Abstract
An accurate assessment of an individual often requires taking their potential into account. Across six studies the authors found that people are more inclined to do so when evaluating themselves than when evaluating others, such that people credit themselves for their perceived potential more than they credit others for theirs. Participants rated potential as a more telling component of the self than of others, and the importance participants placed on their own potential led to attentional biases toward information about their own future potential that did not apply to information about the potential of others. Furthermore, when assessing themselves and other people, participants required more tangible proof that someone else has a given level of potential than they required of themselves, and they relied more on how they would ideally perform in self-assessment but more on how others actually performed in judging them.
We judge ourselves by what we feel capable of doing, while others judge us by what we have already done.
In this article, we subject Longfellow’s contention to empirical scrutiny. We ask whether people give greater weight and prominence to future potential in their conception and evaluation of the self than in their thinking about other people. Toward that end, we examine whether people pay more attention to information about their potential, allow feedback about potential to have more impact on their self-assessments, and believe potential to be a more important issue with which the self should be concerned. When it comes to other people, we assert, much like Longfellow did, that social perceivers instead focus more on already revealed levels of performance and achievement, with relatively less attention paid to the potential of others to improve.
Although we are not aware of any work that examines Longfellow’s contention directly, his claim is consistent with a host of findings in the social psychological literature. Three strands of research indirectly suggest that people might highlight potential in themselves more than they do in others. First, people give considerable weight to their intentions when assessing where they stand on various traits and abilities but judge others more strictly in terms of observed behavior (Kruger & Gilovich, 2004). This tendency is likely exacerbated by people’s belief that mental evidence, such as intentions, expectations, and counterfactuals, is a more valid reflection of who they truly are and what they are capable of doing than it is for other people, who are better understood in terms of their overt behavior (e.g., Andersen & Ross, 1984; Koehler & Poon, 2006; Pronin, 2009). Second, past work shows that the “true self” is a different blend of past, current, and future selves when it comes to oneself than when it comes to others (Williams & Gilovich, 2008). People tend to believe that for someone else to know who they “really are,” it is necessary for that person to know where they are headed and who they plan to be in the future, but that information is less essential to know about others. Because one’s potential is something to be realized in the future, this tendency to weight the future more heavily in self-assessments than in assessments of others should give rise to a tendency for potential to be a bigger part of the self than someone else. Finally, people give more weight to peak performances, those presumably more diagnostic of potential, when evaluating themselves, and more weight to typical or average performances when evaluating someone else (Williams & Gilovich, 2011).
Why might people emphasize potential more in self- than social judgment? There are a number of reasons. One is simply that people have more information relevant to their own potential than they do for the potential of others (e.g., Chambers, Epley, Savitsky, & Windschitl, 2008; Pronin, 2008). People know their own desires and intentions, they know their goals and aspirations (Andersen & Ross, 1984; Johnson, 2009; Pronin, Berger, & Molouki, 2007; Pronin & Kugler, 2007), and they are aware of situational constraints and random events that might have prevented them from realizing their potential in the past.
In addition, they may view their own behavior as a contest between their efforts to succeed and various obstacles thwarting their efforts, such as ineffective strategies (Anderson & Jennings, 1980), situational circumstances (e.g., Dillon & Tinsley, 2008), or pure bad luck. Of course, suboptimal strategies can be changed and unfortunate circumstances are often happenstance events that are likely to be absent in the future or avoided with proper foresight. This may lead people to view their potential performance, in the terminology of measurement theory, as their “true score,” and their actual performance as “true score plus error.” Thus, like political forecasters who cite “close call” counterfactuals to maintain their faith in their predictive acumen and overall worldview in the face of repeated failures to predict world events (Tetlock, 2005), people are likely to seize on mitigating circumstances to explain away their failure to meet their intentions or realize their potential. This can leave them with the sense that their potential self is the truest, least tainted representation of who they are. In contrast, people are much less knowledgeable about others’ intentions, how hard they have tried to fulfill them, the strategies they have adopted to succeed, and any events or circumstances that interfered with success. Without such knowledge, another person’s performance is often taken at face value—as an accurate reflection of the individual in question, not as an incomplete or misleading representation of what he or she is capable of achieving.
Beyond this, potential may feature in people’s self-concepts more because it is more pertinent to the goals they pursue in everyday life. One pervasive human motivation is self-improvement. People want to enhance their skills and enlarge their expertise (e.g., Sedikides, 2009), so it is a safe bet that people are generally more concerned about improving themselves than others. Can I improve? is a question people ask themselves much more often than Can Person X improve? This leads people to pay attention to their potential because it provides information relevant to the personal goal of improvement. If one believes that one has potential in mathematics, that belief is informative about how much one might improve on math tasks or whether math is a skill to which improvement efforts should be devoted. However, people harbor different goals regarding others that could cause them to deemphasize potential in favor of knowledge about others’ past and present abilities. Put simply, people just want to know what others are like, to know how to act around them in the here and now. People generally are not as concerned about possible improvements others might experience, except in special cases such as when the others in question are their children, students in their classroom, or players on the team they are coaching.
Finally, people may emphasize potential more in their self-concepts than in their conceptions of others because they are more motivated to think highly of themselves than of other people (e.g., Kunda, 1990). Thinking about one’s potential can be quite pleasant, and believing that one has potential allows for congenial conclusions about one’s ability and expertise. Thus, the motive toward self-enhancement would exacerbate the pronounced asymmetry in the information available about the self and others, as it leads people to seek out and seize on information about their own potential but not that of others.
Overview of Studies
We report the results of six studies designed to assess whether people consider future potential to be a bigger part of the self than of others. The first four studies ask whether future potential looms larger in people’s self-evaluations than it does in evaluations of others. In Study 1, students specified the extent to which past accomplishments, present talents, and future potential need to be considered to get an accurate sense of what they are really like as students, or what another student is really like. We predicted that participants would rate future potential as more important when it comes to an accurate understanding of themselves than when it comes to an accurate understanding of another student.
In Study 2, we investigated a notable implication of this self–other difference in the weight assigned to future potential. If people receive more credit than others for future potential, they should require more tangible evidence of someone else’s potential before they are willing to grant that person the same level of potential they grant themselves. And if others have to “show” more to be granted a given level of potential, people should judge someone else with the same potential as themselves as being closer to realizing that ultimate potential than they themselves are. In Study 3, we tested another possible consequence of a greater focus on potential for the self than for others, that people may be more likely to label themselves as underachievers than others. Because people see themselves as having unattained potential, they may believe they have failed to accomplish everything they can. Others, however, will seem to more closely match their optimal level of output. Finally, in Study 4 we examined whether this self–other difference in the weighting of potential can lead to the otherwise paradoxical result that people in athletic competitions can think they are “better” than opponents who consistently defeat them.
The last two studies explored the extent to which people care about and give weight to feedback about their potential. In Study 5, we used a dot probe paradigm to examine whether people pay more attention to feedback about their potential than to feedback about their actual performance but show the opposite pattern when evaluating feedback about someone else. In Study 6, we explored whether feedback about potential would prompt people to revise their own self-impressions more than it would their impressions of others.
Study 1
To begin our investigation, we approached the question in the most direct way possible, by simply asking participants how much past performance, current talents, and future potential should factor into an adequate understanding of themselves or another person. We predicted that future potential would be more important for participants assessing the importance of these three types of information for understanding the self rather than another person.
Method
Participants
Eighty-six Cornell undergraduates (60 female, 26 male) completed this study for extra credit in psychology classes.
Procedure
Participants in the self condition received a questionnaire that asked what balance of information another person would need to know to have an accurate understanding of their ability as a student. More specifically, they read:
Suppose that you have been asked to describe yourself as a student to another person, and you are supposed to give them a description that most accurately represents who you are academically. This description can consist of three different aspects of yourself: your past accomplishments, your present talents, and your future potential as a student. How much weight do you think is appropriate to give each of these aspects when describing yourself, i.e., what portion of the whole picture of you as a student would give the other person the truest picture of who you are?
To indicate how much each component should be weighed, participants assigned percentages to their past accomplishments, present talents, and future potential as a student, with these percentages summing to 100%.
Participants in the other condition considered making the same assessment of John, a sophomore at Cornell. Following the same instructions above, they responded to: “How much weight do you think is appropriate for John to give each of [the three] aspects when describing himself . . . ?”
Results and Discussion
Gender had no significant influence on the results of this or any of the following studies and is not discussed further.
As predicted, future potential seemed more important for an adequate understanding of the self (M = 35.8%, SD = 15.1) than an adequate understanding of another student, John (M = 28.9%, SD = 14.4), t(84) = 2.19, p < .05, d = .48. Because participants’ ratings of the importance of past performance, present ability, and future potential must add to 100%, they are not independent, and significance tests performed on participants’ assessments of the importance of past performance and present ability would draw on the effect we just documented. Nevertheless, at a purely descriptive level it is worth noting that participants did not differ in how important they thought it would be for someone to know their own or John’s past performance (Ms = 30.8% and 29.2%, respectively). But they thought that knowledge of present ability was more important in understanding the kind of student John is (M = 41.7%) than the kind of student they themselves are (M = 33.6%). Thus, consistent with our thesis, people appear to believe that future potential is a more important element of their own standing than another person’s.
Study 2
Study 2 provided a second test of whether potential looms larger in people’s self-conceptions than in their conceptions of others. If others are evaluated mainly in terms of their actual performance, whereas the self receives considerable “credit” for intentions and future potential, it follows that others must show more in terms of actual performance than the self to be granted any given level of potential.
We tested this prediction by having participants indicate how far along toward reaching their potential they and someone else with the same ultimate ability as themselves were at the moment. We hypothesized that participants would indicate they are farther from reaching their potential than others with the same potential. Other people must exhibit more concrete evidence that they have a certain level of potential to be granted it, and thus others should appear to be closer than the self to reaching their potential.
Method
Participants
Fifty-seven Cornell undergraduates completed this questionnaire in exchange for extra credit in psychology and human development courses.
Procedure
The questionnaire that participants completed began with the following:
The phrase “life is a journey” is a clichéd but useful way to describe the idea that life is a process of growth and development. We all have an obvious starting point, as well as a destination, the point at which we have become the best person we can be. Think about yourself. Where are you on this journey? How close do you think you are to being the best person you could ever be?
Below this paragraph was a 151-mm line with the endpoints, just starting out and at your destination. Participants were told to make a slash mark on the line at the point at which they thought they fell. After doing so, they thought of a specific classmate with the same overall potential as they thought they themselves had. This was to be someone who “may have different goals in life, a different career path, different wants and desires, and so on. But after accounting for all that, their potential level of accomplishment and fulfillment in life is, overall, the same as” their own. Participants wrote down this person’s initials, to guarantee that they were thinking of a specific person, and then made a slash mark on another 151-mm line labeled with the same endpoints to indicate how close the other person was to reaching his or her potential.
Results and Discussion
Consistent with our hypothesis, participants saw themselves as significantly farther from reaching their full potential (M = 55.6 mm, SD = 27.0) than their acquaintance with the same overall level of ability (M = 63.5 mm, SD = 30.8), paired t(56) = 2.94, p < .01, d = .78. To be granted the same level of potential as the self, other people have to show more concrete evidence that they indeed have that potential. This reinforces our overall thesis that potential is weighted more heavily in self-assessments than in assessments of others, whereas actual performance is weighted more heavily in assessments of others than in assessments of the self.
Study 3
One notable implication of this finding is that if unattained potential is a bigger component of self-evaluation, people are likely to see themselves as failing to have accomplished things they should be able to accomplish, whereas others will seem to better match their supposed level of achievement. In Study 3, we examined whether people would admit to these shortcomings in themselves—seeing themselves as underachievers, in essence—while viewing others as better matching what they should accomplish.
Method
Participants
Sixty-six adults (42 female, 24 male) were recruited via Amazon.com’s Mechanical Turk service, participating in exchange for $.15 in Amazon.com credit.
Procedure
The questionnaire for participants in the self condition began as follows:
Many people’s skill levels and output levels don’t match up. Some people accomplish much less than their true potential suggests they should be able to, and some people accomplish much more.
Participants considered how this might apply to themselves. Were they an underachiever, such that they should be able to accomplish more than they have to date; or an overachiever, such that they have accomplished more than they should have been able to? To clarify what we meant by under- and overachievement, they were reminded that:
. . . in this survey, under- and overachieving is not relative to others, it’s relative to yourself and your own potential. For example, you may be an overachiever relative to others, accomplishing more than your peers are able to, but an underachiever relative to yourself, in that you can accomplish even more than you have already.
Participants responded in two ways. First, they indicated on a continuous scale how they would describe themselves on a scale from 1 (I’m an underachiever: I have potential beyond what I’ve managed to achieve thus far) to 4 (I’m an “achiever”: My accomplishments so far are exactly what you would expect from my level of potential) to 7 (I’m an overachiever: I have achieved more than I should be able to, given my true potential). Participants also responded to “if you had to categorize yourself as one or the other, which one would best describe you,” by checking one of three boxes corresponding to the anchor points in the previous question.
Participants in the other condition first thought of an acquaintance of their gender and approximate age and reported this person’s initials, and then indicated whether that person is an under- or overachiever relative to his or her potential, using both measures.
Results and Discussion
Participants were more likely to see themselves as underachievers than their acquaintances. Their self-ratings were lower on the underachiever–overachiever scale (M = 3.4, SD = 1.5) than their ratings of their acquaintance (M = 4.3, SD = 1.7), t(64) = 2.29, p < .05, d = .57. The average self-rating was significantly below the midpoint of the scale, t(33) = 2.56, p < .05, d = .44, whereas the average rating of the acquaintance did not differ from the midpoint, t(31) = .83, ns, d = .15. In addition, participants were more likely to definitively categorize themselves as underachievers than their acquaintances: Fifty-six percent of participants categorized themselves in this way, whereas only 28% categorized their acquaintances similarly, χ2(1, N = 66) = 5.20, p < .05, φ = .28.
Study 3 indicates that people see a sizable gap between what they have accomplished and what they should be able to accomplish but see more of a match between other people’s accomplishments and potential. They believe in this gap strongly enough that they are willing to apply to themselves a fairly negative label—underachiever—while labeling their acquaintances in more neutral terms.
Study 4
Study 4 was a final test of whether people feature potential more in self- than in social impressions. The tighter relation between actual performance and overall evaluation in assessments of others leads to the seemingly paradoxical possibility that individuals might see themselves as more talented than those who have consistently outperformed them. After all, if the self receives more credit for future potential than others do, the extra edge provided by including potential in self-assessments can make up for a history of coming up short in the past and present. Study 4 examined this possible mismatch between past performance and perceived ability on the part of the self and others.
We suspect that it is a rare reader indeed who cannot think of an instance of being outperformed by someone of seemingly inferior ability. One domain in which this seems especially common is the world of sports. Nearly everyone can recall instances in which their athletic performance failed to live up to their perceived potential. But will people maintain that they are better than someone else at a particular sport even if that person consistently outperforms them? In Study 4, we asked a sample of tennis players whether they could think of any players who generally beat them who, nevertheless, were inferior tennis players. We also asked whether they could think of any players they generally beat who, nevertheless, were superior tennis players. We predicted that our respondents would recall significantly more instances of the former than the latter: Other players may typically beat the self, but the self can still be considered a better player when potential is taken into account.
Method
Participants
Thirty-four tennis players in Ithaca, New York, volunteered to participate.
Procedure
Participants were approached at tennis courts on or near the Cornell University campus and asked if they were willing to answer a series of simple questions about their tennis history. Those who agreed to participate were asked the following questions: “Think about the people you have played tennis against. Can you think of someone who has beaten you more times than you have beaten them, but you still believe overall you are a better tennis player than them?” and “Think about the people you have played tennis against. Can you think of someone you have beaten more times than they have beaten you, but you still believe that overall they are a better tennis player than you?” The order of the two questions was counterbalanced. Participants responded to each question by reporting the initials of the particular player they had in mind. After each response, participants were asked whether there were any other players who fit that description until they could not think of anyone else. They were then asked the other question, in the same fashion.
After answering these questions, participants indicated whether they knew their U.S. Tennis Association (USTA) rating (a structured but self-generated assessment of tennis skills and experience used to equate players for tournament play) and what it was if they knew. If they did not know, they rated their skills on a 1 to 9 scale, with 1 corresponding to a novice player and 9 corresponding to high school varsity level or better. They were then thanked and debriefed.
Results and Discussion
As predicted, our sample of tennis players was able to think of more players who regularly beat them despite being inferior players (M = 3.0, SD = 1.9) than players who regularly lost to them despite being better players (M = 2.0, SD = 1.6), paired t(33) = 3.65, p < .001, d = .57.
Was this trend more pronounced among players with more positive self-assessments? The two measures of respondents’ self-rated skill level that we used are mutually exclusive: respondents only rated their skills on the 9-point scale if they did not know their USTA rating. We therefore examined the relation between skill level and the effect reported above for two separate subsets of our respondents. Despite the small sample in each analysis, there was a notable positive relation between self-assessed skill level and the difference between the two types of opponents recalled. Respondents who rated themselves as having greater skill were particularly likely to think of more “inferior” players who regularly outperformed them than “superior” players they regularly beat, rUSTA(16) = .41, p < .12; r1-9 rating (18) = .47, p = .05, with combined result, Stouffer’s Z = 2.68, p < .01.
These findings only make sense if (and thus further support our contention that) people give themselves more credit for their potential than they are willing to credit others for theirs: Because their own potential is factored into who they are, but others are judged more exclusively on their actual performance, it can seem sensible to the individuals involved to maintain that they are better players than opponents who regularly beat them. This tendency, furthermore, is likely to be especially pronounced for people who rate their skills highly because that same tendency to credit themselves for their potential is likely to lead them to report higher levels of skill than their current abilities alone would warrant.
Study 5
In light of the self–other differences documented in the first four studies, we might expect people to be differentially attentive to information about actual performance versus future potential when that information pertains to the self than to someone else. People might be more interested in finding out about their own future potential than their actual current performance. At the same time, they might be more interested in finding out about someone else’s actual current performance than that person’s future potential.
In a test of this idea, participants completed several rounds of a word search game, after which they received feedback about their actual ability and their potential relative to others, and the same information about another participant to whom they had supposedly been yoked. They then completed a dot probe task (see Mather & Carstensen, 2003) in which this feedback was presented just before the dot appeared. If participants pay more attention to their own potential than their actual performance, they respond faster to the dot probe when appears in the same place where the feedback about their potential had been than when it appears where the feedback about their actual performance had been. They should exhibit no such asymmetry when it comes to information about the actual performance and future potential of their yoked counterpart.
Method
Participants
One hundred forty-five students at the University of Florida participated in exchange for extra credit in introductory marketing and business classes. One participant was unable to complete the dot probe task because of computer error, and another participant expressed suspicion about actually being yoked to another participant. Their data were eliminated from the analyses, as were the data from 12 others who failed to follow instructions, as indicated by extremely slow reaction times (greater than 3 SD beyond the mean; 3 participants) or high numbers of errors (greater than 40%; 8 participants), or simply failing to complete the task to the end (1 participant), leaving a final sample of 131 participants.
Procedure
Participants completed this experiment in groups of 7 to 24, with each participant seated in an individual computer cubicle. The first part of the experiment consisted of a word search game, similar to the game Boggle. On each of five rounds, participants were presented with a 5 × 5 grid of random letters. Their task was to find words of three or more letters, with each letter in a word contiguous to one another and no individual letter used more than once in each word. They had 2 min to find as many words as they could in each round.
To increase participants’ motivation to do well and pay attention to the feedback they received after all five rounds, they were told that the task was often used to assess students’ verbal abilities, which can “determine a person’s academic and professional future.” They also learned that they would be competing with another participant at the task, whose final scores would be compared to their own; they would not receive information about how they or the other participant were doing until they completed all five rounds. (In reality, there was no other person, and the feedback they received about their own this “person’s” performance was the same for all participants, as was their own.) Participants also learned that the completion of multiple rounds would allow us to track variation in their performance and enable us to “calculate . . . what [their verbal ability] may be in the future if [they] were to train, practice, or grow—in other words, what [their] true potential is.” This bit of fiction made it possible for us to later present information about their potential verbal facility.
Once they completed all five rounds of the game, roughly half of the participants (n = 78) received feedback on the computer indicating that they scored at the 71st percentile on the test and could potentially score at the 93rd percentile with “further skill building.” The computer also indicated that their yoked counterpart scored at the 67th percentile and could potentially score at the 88th percentile with further skill building. The remaining participants (n = 67) received the complementary pattern of feedback—that is, that the self’s actual and potential scores were at the 67th and 88th percentiles, respectively, and the other person’s were at the 71st and 93rd percentiles. 1 The feedback stayed on the screen for at least 60 s; after that, participants could press the space bar to proceed to the next part of the experiment.
Participants then completed 24 trials of the dot probe task. On each trial: (a) an asterisk appeared in the center of the screen for 500 ms as a fixation point; (b) two of the four pieces of feedback (self-actual, self-potential, other-actual, other-potential) were displayed for 750 ms, one on each side of the screen; (c) the feedback was erased from the screen; (d) a black dot immediately appeared on one side of the screen in the previous location of one of the two pieces of feedback; and (e) the dot remained until the participant pressed either of the two response keys. 2 Participants indicated, as quickly as possible, where on the screen the dot appeared, pressing the C key if it appeared on the left side and the M key if it appeared on the right. They were also told that before the dot appeared, portions of the feedback they received earlier would appear and disappear on each side of the screen. They had four practice trials to get used to the task.
Results
If participants pay more attention to feedback about their potential than feedback about their actual performance, they should be faster to indicate where the dot appeared if it appeared where the information about their potential had been than if it appeared where the information about their actual performance had been. Likewise, if people pay more attention to feedback about another person’s actual performance than their potential, this pattern should be eliminated or reversed. Participants’ reaction times on trials in which both types of information about the self or about the other person were presented were entered in a repeated measures ANOVA with self or other as one factor and actual or potential feedback as the second. 3 This analysis revealed no main effects but a significant interaction between the two factors, F(1, 128) = 15.49, p < .001, η2 = .11.
Specifically, participants were significantly faster to react to the dot probe when it appeared where information about their own potential had been (M = 334 ms, SD = 88) than when it appeared where information about their actual performance had been (M = 360 ms, SD = 133), paired t(128) = 3.51, p = .001, d = .44. In contrast, participants were slower to react to the dot when it appeared where the other person’s potential feedback had appeared (M = 368 ms, SD = 223) than when it appeared where information about the other person’s actual performance had been (M = 346 ms, SD = 166), paired t(128) = −2.10, p < .05, d = .27 (see Figure 1).
Mean reaction times (in milliseconds) on the dot-probe task
Discussion
These results reinforce our contention that people are much more attuned to their potential than the potential of others. When the dot appeared where information about their own potential performance had been, they were faster to specify its location—indicating that their attention had been in that spot all along. In contrast, they were faster to specify the dot’s location when it appeared where information about someone else’s actual performance had been. This asymmetry in the amount of attention devoted to potential versus actual performance when it comes to the self versus others supports our claim that people consider potential to be a bigger component of the self than of someone else.
To be sure, people do not ignore potential entirely when evaluating others. Most people acknowledge that others will change and grow and try to develop their capacities, just as they themselves will do. But thoughts of their own potential are heavily influenced, not just by what they have done thus far, but by knowledge of their desires and intentions (Kruger & Gilovich, 2004; Williams & Gilovich, 2008). Their thoughts about the potential of others, in contrast, are based largely on others’ past behavior and on population base rates (e.g., Balcetis, 2009; Epley & Dunning, 2000; Kruger & Gilovich, 2004).
Study 6
If potential is a more meaningful component of the self than of others, we should also find that people are more likely to believe and incorporate feedback about potential into their self-assessments. Our final study simulated such an experience. Participants imagined a scenario in which they or a friend received feedback that they had a considerable amount of potential dancing ability, and then reassessed their or their friend’s standing on that skill. We expected to find that participants would be more likely to shift their subsequent self-assessments up to match the feedback they received than they would be for their friend.
Method
Participants
Seventy-eight Cornell undergraduates (57 female, 21 male) volunteered to fill out this questionnaire as they entered the lab.
Procedure
Participants in the self condition were asked to assume that they were perfectly average dancers among students at Cornell. With that in mind, they read that:
. . . one day you’re with your friends waiting for a table at a restaurant, and, just joking around, you start showing your friends some of your “moves.” After dinner a woman comes up to you, and tells you that she saw you dancing earlier. The woman is the head dance coach at Cornell, and she tells you that, although your dancing is rough right now, if you were to take lessons and practice several times a week, she thinks you have the potential to become one of the best dancers on campus, better than 95% of the other students. You take her card and thank her, and promise to mull over what she said.
Next, they imagined reassessing their dancing skills right after they left the restaurant, and with the feedback in mind, provided a new assessment of where they stand relative to other Cornell students. They did so by making a slash mark on a 151-mm line with the endpoints worse than all Cornell students and better than all Cornell students. Lastly, they reported how likely they were to follow up on the coach’s advice to take lessons and practice to become an excellent dancer, using a 1 (not at all likely) to 7 (extremely likely) scale.
In the other condition, participants imagined that they had a friend named Jamie, whom they considered to be a perfectly average dancer, who received the same feedback. They indicated how they would reassess Jamie’s dancing skills once they learned about the feedback from the instructor and how likely Jamie would be to follow the coach’s advice.
Results
If participants take the feedback into account when making their reassessment, we should see their reassessments shift up from their provided starting point, at “average.” Participants report a shift up from average (50%), both for themselves (M = 101.8 mm, SD = 15.1, equivalent percentile = 67.4%), t(37) = 10.76, p < .001, d = 3.54, and for Jamie (M = 83.9 mm, SD = 19.0, equivalent percentile = 55.5%), t(39) = 2.78, p < .01, d = .89. The shift for the self, though, was significantly greater than that for Jamie, t(76) = 4.60, p < .001, d = 1.06. Even though the starting point and the feedback were exactly the same for both, the flattering feedback about potential was incorporated to a greater extent into participants’ self-assessments than into their assessments of Jamie.
Another indication of how much participants consider potential to be a part of themselves or someone else can be seen in their beliefs about following the coach’s advice to become a better dancer. If people who receive feedback that they have a certain level of potential believe that potential is already a part of them, they may also believe they do not need to work as hard to reach that potential. The data are suggestive of that idea. Participants thought Jamie would be significantly more likely to follow the coach’s advice (M = 5.1, SD = 1.1) than they themselves would be in the same situation (M = 3.9, SD = 1.8), t(57.1) = 3.38, p = .001, d = .89. This is not conclusive evidence in support of our hypothesis, to be sure, but it is a logical extension of the idea that potential is seen as an existing part of the self.
One might argue that participants did not buy into our request to imagine that they were initially perfectly average dancers, and so they merely reported preexisting beliefs about their dancing skills instead of a reassessment incorporating the new feedback. As a result, our findings may reflect simple self-enhancement rather than any tendency for people to incorporate potential more readily into their self-assessments than assessments of others. Note, however, that we chose dancing because we thought most students would harbor doubts and insecurities about their dancing abilities and thus be less prone to an above-average effect in this domain. To examine our suspicion, we pretested Cornell students’ beliefs about their dancing skills. A separate set of 29 Cornell students were recruited in the same way as in the study proper. They assessed their dancing skills relative to the other students at Cornell, by making a slash mark on the same line as in the original study, without the corresponding scenario. We found that our effect does not seem to be due to simple self-enhancement, as Cornell students tend to think they are actually below-average dancers (M = 64.8 mm, SD = 29.9, equivalent percentile = 42.9%), t(28) = −1.93, p = .06, d = .73.
Discussion
When participants imagined themselves or another person receiving feedback about their potential, that feedback seemed to be incorporated into their own self-concepts to a greater extent than it was in their conceptions of another person. This prompted people to report more favorable self-assessments than assessments of others despite equal starting points and equivalent feedback. It also led participants to report a belief that they have less need to work to attain their potential—their potential lies closer to the surface. People’s tendency to see their potential as part of who they are thus extends to the incorporation of feedback about their potential provided by others, not just their own estimates of their ultimate level of ability.
General Discussion
Across six studies, we demonstrated that potential is an important component of people’s self-concepts, and that potential plays a bigger role in people’s self-assessments than in their assessments of others. The first four studies revealed, in converging ways, that potential figures more prominently in people’s self-concepts than it does in impressions of other people. In Study 1, when weighting the importance of past accomplishments, current abilities, and future potential in understanding an individual, participants stated that future potential was more important to an accurate understanding of who they themselves are than who someone else is. Study 2 found that people thought they had more of “a journey” ahead of them to reach their potential than did another person of equal competence and achievement. Study 3 revealed that people see themselves as underachievers, having (so far) failed to achieve their full potential, whereas others seem to better match their true level of capability. Study 4 showed that respondents rate their ability, bolstered by their sense of their future potential, ahead even of people who consistently outperform them. It appears they evaluated themselves in terms of how they might ideally perform but others in terms of past performance—presumably because respondents gave greater weight to as-of-yet unrealized potential in their self-assessments than in assessments of their competitors.
The last two studies showed that people care more about feedback when it relates to their own potential than when it relates to someone else’s potential. Study 5 revealed that people attend more to information about their future potential than to information about the future potential of others. In a dot-probe paradigm, we found that participants focused on information about their own potential but focused on information about another person’s past performance. Finally, Study 6 showed that people use information about their potential to update their assessments of their own ability more readily than they use information about potential to update their assessments of others.
Taken together, this research indicates that people weigh criteria differently when evaluating the self rather than others. People appear to assess themselves based on their expectations of what could or may be, whereas they assess others based on what is or was. That said, it seems highly likely, given people’s well-documented tendency to see themselves in a positive light, especially with regard to intentions and plans (e.g., Epley & Dunning, 2000; Kruger & Gilovich, 2004; Pronin et al., 2007; Sedikides, 1993), that people would not only use information about potential more in their self assessments than in their assessments of others but that they would also see themselves as having more potential, period.
The work presented in this article represents a step beyond any such tendency, showing that even when potential is considered equal between self and other, it still counts more for the self. We agree that the idea that people might see themselves as simply having more potential than others has merit, but we believe that the more interesting question is what this increased potential entails for how people see themselves in the present. It may not be people’s thoughts about how much potential they ultimately have that drives their immediate behavior, but how readily attainable they consider that potential to be and what it reflects about the present self that is decisive.
The fact that people understand themselves and others according to different standards should not be taken to indicate that people are mistaken in doing so. Indeed, it may be wise to downplay, and sometimes even ignore, statements of future plans and intentions by others (Gilbert & Malone, 1995; Pronin, 2008). We know for sure what our own intentions and aspirations are, how much we yearn to bring them to fruition, and what it might take to do so. We can only guess at the content of such thoughts and yearnings of others. The intentions of others are more of an unknown entity, and it is only prudent to treat them more conservatively. Even in cases in which the substance of someone else’s potential is well known, as in Study 6, it may be sensible to doubt whether that person intends to pursue that potential, or how seriously he or she took the feedback, and to be concerned about any number of mental states that may intercede between the present reality and the desired future.
Boundary Conditions and Underlying Mechanisms
To fully understand the influence of potential on assessments of self and others, it is necessary to consider why and when we might find differences in the degree to which people consider potential when assessing self and others. There are two plausible explanations for why people might believe that their own potential is more informative than others’: (a) people find it desirable to see their yet-unattained potential as part of who they are but have less of a stake in seeing the same in others, and (b) people have greater access to evidence supporting the importance of their own potential than others’. It is easy to see why people might be motivated to factor in potential more for themselves than for others. Seeing potential as part of oneself is a simple way to maintain a positive self-impression. For example, in Study 4, tennis players saw themselves as better players than their record of play would suggest, allowing them to make more generous self-assessments than they could otherwise make based strictly on past and present performances.
Although the desire to view oneself favorably is surely important, we believe that the results of several of our studies speak against the idea that this effect is purely motivational in nature. In Study 3, participants ascribed the negative label of underachiever to themselves, admitting they had failed to accomplish as much as they should have been able to. If participants simply wanted to see themselves in the most positive light, they could have instead labeled themselves as having extracted the exact amount their abilities would allow.
Furthermore, the results of Study 5 suggest that people find their potential more informative and relevant even before any reasoned or motivated process has a chance to kick in: Participants paid more, and faster, attention to information relevant to their own potential performance than to information about their own actual performance—and did just the opposite for feedback about others. Indeed, previous research has shown that the evidence people bring to mind most quickly and most assuredly when they evaluate themselves is often quite different from the evidence they consider when evaluating others, such that information pertaining to potential (intentions, plans, and desires) is so much more readily available about the self than others (e.g., Andersen & Ross, 1984; Koehler & Poon, 2006; Pronin, 2009).
Beyond any motivated or cognitive reasons why we might find a self–other difference in reliance on potential, there are circumstances under which we may be more or less likely to find this difference. Age is one such factor. The amount of potential one has, and the quality of that potential, will change substantially over the course of a lifetime. As one gets older, time erodes the extent to which one has potential to attain, and the quality of that potential is likely to be less flattering. It seems plausible that the self–other difference might diminish or even reverse with age, as people are less inclined to see their likely decline (i.e., diminished potential) as representative of who they are. We have preliminary evidence regarding this idea. Eighty-seven Mechanical Turk users age 50 and up completed a general version of Study 1, in which they apportioned their own or a matched-age acquaintance’s past, present, and potential into an overall description of the target person’s life. We replicated the self–other difference from Study 1, such that our participants placed more weight on their own future potential (M = 24.0%, SD = 11.9) than their acquaintance’s (M = 18.9%, SD = 9.5), t(85) = 2.20, p < .05, d = .47. However, in the self condition, there was a negative correlation between participant age and how heavily the future was weighed, r(45) = –.31, p < .05, but no such correlation in the other condition, r(42) = .07, ns. This suggests that as people grow older, potential becomes a less important barometer of their own ability, whereas its perceived informativeness about the abilities of others changes much less over the same period.
Implications
If people believe that potential is central to their own identity, but more peripheral in others, they may also have different standards for what they consider to be acceptable methods for bringing potential to the fore. For example, research on attitudes about pharmaceuticals that enhance rather than heal revealed that people can be more hesitant to consume drugs that promise to make them “become more than who [they] are” than those that promise to make them “become who [they] are” (Riis, Simmons, & Goodwin, 2008). The more a drug purports to change someone into a being they would not become otherwise, the less moral they believe the use of that drug to be.
But consider our finding that people consider their potential a more natural and prominent component of the self than of other people. Because people believe potential is less reflective of others’ true abilities than the self’s, they may view artificial attempts to improve performance to be less acceptable on the part of others. However, they may think of taking such a drug as something that merely unlocks or reveals their true self and thus feel it is more appropriate for their own use than it is for others (Williams & Steffel, 2011).
Another implication of our findings is that the challenges people are inclined to attempt themselves may differ from what they would advise others to do. As work on the planning fallacy makes clear (e.g., Buehler, Griffin, & Ross, 1994), people’s intentions often swamp knowledge of their past failures, making them more ambitious than might be wise. Sally registers for the triathlon because she believes she can get in shape in the next 3 weeks; Simon takes an extra class this semester because he remembers how diligently he studied at his most industrious and expects that he can be that industrious again. People’s ambitions frequently outstrip their performance and their talents, often because their views about their true talents are tainted by their hopes and ideals.
However, if Sally or Simon were to ask us for advice about whether they should undertake these challenges, our advice might be much more cautious—and appropriate. Just as we tend to broker advice based on others’ perceived stable and enduring qualities (e.g., Schoeneman & Rubanowitz, 1985), we are unlikely to augment our sense of their current abilities with imagined future potential when determining what advice to give. We may not expect people to conquer challenges more difficult than they have proven capable of conquering in the past.
The broadest implication of this work is that people think about the self and others in very different ways. When thinking of the self, one often starts in the future—with intentions, hopes, and expectations—and works backward, interpreting the past and present in this light. People’s theories of their own forward movement or intended behavior can alter their understanding of the past and the present (e.g., Epley & Dunning, 2000; Kruger & Gilovich, 2004; Williams & Gilovich, 2008; Wilson & Ross, 2001). When we think about other people, however, we typically start in the past and work our way forward (Blakemore & Decety, 2001). We have seen throughout this article that people’s assessments of others are rooted in observable behavior in the past and present. To be sure, we sometimes think about what someone else may yet be capable of attaining, but it is not the starting point with which we often begin when thinking about the self.
The divide between self and other is especially apparent when people try to understand the expectations and intentions of others and predict their future behavior. In critiques of Ajzen and colleagues’ work on the theory of planned behavior (e.g., Ajzen & Madden, 1986; Fishbein & Ajzen, 1975), Warshaw and Davis (1985) and Gordon (1990) proposed that the researchers before them had conflated intentions and expectations, at times examining whether their participants intended to do something and at other times examining whether they expected to do something. These critics maintain that people answer the two questions very differently: People who are asked about intentions rely more on knowledge of their plans and their general beliefs and preferences about the behavior, whereas those asked about expectations rely on their past behavior and the surrounding circumstances (Gordon, 1990).
These differences parallel differences in the assessments people make of self and others. Gordon (1990) and others argue that when making self-predictions, people anchor on their intentions to perform a certain behavior and add in circumstances and other knowledge to estimate the likelihood that they will actually do so. But when making predictions about others, people have precious little information about intentions on which to anchor and so they tend to base their predictions on base rates, existing circumstances, and past behavior. Thus, when making predictions for self and others, people may ask themselves two very different questions: “Do I intend to perform the behavior?” versus “Do I expect him or her to perform the behavior?” One question taps into knowledge of intentions and ideals, and the other into past and present behavior—the very asymmetry we have documented here. The result is a view of the self that grants all sorts of potential but a view of others that is limited largely to past actions and accomplishments. People thus think of themselves as all that they can be but think of others in terms of what they have been.
Footnotes
Acknowledgements
We thank Chrissy Carino and Michael Smiles for their help in data collection.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: National Science Foundation Research Grants SES0542486, SES0241638, BCS0745086, and SES0922323.