. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 30, pp. 1– 46). New York: Academic Press. Higgins, E. T. (1999). Self-discrepancy: A theory relating self and affect. In R. F. Baumeister (Ed.), The self in social psychology (pp. 150 –181). Philadelphia: Psychology Press. Higgins, E. T., Shah, J. Y., & Friedman, R. (1997). Emotional responses to goal attainment: Strength of regulatory focus as moderator. Journal of Personality and Social Psychology, 72, 515–525. Hinkley, K., & Andersen, S. M. (1996). The working self-concept in transference: Significant-other activation and self change. Journal of Personality and Social Psychology, 71, 1279 –1295. Horney, K. (1946). Our inner conflicts: A constructive theory of neurosis. London: Routledge & Kegan Paul. Huston, T. L., & Chorost, A. F. (1994). Behavioral buffers on the effect of negativity on marital satisfaction: A longitudinal study. Personal Relationships, 1, 223–239. Kelley, H. H. (1952). Two functions of reference groups. In G. E. Swanson, T. M. Newcomb, & E. L. Hartley (Eds.), Readings in social psychology (2nd ed., pp. 410 – 420). New York: Holt, Rinehart & Winston. Kruglanski, A. W. (1996) Goals as knowledge structures. In P. M. Gollwitzer & J. A. Bargh (Eds.), The psychology of action: Linking cognition and motivation to behavior (pp. 599 – 619). New York: Guilford Press. Kruglanski, A. W., Thompson, E. P., Higgins, E. T., Atash, M. N., Pierro, A., Shah, J. Y., & Spiegel, S. (2000). To “do the right thing” or to “just do it”: Locomotion and assessment as distinct self-regulatory imperatives. Journal of Personality and Social Psychology, 79, 793– 815. Kuhl, J., & Beckman, J. (Eds.). (1994). Volition and personality. Toronto, Canada: Hogrefe. Markus, H., & Cross, S. (1990). The interpersonal self. In L. A. Pervin (Ed.), Handbook of personality: Theory and research (pp. 576 – 608). New York: Guilford Press. Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224 – 253. Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press. Mikulincer, M. (1998). Attachment working models and the sense of trust: An exploration of interaction goals and affect regulation. Journal of Personality and Social Psychology, 74, 1209 –1224.

681

Moretti, M. M., & Higgins, E. T. (1999a). Internal representations of others in self-regulation: A new look at a classic issue. Social Cognition, 17, 186 –208. Moretti, M. M., & Higgins, E. T. (1999b). Own versus other standpoints in self-regulation: Developmental antecedents and functional consequences. Review of General Psychology, 3, 188 –223. Perdue, C. W., Dovidio, J. F., Gurtman, M. B., & Tyler, R. B. (1990). Us and them: Social categorization and the process of intergroup bias. Journal of Personality and Social Psychology, 59, 475– 486. Radvansky, G. A. (1999). The fan effect: A tale of two theories. Journal of Experimental Psychology: General, 128, 198 –206. Read, S. J., & Miller, L. C. (1989). Inter-personalism: Toward a goal-based theory of persons in relationships. In L. A. Pervin (Ed.), Goal concepts in personality and social psychology (pp. 413– 472). Hillsdale, NJ: Erlbaum. Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25, 54 – 67. Schaffer, R. (1968). Aspects of internalization. New York: International Universities Press. Shah, J. Y. (2002). The motivational looking glass: How significant other may implicitly influence how goals are construed, pursued and experienced. Manuscript submitted for publication. Shah, J. Y., Friedman, R., & Kruglanski, A. W. (2002). Forgetting all else: On the antecedents and consequences of goal shielding. Journal of Personality and Social Psychology, 83, 1261–1280. Shah, J. Y., Higgins. E. T., & Friedman, R. F. (1998). Performance incentives and means: How regulatory focus influences goal attainment. Journal of Personality and Social Psychology, 74, 285–293. Shah, J. Y., & Kruglanski, A. W. (2000). Aspects of goal networks: Implications for self-regulation. In M. Boekaerts, P. R. Pintrich, & M. Zeidner (Eds.), Handbook of self regulation (pp. 85–110). San Diego, CA: Academic Press. Shah, J. Y., & Kruglanski, A. W. (2002). Priming against your will: How goal pursuit is affected by accessible alternatives. Journal of Experimental Social Psychology, 38, 368 –383. Shah, J. Y., & Kruglanski, A. W. (in press). When opportunity knocks: Bottom-up priming of goals by means and its effects on self regulation. Journal of Personality and Social Psychology. Shah, J. Y., Kruglanski, A. W., & Friedman, R. (2002). A goal systems approach to self-regulation. In M. Zanna, J. Olson, & C. Seligman (Eds.), The Ontario Symposium on personality and social psychology (pp. 247–276). Mahwah, NJ: Erlbaum. Sherif, M. (1948). An outline of psychology. New York: Harper. Sullivan, H. S. (1940). Conceptions of modern psychiatry. New York: Norton. Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: Norton. Thomsen, C. J., Lavine, H., & Kounios, J. (1996). Social value and attitude concepts in semantic memory: Relational structure, concept strength, and the fan effect. Social Cognition, 14, 191–225. Triplett, N. (1898). Dynamogenic factors in pacemaking and competition. American Journal of Psychology, 9, 507–533.

Received April 1, 2002 Revision received October 14, 2002 Accepted October 30, 2002 䡲