import pymupdf
import re

doc = pymupdf.open("testPIT.pdf")

out = open("output.txt", "wb")                   # creates a text output
for page in doc:                                 # iterate the document pages
    text = page.get_text().encode("utf8")        # get plain text (is in UTF-8)
    out.write(text)                              # write text of page
    out.write(bytes((12,)))                      # write page delimiter (form feed 0x0C)
out.close()

text = open('output.txt','r')
content = text.read()

content

'A.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\nSocialP sychology 2013; Vol. 44(2):84–94\n© 2013 Hogrefe Publishing\nOriginal Article\nPreference for Other Persons’\nTraits Is Dependent on the\nKind of Social Relationship\nAndrea E. Abele and Susanne Brack\nUniversity of Erlangen-Nuremberg, Erlangen, Germany\nAbstract. Three studies test the proposition that people appreciate others’ traits relative to the kind of relationship they have. Hypotheses\nare derived from the dual-perspective model, according to which the fundamental content dimensions of traits, agency, and communion\nare linked differently to the perspectives of actor versus observer; and that the kind of relationship between both persons is a moderator\nof the perspective-trait link. Study 1 shows that people generally prefer communal traits in others, and that agency traits are chosen less\nfrequently in independence relationships than in unilateral dependence relationships than in mutual dependence relationships. Study 2\nreplicates the communion finding and reveals that – independent of respondents’ own agency and communion – agency traits are chosen\nless frequently in communal relationships than in exchange relationships. Study 3 shows that in both kinds of relationship well-being\ngoals enhance the selection of communion traits, whereas mastery goals enhance the selection of agency traits.\nKeywords: social relationships, dual-perspective model, fundamental dimensions of agency and communion, trait selection\nImagine the people you interact with on an average day. Your\ninteractions may include your partner, your child, a friend, a\ncolleague, a supervisor, a teammate, a cashier in the super-\nmarket, an opponent in a negotiation, a server in a restaurant.\nYou may, for instance, appreciate your supervisor’s decisive-\nness, your friend’s trustworthiness, and the cashier’s punctu-\nality. Your relationship to these persons is different, ranging\nfrom long-time emotional bonds to chance encounters, from\ncloseness to distance, from interdependence to independ-\nence, or from exchange orientation to need orientation.\nThe present paper is concerned with how we appreciate\nspecific characteristics of other persons in dependence on\nthe kind of relationship we have with them. Our general\nhypothesis states that one and the same characteristic may\nbe highly appreciated in one kind of relationship and less\nso in another kind of relationship. We further assume that\nthe distinction of traits into two broad content classes, the\nso-called fundamental dimensions or “Big Two” of agency\nand communion (Abele, Uchronski, Suitner, & Wojciszke,\n2008; Abele & Wojciszke, 2007; Fiske, Cuddy, & Glick,\n2007; Judd, James-Hawkins, Yzerbyt, & Kashima, 2005;\nYbarra et al., 2008), helps to more specifically predict\nwhich traits are appreciated in what kinds of relationships.\nWe derive our hypotheses from the dual-perspective model\nof person perception (DPM; Abele & Wojciszke, 2007;\nWojciszke, Baryla, Parzuchowski, Szymkow, & Abele,\n2011). It states that agency and communion are differen-\ntially linked to the perspectives of actor versus observer or\nrecipient of an action. It also states that the link of agency\nand communion to the actor versus observer perspective is\nmoderated by the specific relationship both persons have\nwith each other. Whereas previous research has revealed\nsupporting evidence for the differential link of agency and\ncommunion to the actor versus observer perspective (Abele\n& Wojciszke, 2007; Wojciszke et al., 2011), research on the\nimpact of different kinds of social relationships on trait ap-\npreciation is still scarce (see Cislak, 2013; Wojciszke &\nAbele, 2008). The present study tests the latter prediction\nof the DPM. We apply reasoning from interdependence\ntheory (see Kelley & Thibaut, 1978; Rusbult, 2000; Rus-\nbult & Van Lange, 1996; Thibaut & Kelley, 1959) and from\nthe differentiation of exchange versus communal relation-\nships (Clark, 1984; Clark & Mills, 1979, 1993). We also\nanalyze whether individual differences in own agency and\ncommunion and individual differences in interaction goals\nmoderate the impact of kind of relationship on preference\nfor others’ traits. We present three studies that test our hy-\npotheses.\nKinds of Social Relationships\nThibaut and Kelley (1959) were among the first to argue\nthat social behavior should be understood by studying the\nfabric of interdependence characterizing social situations.\nDOI: 10.1027/1864-9335/a000138\nSocial Psychology 2013; Vol. 44(2):84–94\n© 2013 Hogrefe Publishing\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0cThey defined interdependence as the strength and quality\nof the effects that interacting individuals exert on one an-\nother in terms of preferences, motives, behavior, and out-\ncomes. Interdependence differs in level and mutuality of\ndependence, in the basis of the dependence, and in the co-\nvariation of interests (Kelley & Thibaut, 1978; Rusbult,\n2000; Rusbult & Van Lange, 2003). We concentrate here\non mutuality of dependence.\nApplied to two of our above examples, the level of in-\nterdependence with the cashier at a supermarket is low,\nsince the person who goes shopping may choose one of\nseveral cashiers to pay the due amount. The mutuality of\ndependence is also low, as the cashier has many clients and\nis not dependent on a single one in order to, say, keep his\nor her job. On the other hand, the level of interdependence\nwith a supervisor is high, since the employee has to fulfill\nthe supervisor’s orders and cannot simply decide to look\nfor another supervisor. There is also a certain degree of\nmutuality of dependence because the supervisor wants the\nwork to be done efficiently by his/her employee.\nAnother relevant distinction of social relationships was\nintroduced by Clark and Mills (Clark, 1984; Clark & Mills,\n1979, 1993), who argue that social relationships are either\nreward-based, something they call exchange relationships;\nor they are needs-based, something they call communal re-\nlationships. In an exchange relationship, benefits are given\nwith the expectation of receiving a benefit in return. Each\nperson is concerned with how much he or she receives in\nexchange for benefiting the other, and how much he or she\nowes the other for the received benefits. Members of a\ncommunal relationship assume that each is concerned\nabout the welfare of each other. They have a positive atti-\ntude toward benefiting the other when a need for the benefit\nexists and the receipt of a benefit does not create a specific\nobligation to return a comparable benefit (Clark & Mills,\n1979). Exchange relationships typically exist between\nstrangers and casual acquaintances as well as in certain\nlong-term arrangements such as business partnerships. In\ncontrast, strong communal relationships are usually limited\nto family members, friends, and romantic partners (Clark\n& Mills, 1993).\nApplied to our above examples, the server is an example\nof an exchange relationship in which one benefit (good ser-\nvice) is expected to be reciprocated by another benefit (a\ngood tip). A parent-child relationship is, of course, a com-\nmunal one in which the parent responds to the child’s need\nand does not expect a direct compensation.\nThe Fundamental Dimensions of\nSocial Perception and the\nDual-Perspective Model\nResearch has revealed thatthere are two basiccontentdimen-\nsions in social perception (the “fundamental dimensions” or\n“Big Two”)thatunderlie judgmentsofvarioustargetslikethe\nself, other persons, groups, etc. (Abele et al., 2008; Abele &\nWojciszke,2007;Fiskeetal.,2007;Judd etal.,2005; Paulhus\n& Trapnell, 2008; Peeters, 1992; Ybarra et al., 2008). Agency\n(also-called competence) comprises competence, goal\nachievement, individuality, and self-assertion and focuses on\nown goals and their achievement. Communion (also called\nwarmth or morality), by contrast, emphasizes the individual\nas a social being and compromises cooperation, morality,\nwarmth and trustworthiness. These two dimensions reflect\nthe “duality of human existence” (Bakan, 1966) and the basic\ngoals of social information processing, namely, forming and\nmaintaining social connections (communion) and pursuing\ngoals and manifesting skills and accomplishments (agency;\nFiske et al., 2007; Ybarra et al., 2008).\nIn addition to being a classification of content, the di-\nmensions of communion and agency can also be distin-\nguished with respect to the social value they have, i.e., their\n“profitability” for other people interacting with the trait\npossessor versus their profitability for the trait possessor\nhim-/herself (Peeters, 2001, 2008; Peeters & Czapinski,\n1990; see also Beauvois & Dubois, 2009). Communion\ntraits tend to be other-profitable because they inform the\nperceiver about attributes of the target that first and fore-\nmost have consequences (positive or negative) for other\npeople interacting with the target. Agency traits tend to be\nself-profitable because they first and foremost have conse-\nquences (positive or negative) for the trait possessor and\nthe pursuit of his or her goals.\nThe DPM (Abele & Wojciszke, 2007; Wojciszke et al.,\n2011) predicts that the interpretation of behavior in terms\nof traits as well as the appreciation and relevance of traits\ndiffer between perspectives. Communion generally is the\nprimary of the two content classes, because a target’s com-\nmunal traits inform the observer about this person’s inten-\ntions and whether it is safe to approach him/her or not. The\nprimacy of communion is revealed in processing speed,\ni.e., communal content is processed faster than agency con-\ntent (Abele & Bruckmüller, 2011; Ybarra, Chan, & Park,\n2001) and in the weighting of others’ traits because people\nplace more weight on communion traits in forming overall\nimpressions of others than on agency traits (Abele & Woj-\nciszke, 2007; Asch, 1946; Brambilla, Rusconi, Sacchi, &\nCherubini, 2011; De Bruin & Van Lange, 1999, 2000; Kel-\nley, 1950; Kenworthy & Tausch, 2008; Scholer & Higgins,\n2008; Wojciszke & Abele, 2008; Wojciszke, Bazinska, &\nJaworski, 1998). When asked to describe themselves, peo-\nple also usually indicate higher communion than agency\n(e.g., Abele, 2003; Uchronski, 2008; Wojciszke et al.,\n2011; Ybarra, Park, Stanik, & Lee, 2012) again indicating\nthe primacy of communion.\nApart from this general primacy of communion, howev-\ner, there are clear differences contingent on the actor-ver-\nsus-observer perspective. Communion content being oth-\ner-profitable is more relevant in the observer than in the\nactor perspective and agency content being self-profitable\nis more relevant in the actor than in the observer perspec-\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\n85\n© 2013 Hogrefe Publishing\nSocial Psychology 2013; Vol. 44(2):84–94\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0ctive. Wojciszke and colleagues (2011), for instance,\nshowed that people base their self-esteem more on their\nagency than their communion, whereas the evaluation of\nothers is based more on their communion than their agency.\nKind of Social Relationship and\nPreference for Others’ Traits\nThere is only limited research concerning the moderating im-\npact the kind of social relationship between actor and observ-\ner/recipient mighthave inascribingor evaluating agencyver-\nsus communion traits. Wojciszke et al. (1998) found that re-\nspondents preferred others’ communal traits when they were\nseeking a person to share a secret with; however, they pre-\nferred agentic traits when they were looking for a good nego-\ntiator. Abele and Wojciszke (2007, study 4) showed that\nagency traits were rated as more important for a close friend\nthan for a more distant acquaintance. Wojciszke and Abele\n(2008) found that liking a supervisor was dependent on this\nsupervisor’s communion qualities when the supervisor had\nonly limited influenceonthe respondent’soutcomes.If,how-\never,the supervisor’s influenceontherespondent’soutcomes\nwas strong, then liking the supervisor was more influenced\nby this person’s agency qualities. Cislak (2013) also analyzed\nobservers’ evaluations of other persons’ agency versus com-\nmunion traits. The respondents were primed with roles or\nmemories that were always connected with having or not\nhaving power. The findings revealed that having power led\nto an increased interest in others’ agency traits, whereas not\nhaving power led to an increased interest in others’ commu-\nnion traits.\nRelationship researchers not concerned with the funda-\nmental dimensions of agency and communion have also\nfound evidence that fits the present reasoning. Cottrell,\nNeuberg, and Li (2007), for instance, analyzed the impor-\ntance of different traits for different kinds of targets (gen-\neralized “ideal person”; “ideal persons” with respect to dif-\nferent social relationships; persons differently interrelated\nwith the judging person). They hypothesized and found that\ntraits like trustworthiness and cooperativeness are general-\nly highly valued – a finding that can be covered under the\npresent heading of “primacy of communion.” Characteris-\ntics such as intelligence were differently valued across\ntasks, goals, and functions. Assuming that “intelligence”\nbelongs to the agency dimensions, this finding is also well\nin line with our prediction that the importance of agency\ntraits varies with the kind of relationship.\nIndividual Differences and Preference\nfor Others’ Traits\nPreference for specific traits of other people we interact\nwith is, of course, influenced not only by the kind of rela-\ntionship we have with these persons, but also by individual\ndifferences. From the well-known “similarity–attraction”\nparadigm (see Byrne, 1971) we know that people like oth-\ners who are similar to themselves. It may, hence, be asked\nwhether a person’s own traits related to agency and com-\nmunion influence their preference for others’ agency and\ncommunion traits. If the kind of relationship is in fact an\nimportant moderator of the preference for others’ traits,\nthen this effect should remain even when individual differ-\nences in agency and communion are controlled for.\nAnother variable relevant to individual differences is the\ngoal people pursue in an interaction. Apart from general\ndifferences in interaction goals contingent on the kind of\nrelationship, such as exchanging rewards in an exchange\nrelationship versus fulfilling needs in a communal relation-\nship, every kind of relationship may serve further goals (see\nHogan, 1982). More specifically, an exchange relationship\nmight, in addition to exchanging rewards, also serve well-\nbeing in the interaction situation, and people may differ in\nthe weight they assign this goal in an interaction. On the\nother hand, a communal relationship may, in addition to\nfulfilling needs, also serve intellectual goals of interperson-\nal stimulation and learning from each other. Again people\nmay differ in the weight they assign this goal in an inter-\naction. Regarding our present research questions, it is im-\nportant to see whether the influence of the kind of relation-\nship upon the preference for others’ traits holds, even when\nthese differences in interaction goals are controlled for.\nPresent Research\nThe present research tests the moderating impact that the\nkind of relationship between actor and observer might have\non the preference for other persons’ agency and commu-\nnion traits. Whereas previous research studied specific\nkinds of relationships (person who keeps a secret versus\nperson who negotiates well; friend versus acquaintance; su-\npervisor on whom one is dependent or not; having power\nversus not having power), here we base our analyses on a\nbroader conceptualization of the kind of relationship.\nStarting from interdependence theory (Kelley & Thi-\nbaut, 1978; Rusbult & Van Lange, 1996; Thibaut & Kel-\nley, 1959), we distinguish between interdependent rela-\ntionships, where both partners are mutually dependent,\nand unilateral dependence relationships, where one part-\nner is dependent on the other, but the other is not (or at\nleast less) dependent on this person; and independent re-\nlationships, where both persons pursue their goals inde-\npendently of the other. Study 1 is based on this distinc-\ntion. Our first hypothesis states that averaged across all\nrelationships there will be a primacy of communion, i.e.,\nrespondents will choose more communion traits than\nagency traits. We base this prediction on the above-men-\ntioned findings that people weight others’ communal\ntraits more heavily in forming overall impressions (Abele\n86\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\nSocial Psychology 2013; Vol. 44(2):84–94\n© 2013 Hogrefe Publishing\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0c& Bruckmüller, 2011; Abele & Wojciszke, 2007; Asch,\n1946; Brambilla et al., 2011; De Bruin & Van Lange,\n1999, 2000; Kelley, 1950; Kenworthy & Tausch, 2008;\nScholer & Higgins, 2008; Wojciszke & Abele, 2008;\nWojciszke et al., 1998; Ybarra et al., 2001) and on the\nreasoning that a target’s communal traits reveal that per-\nson’s intentions better than agency traits. Hypothesis 2\npredicts that – apart from the general primacy of commu-\nnion – the partner’s agency traits are chosen more fre-\nquently in mutual dependence relationships than in uni-\nlateral dependent relationships than in independent rela-\ntionships. If a person is in a mutual dependence\nrelationship – and to a lesser degree also in a unilateral\ndependence relationship – then besides the other’s com-\nmunal traits also that person’s ability to achieve mutual\ngoals, i.e., his/her agency traits, become important as\nwell. In independent relationships, in contrast, it is main-\nly important that the other is friendly and trustworthy,\ni.e., communal, and that his/her agentic traits are of minor\nrelevance for the observer.\nStudies 2 and 3 are based on the distinction between\nexchange relationships and communal relationships (Clark,\n1984; Clark & Mills, 1979, 1993): Hypothesis 1 again says\nthat people generally choose more communion than agency\ntraits. Hypothesis 3 says that the share of agency traits cho-\nsen is higher in exchange relationships than in communal\nrelationships. The reasoning for Hypothesis 3 is that ex-\nchange relationships are characterized by interests of pur-\nsuing certain common goals, and that, in addition to trust-\nworthiness and friendliness (examples for communal\ntraits), competence and decisiveness (examples for agency\ntraits) are positively valued too.\nBesides these hypotheses on the impact of the kind of\nrelationship we are also concerned with individual differ-\nences as outlined above. Study 2 includes a measure of\nown agency and communion. We test whether a partici-\npant’s own agency and communion influence the prefer-\nence for the agency and communion traits of others in the\nsense of similarity-attraction (Byrne, 1971), and whether\nthis holds for both types of relationship. In Study 3 we\ninclude measures of two interaction goals that may be\nrelevant in all kinds of interactions. These are the goals\nof well-being in a relationship and of learning and mas-\ntery in a relationship. As communion traits are other-\nprofitable (Peeters, 2008), the well-being goal should be\nespecially fulfilled when the person one interacts with\npossesses positive communion traits. We therefore pre-\ndict that goals of well-being should lead to a higher\nchoice of others’ communion traits (Hypothesis 4). In the\ncase of learning and mastery goals, a person should be\nespecially interested in how competent and efficient the\ninteractant is in pursuing his/her goals. Hence, the agency\ntraits of others should become more relevant. We predict\nthat goals of mastery and learning lead to a higher num-\nber of agency traits selected (Hypothesis 5). These indi-\nvidual differences, however, should not moderate the im-\npact of the kind of relationship.\nStudy 1\nOverview\nWe tested Hypothesis 1 and Hypothesis 2 by presenting our\nrespondents six situations. Two situations were always con-\ncerned a mutual dependence relationship, a unilateral de-\npendence relationship, and an independence relationship.\nWe asked our respondents to imagine the respective situa-\ntion and then to choose out of a list of 16 traits those eight\ntraits that they would like their partner to have in the re-\nspective situation.\nPretest\nWe first performed a pretest with 30 participants (equal\nnumber of men and women; mean age 26 years). We ana-\nlyzed whether the situations we had constructed for the\nmain study were valid operationalizations of the intended\nthree types of relationship. Participants read the situation\ndescriptions (order balanced between participants) and an-\nswered questions on dependency. The mutual dependence\nsituations concerned a joint preparation of an exam and\nteammates in a tennis team. The unilateral dependence sit-\nuations concerned needing the lecture notes from another\nstudent and taking a lift for a longer trip from a driver one\ndoes not yet know. The independence situations concerned\nattending a weekend seminar together with another student\nand having a new neighbor in one’s apartment building.\nThe exact wording of the situations is listed in the Appen-\ndix. As an example, participants read the following situa-\ntion (mutual dependence):\nImagine that you have to prepare for an exam. The study ma-\nterial is substantial and you decide to ask another student, who\nhas to pass the same exam, but whom you don’t know yet to\nprepare for this test together. Both of you will prepare half of\nthe study material and will then teach it to one another.\nParticipants then rated own dependence (“How dependent\nare you upon the other person in this situation?”) and – in\ncase of interdependence – the other’s dependence (“How\nmuch is the other person dependent on you?”) on 7-point\nresponse scales each (Not at all to Very much). Supporting\nour intended manipulations, the ratings for own depend-\nence were higher in the mutually dependent (M = 5.48, SD\n= 1.14) and unilateral dependent (M = 5.88, SD = 0.94)\nsituations than in the independent situation (M = 2.38, SD\n= 1.18), F(2, 58) = 130.41, p < .001, η2partial = .82. The rat-\nings for the other person’s dependence also followed the\nintended pattern with higher means for the interdependent\nsituations (M = 5.33, SD = 1.18) than for the unilateral\ndependence situation (M = 2.18, SD = 1.27), t (29) = 8.70,\np < .001.\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\n87\n© 2013 Hogrefe Publishing\nSocial Psychology 2013; Vol. 44(2):84–94\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0cMain Study\nParticipants and Procedure\nA total of 65 female and 65 male students (mean age = 22\nyears, SD = 1.9) participated voluntarily in a study on “traits\nwe regard as important in others.” They received sweets for\ntheir participation. Participants were approached at various\nlocations on the campus of a large German university. After\nhaving agreedtoparticipate,they received a booklet thatcon-\ntained the six situation descriptions as outlined above. Differ-\nent versions of the booklet with different orders of the situa-\ntion descriptions were provided. The descriptions were al-\nways presented in a same-sex version (male participants:\nmale target; female participants: female target). Participants\nwere asked to imagine the respective situation. They then\nreceived a list of 16 traits (different orders per respondent)\nand chose eight of them. Finally, they provided their demo-\ngraphics (sex, age). Answering all questions took about 10\nminutes.\nDesign and Measures\nThe design was 3 (Relationship: mutual dependence, uni-\nlateral dependence, independence) × 2 (Dimension: agen-\ncy, communion) factorial with repeated measures on both\nfactors.\nAgency and Communion\nWe selected eight agency and eight communion traits from\nan international item pool (Abele et al., 2008). These items\nwere carefully matched for valence (agency traits M = 1.86;\ncommunion traits M = 1.79; scales from –3 = very negative\nto +3 = very positive) because valence must be controlled\nwhen trait selection is based on content dimension (see Suit-\nner & Maas, 2008). The traits were also carefully matchedfor\ntypicality regarding the respective dimension (agency traits\nM = 1.93, communion traits M = 1.96; scales from –3 = not\ntypical to +3 = very typical). The agency traits were active,\nassertive, creative, intelligent, knowledgeable, rational, self-\nconfident, and autonomous. The communion traits were sen-\nsitive, caring, helpful, polite, empathic, loyal, understanding,\nand trustworthy. Previous research has shown that the selec-\ntion of these two trait dimensions forms reliable scales (α’s\nfrom .78 to .83). Dependent measures were the number of\nagency (possible range from 0 to 8) and communion traits\n(possible range from 0 to 8) selected per situation.\nResults and Discussion\nPreliminary Analyses\nThe data of 8 participants (5 men, 3 women) had to be dis-\ncarded because they had selected less than the required eight\ntraits per situation. We also checked for possible differences\nin the answers of female and male participants. We found\nsome main effects that are irrelevant here,butno interactions.\nHence, we did not consider participants’ sex further.\nHypotheses Testing\nWe first averaged the number of agency and communion\ntraits selected over all situations. The average number of\ncommunion traits was higher (M = 4.12, SD = 0.55) than\nthe number of agency traits (M = 3.88, SD = 0.55), t(121)\n= 2.46, p < .02.\nWe then averaged the number of agency and communion\ntraits chosen over the always two mutual dependence, unilat-\neraldependenceand independencesituations.Sincethe num-\nber of agency traits selected is 8 minus the number of com-\nmunion traits selected it was sufficient to analyze the number\nof agency traits chosen by means of a three (kind of relation-\nship) factorialMANOVAwith repeatedmeasures.Itrevealed\na significant repeated measures effect, F(2, 242) = 52.42, p <\n.001, η2partial = .30, with a higher number of agency traits\nchosen in the mutual dependence relationship (M = 4.44, SD\n= 0.82) than in the unilateral dependence relationship (M =\n3.70, SD = 0.84), t(121) = 6.94, p < .001, than in the inde-\npendence relationship (M = 3.49, SD = 0.83), t(121) = 2.25,\np < .03. Figure 1 shows the findings. The number of agency\ntraits selected is higher than the number of communion traits\nselected when respondents imagine a mutual dependence sit-\nuation, but the number of communion traits selected is higher\nwhen respondents imagine a unilateral dependence situation\nand especially when they imagine an independent relation-\nship situation.\nSpecific Traits Selected\nWe also analyzed which specific traits were chosen most\noften. These were trustworthy (selected in 91% of the\ncases), helpful (85%), independent (75%), polite and intel-\nligent (both 72%).\nFigure 1. Number of agency and communion traits selected\nin the mutual dependence, unilateral dependence, and in-\ndependence situations (Study 1).\n88\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\nSocial Psychology 2013; Vol. 44(2):84–94\n© 2013 Hogrefe Publishing\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0cDiscussion\nStudy 1 supported our hypotheses. In line with Hypothesis\n1, participants generally selected more communion traits\nthan agency traits. Most importantly, the data supported\nHypothesis 2 regarding the moderating impact of kind of\nrelationship. Respondents selected more agency traits,\nwhen in a mutual dependence relationship, than when in a\nunilateral dependence relationship than when in an inde-\npendent relationship.\nStudy 2\nOverview\nStudy 2 addressed the distinction between exchange rela-\ntionships versus communal relationships (Clark, 1984;\nClark & Mills, 1979, 1993). Participants were presented\nwith one situation in which they are working with a super-\nvisor (exchange relationship) and another situation in\nwhich they are seeking a friend (communal relationship).\nHypothesis 1 again says that communion traits are gener-\nally chosen more often than agency traits; and Hypothesis\n3 says that the share of agency traits chosen is larger for\nthe exchange relationship than for the communal relation-\nship. We also assessed our participants’ self-assessed agen-\ncy and communion and tested whether the above hypothe-\nses are supported even when these individual differences\nare controlled for.\nMethod\nParticipants and Procedure\nA total of 72 female and 79 male students (mean age = 23\nyears, SD = 2.3) participated voluntarily in a study on\n“traits we regard as important in others.” They received\nsweets for their participation. Recruitment was the same as\nin Study 1. After having agreed to participate, participants\nreceived a booklet that contained two situation descrip-\ntions. Half of the participants first answered the supervisor\nsituation and the other half first answered the friend situa-\ntion. Both supervisor and friend description were formulat-\ned in a gender-neutral language (exact wording see Appen-\ndix). Participants were asked to imagine being in this situ-\nation and to choose 8 traits their interacting partner should\nhave. They then received the same list of 16 traits as in\nStudy 1. At the end of the questionnaire they provided their\ndemographics (sex, age) and rated their own agency and\ncommunion. It took about 5 minutes to answer the ques-\ntions.\nDesign and Measures\nThe design was 2 (Exchange Relationship: supervisor;\ncommunal relationship: friend) × 2 (Dimension: agency,\ncommunion) factorial with repeated measures on both fac-\ntors.\nAgency and Communion Traits\nThe agency and communion traits provided were the same\nas in Study 1. Dependent measures were the number of\nagency and communion traits selected per situation (range\nfrom 0 to 8).\nOwn Agency and Communion\nParticipants rated their own agency and communion by\nmeans of the same traits as provided in the trait choice task.\nThe response format was from 1 (not at all) to 7 (very\nmuch). The ratings were averaged across the always 8\nagency and communion traits. The internal consistency of\nboth scales was acceptable (agency α = .73; communion:\nα = .78).\nResults and Discussion\nPreliminary Analyses\nThe data of four participants (3 men, 1 woman) had to be\ndiscarded because they had selected too little (2 persons)\nor too many (2 persons) traits. We again found no differ-\nences between female and male participants.\nFigure 2. Number of agency and communion traits selected\nin the exchange versus communal relationship (left: Study\n2, right: Study 3).\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\n89\n© 2013 Hogrefe Publishing\nSocial Psychology 2013; Vol. 44(2):84–94\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0cHypotheses Testing\nSupporting Hypothesis 1, participants overall chose more\ncommunion (averaged over both situations M = 4.44, SD =\n1.03) than agency traits (M = 3.56, SD = 1.03), t(146) =\n5.12, p < .001. Supporting Hypothesis 3, participants chose\nmore agency traits for the exchange relationship (M = 3.94,\nSD = 1.27) than for the communal relationship (M = 3.18,\nSD = 1.34), F(1, 146) = 32.27, p < .001, η2\npartial = .18. This\nmeans that the number of agency and communion traits\nselected was the same in the exchange relationship and dif-\nfered with more communion traits selected in the commu-\nnal relationship (Figure 2, left).\nImpact of Own Agency and Communion\nParticipants rated their own communion as higher (M =\n5.59, SD = 0.74) than their agency (M = 5.04, SD = 0.73),\nt(145) = 7.15, p < .001. We computed a linear multiple\nregression with the number of agency traits chosen re-\ngressed on own agency and communion. This regression\nresulted in a significant positive β = .26, p = .001, for agen-\ncy; and a significant negative β = –.27, p = .001, for com-\nmunion. This means that persons high in own agency se-\nlected more agency traits than persons low in own agency.\nConversely, persons high in own communion selected less\nagency traits – and hence more communion traits, as the\nnumber of communion traits chosen is 8 minus the number\nof agency traits chosen – than persons low in own commu-\nnion.\nWe also computed this regression separately for both\nkinds of relationship and found the same results (exchange\nrelationship: agency, β = .22, p = .01, communion, β = –.18,\np = .03; communal relationship: agency, β = .20, p = .01;\ncommunion, β = –.24, p < .01).\nSpecific Traits Selected\nFinally we analyzed which specific traits were chosen most\noften. These were again trustworthy (selected in 87% of the\ncases) and helpful (81%), then polite (71%), and then in-\ntelligent (62%).\nDiscussion\nAgain supporting Hypothesis 1, participants generally se-\nlected more communion traits than agency traits; and sup-\nporting Hypothesis 3, they chose more agency traits in the\nexchange relationship than in the communal relationship.\nParticipants’ own agency had a positive influence on the\nselection of agency traits, and participants’ communion\nhad a positive influence on the selection of communion\ntraits and this influence was the same for both kinds of\nrelationship.\nStudy 3\nOverview\nStudy 3 was concerned with processes underlying the\nchoice of agency vs. communion traits in others. We used\nthe same material as in Study 2 and we included measures\nfor the importance respondents assign to two kinds of goals\nin a social relationship, i.e., goals for well-being and goals\nfor learning and mastery. Hypothesis 1 again says that com-\nmunion traits are generally chosen more often than agency\ntraits; Hypothesis 3 says that the share of agency traits cho-\nsen is larger for exchange relationships than for communal\nrelationships. Hypothesis 4 says that independent of the\nkind of social relationship goals for well-being influence\nthe choice of others’ communion traits; and Hypothesis 5\nsays that goals for mastery and learning influence the\nchoice of others’ agency traits.\nMethod\nParticipants and Procedure\nA total of 58 female and 23 male students (mean age = 21\nyears, SD = 2.9) participated in a group-testing session vol-\nuntarily and without payment. They received a booklet that\ncontained the two situation descriptions used in Study 2.\nAfter having provided their demographics (sex, age) half\nof the participants first answered the supervisor situation\nand the other half first answered the friend situation.\nDesign and Dependent Measures\nThe design was 2 (Exchange Relationship: supervisor;\ncommunal relationship: friend) × 2 (Dimension: agency,\ncommunion) factorial with repeated measures on both fac-\ntors. Dependent measures were the number of agency and\ncommunion traits selected per situation (range from 0 to 8).\nInteraction Goals\nParticipants rated the importance of 4 interaction goals per\nsituation on a scale from 1 (not important) to 7 (very im-\nportant). Two ratings pertained to well-being goals (“that\nhe/she takes care of my well-being”; “that I think that we\ngo well together”) and two ratings pertained to mastery and\nlearning goals (“that he/she can teach me something”; “that\nI can learn something from him/her”). The ratings were\naveraged across the two situations. The internal consisten-\ncy of the learning/mastery scale was, α = .82 (well-being\nscale α = .68).\n90\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\nSocial Psychology 2013; Vol. 44(2):84–94\n© 2013 Hogrefe Publishing\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0cResults and Discussion\nPreliminary Analyses\nWe had to discard the data of 6 participants (all women)\nbecause they had selected more than 8 traits per situation.\nWe found no gender differences in the participants’ an-\nswers.\nHypotheses Testing\nSupporting Hypothesis 1, participants overall chose more\ncommunion (averaged over both situations M = 4.56, SD =\n1.08) than agency traits (M = 3.44), t(74) = 4.51, p < .001.\nSupporting Hypothesis 3, participants chose more agency\ntraits for the supervisor (M = 3.96, SD = 1.16) than for the\nfriend (M = 2.92, SD = 1.44), t(74) = 6.02, p < .001 (Figure\n2 right).\nImpact of Interaction Goals\nThe overall means of the two interaction goals did not differ\n(learning M = 4.82, SD = 1.01; well-being M = 4.82, SD =\n0.98), t < 1. However, mastery goals were rated as more\nimportant (M = 6.11, SD = 1.01) than well-being goals (M\n= 4.24, SD = 1.10), t(74) = 12.33, p < .001, in the exchange\nrelationship. And mastery goals were rated as less impor-\ntant (M = 3.53, SD = 1.41) than well-being goals (M = 5.39,\nSD = 1.16), t(74) = 8.98, p < .001, in communal relation-\nships.\nWe computed a linear multiple regression with the num-\nber of agency traits chosen regressed on the importance of\nboth goals. This regression resulted in a significant positive\nβ = .23, p = .05, for learning/mastery goals; and a signifi-\ncant negative β = –.30, p = .01, for well-being goals. This\nmeans that persons with higher learning/mastery goals se-\nlected more agency traits, whereas persons with higher\nwell-being goals selected more communion traits.\nWe also computed this regression separately for both\nkinds of relationship and found comparable effects.\nSpecific Traits Selected\n“Trustworthy” was again the trait selected most often (se-\nlected in 93% of the cases), then understanding (84%), then\nhelpful (81%), and then active (61%).\nDiscussion\nThe results of Study 3 replicate those of Study 2. Partici-\npants generally selected more communion than agency\ntraits (Hypothesis 1); and they selected less agency traits\nfor the communal relationship than for the exchange rela-\ntionship (Hypothesis 3). Moreover, participants selected\ntraits according to their interaction goals: Well-being goals\nled to the selection of more communion traits, and mastery\ngoals led to the selection of more agency traits. The direc-\ntion of influence of these two interaction goals was the\nsame across both types of relationship. However, the effect\nsize was smaller for the exchange relationship than for the\ncommunal relationship. We interpret this slight divergence\nwith methodological reasons. We conclude that our Hy-\npotheses 4 and 5 were supported by the present findings.\nOn the single-trait level “trustworthy” was again selected\nmost often.\nGeneral Discussion\nThe present research was concerned with the influence dif-\nferent kinds of relationships have on the preference for oth-\ners’ traits. Starting from the assumptions of the DPM\n(Abele & Wojciszke, 2007; Wojciszke et al., 2011), we pre-\ndicted and found that the kind of relationship between two\npersons moderates the preference for traits the other person\nshould have.\nIn support of Hypothesis 1, we replicated the “primacy\nof communion” effect. Whereas previous studies revealed\nthis primacy of communion with respect to information\nprocessing (Abele & Bruckmüller, 2011; Ybarra et al.,\n2001) and social judgments (Wojciszke et al., 1998), the\npresent findings extend it to trait selection. Across all three\nstudies participants selected more communion traits than\nagency traits. It should be noted that these results emerged\nwith traits that were carefully balanced for valence and di-\nmension typicality. Hence, differential favorability or dif-\nferential typicality of the agency versus communion trait\nlists are no confounds of the present findings.\nThe most frequently selected single traits were “trust-\nworthy” and “helpful.” This finding is well in line with\nother areas of research, such as research on social values\n(Schwartz & Sagie, 2000) as well as a sociofunctional anal-\nysis of relationships (Cottrell et al., 2007). Trustworthiness\nand cooperation are universal values, and interdependent\nhumans are in need of trustworthy and cooperative rela-\ntionships. “Trustworthy” and “helpful” also belong to the\n“morality” subdomain of communion traits, which has\nbeen shown to influence judgments of others more strongly\nthan traits belonging to the “warmth” subdomain of com-\nmunion (like friendly or sensitive) (Brambilla et al., 2011).\nIn support of Hypothesis 2, we found that the choice of\ntraits varies with interdependence in the relationship. Peo-\nple select more agency traits in mutual dependent relation-\nships than in relationships with unilateral dependence than\nin independent relationships. Stated differently, in accor-\ndance with the DPM, the more other people pursue com-\nmon goals, the more we prefer agency traits for them. Un-\nder these circumstances agency traits, which are self-prof-\nitable (Peeters, 1992, 2001) because they serve the pursuit\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\n91\n© 2013 Hogrefe Publishing\nSocial Psychology 2013; Vol. 44(2):84–94\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0cof own goals also become other-profitable traits, because\nthey serve the pursuit of common goals.\nIt might be argued that the present interdependence rela-\ntionship operationalizations, such as jointly preparing for an\nexam, were mainly “agency”-related interdependence rela-\ntionships in which traits like competence or decisiveness are\nespecially important – whereas other interdependence situa-\ntions, such as sharing a flat with somebody else in order to\nreduce expenses, are more “communion” interdependence\nsituations. In more “communion interdependence situations”\npeople may also chose more communion than agency traits\nfor the other. We, however, think that in every interdepen-\ndence situation both the other’s communal traits as well as\nhis/her agency traits are important. A flatmate, for instance,\nshould be trustworthy, friendly, and helpful, but that person\nshould also be competent in pursuing household duties or\nassertive when talking to the flatowner. Hence, we argue that\ninterdependence should always lead to more preference for\nanother’s agency traits than independence, in which the oth-\ner’s communion is the sole focus.\nWe also found that people select more agency traits in\nexchange-oriented relationships than in communal relation-\nships, thus supporting Hypothesis 3. Exchange-oriented rela-\ntionships serve the pursuit of clearly defined common goals\nand, hence, self-profitable agency traits become other-profit-\nable, as well. Whereas previous research (see Abele & Woj-\nciszke, 2007; Cislak, 2013; Wojciszke & Abele, 2008; Woj-\nciszke etal., 1998) wasconcerned withspecificrelationships,\nwe supported the DPM reasoning for broad classes of rela-\ntionships.\nThe present findings extend the DPM as they reveal some\nfurther variables that underlie the preference for others’ com-\nmunal versus agentic traits. First, we found that people prefer\nsimilar others, because own agency and own communion are\ncorrelated with the selection of respective traits. People high\nin own agency selected more agency traits in others, and peo-\nple high in own communion selected more communion traits\nin others. This finding is in line with research in the context\nof the similarity–attraction paradigm (Byrne, 1971). Second,\nit extends this research by showing that similarity–attraction\nalso works with respect to the fundamental dimensions of\nperson perception. Finally, since own agency and commu-\nnion did not moderate the impact of kind of relationship on\ntrait selection, it shows that this impact is a strong one that\nremains significant when individual differences in agency\nand communion are controlled for.\nStill more important for the extension of the DPM are the\ndata of Study 3: They show that interaction goals vary with\nthe kind of relationship (higher mastery goals in interactions\nwith a supervisor; higher well-being goals in interactions\nwith a friend). Irrespective of their different levels, these in-\nteraction goals determine the preference for others’ traits in\ndifferent kinds of relationship. People who endorse more\nwell-being goals are more interested in others’ communal\ntraits both in case of a supervisor and in case of a friend. And\npeople who endorse more mastery goals are more interested\nin others’ agency traits again both in case of a supervisor and\nin case of a friend. Thus, on a very broad scale, our findings\nshow that both the situation (kind of relationship and related\nlevel of interaction goals) and the person (individual differ-\nences in traits and goals) influence the preference for certain\ncharacteristics in others.\nOur theorizing and findings support and extend other ap-\nproaches in relationship research, such as the sociofunctional\nanalysis (Cottrell et al., 2007) according to which people\nforemostseek trust and cooperation (“communion”)inothers\nand only thereafter evaluate these others’ further traits in de-\npendence on the kind of relationship they have with them.\nBased on the present analyses, the distinction between type\nof relationship, perspective, and the two fundamental dimen-\nsions would allow even more precise predictions.\nThe present findings also have some implications for ap-\nplied issues. Knowing which traits are preferred in others in\nwhich kinds of relationships might help to place and match\npeople, for instance, in work contexts. Knowledge about the\nupmost importance of certain communion traits might help\nin self-presentation and impression monitoring (see also\nWortman & Wood, 2011). However, further research is need-\ned to analyze these preferences not only on the abstract level\nof traits, but also on more concrete levels of behavior selec-\ntion. These behaviors should be clearly interpretable as com-\nmunal or agentic, and we are confident that the findings\nwould be comparable to the present ones.\nConclusion\nTo sum up, the present research supported the notion that\nthe preference for another person’s traits is dependent on\nthe kind of relationship we have with this person as well as\non individual differences related to our similarity with the\nother (trait similarity) and related to interaction goals (well-\nbeing versus mastery). The combination of interdepen-\ndence theory (Kelley & Thibaut, 1978; Rusbult, 2000; Rus-\nbult & Van Lange, 2003), of the distinction of exchange\nrelationships versus communal relationships (Clark, 1984;\nClark & Mills, 1979, 1993), of the conceptualization of the\nfundamental dimensions of social judgment (Abele et al.,\n2008; Abele & Wojciszke, 2007; Fiske et al., 2007; Judd et\nal., 2005; Paulhus & Trapnell, 2008; Peeters, 1992; Ybarra\net al., 2008), and of the dual-perspective model (Abele &\nWojciszke, 2007; Wojciszke et al., 2011) proved fruitful for\nanalyzing these issues.\nAcknowledgments\nThe present research was supported by a grant from the\nGerman Research Council to the first author (DFG Ab\n45/10-2). We thank Ulla Hacker and Dilara Yüksel for their\nhelp in collecting the data for Study 2. Susanne Bruckmül-\nler, Aleksandra Cislak, and Daniela Bernhardt gave helpful\ncomments on an earlier version of this paper.\n92\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\nSocial Psychology 2013; Vol. 44(2):84–94\n© 2013 Hogrefe Publishing\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0cReferences\nAbele, A.E. (2003). The dynamics of masculine-agency and fem-\ninine-communion traits: Findings from a prospective study.\nJournal of Personality and Social Psychology, 85, 768–776.\nAbele, A.E., & Bruckmüller, S. (2011). The bigger one of the\n“Big Two”? Preferential processing of communion informa-\ntion. Journal of Experimental Social Psychology, 47, 935–948.\nAbele, A.E., Uchronski, M., Suitner, C., & Wojciszke, B. (2008).\nToward an operationalization of the fundamental dimensions\nof agency and communion: Trait content ratings in five coun-\ntries considering valence and frequency of word occurrence.\nEuropean Journal of Social Psychology, 38, 1202–1217.\nAbele, A.E., & Wojciszke, B. (2007). Agency and communion\nfrom the perspective of self versus others. Journal of Person-\nality and Social Psychology, 93, 751–763.\nAsch, S.E. (1946). Forming impressions of personality. Journal\nof Abnormal and Social Psychology, 41, 258–290.\nBakan, D. (1966). The duality of human existence. An essay on\npsychology and religion. Chicago, IL: Rand McNally.\nBeauvois, J.-L., & Dubois, N. (2009). Lay psychology and the\nsocial value of persons. Social and Personality Compass, 3,\n1082–1095.\nBrambilla, M., Rusconi, P., Sacchi, S., & Cherubini, P. (2011).\nLooking for honesty: The primary role of morality (versus so-\nciability and competence) in information gathering. European\nJournal of Social Psychology, 41, 135–143.\nByrne, D. (1971). The attraction paradigm. New York, NY: Ac-\nademic Press.\nCislak, A. (2013). Effects of power on social perception: All your\nboss can see is agency. Social Psychology, 44, 139–147.\nClark, M.S. (1984). Record keeping in two types of relationships.\nJournal of Personality and Social Psychology, 47, 549–557.\nClark, M.S., & Mills, J. (1979). Interpersonal attraction in ex-\nchange and communal relationships. Journal of Personality\nand Social Psychology, 37, 12–24.\nClark, M.S., & Mills, J. (1993). The difference between commu-\nnion and exchange relationships: What it is and is not. Person-\nality and Social Psychology Bulletin, 19, 684–691.\nCottrell, C.A., Neuberg, S.L., & Li, N.P. (2007). What do people\ndesire in others? A sociofunctional perspective on the impor-\ntance of different valued characteristics. Journal of Personality\nand Social Psychology, 92, 208–231.\nDe Bruin, E., & Van Lange, P. (1999). Impression formation and\ncooperative behavior. European Journal of Social Psychology,\n29, 305–328.\nDe Bruin, E., & Van Lange, P. (2000). What people look for in\nothers: Inferences of the perceiver and the perceived on infor-\nmation selection. Personality and Social Psychology Bulletin,\n26, 206–219.\nFiske, S.T., Cuddy, A., & Glick, P. (2007). Universal dimensions\nof social cognition: Warmth and competence. Trends in Cog-\nnitive Science, 11, 77–83.\nHogan, R. (1982). A socioanalytic theory of personality. Nebras-\nka Symposium on Motivation, 55–89.\nJudd, C.M., James-Hawkins, L., Yzerbyt, V., & Kashima, Y.\n(2005). Fundamental dimensions of social judgment: Under-\nstanding the relations between judgments of competence and\nwarmth. Journal of Personality and Social Psychology, 89,\n899–913.\nKelley, H.H. (1950). The warm cold variable in first impressions\nof persons. Journal of Personality, 18, 431–439.\nKelley, H.H., & Thibaut, J.W. (1978). Interpersonal relations: A\ntheory of interdependence. New York, NY: Wiley-Intersci-\nence.\nKenworthy, J., & Tausch, N. (2008). Beliefs about the utility and\nstability of trait attributions in an intergroup context: Differ-\nences between warmth and competence. European Journal of\nSocial Psychology, 38, 1121–1129.\nPaulhus, D.L., & Trapnell, P.D. (2008). Self-presentation of per-\nsonality: An agency-communion framework. In O.P. John,\nR.W. Robins, & L.A. Pervin (Eds.), Handbook of personality\npsychology: Theory and research (3rd ed., pp. 492–517). New\nYork, NY: Guilford.\nPeeters, G. (1992). Evaluative meanings of adjectives in vitro and\nin context: Some theoretical implications and practical conse-\nquences of positive-negative asymmetry and behavioral-adap-\ntive concepts of evaluation. Psychologica Belgica, 32,\n211–231.\nPeeters, G. (2001). In search for a social-behavioral approach-\navoidance dimension associated with evaluative trait mean-\nings. Psychologica Belgica, 41, 187–203.\nPeeters, G. (2008). The evaluative face of a descriptive model:\nCommunion and agency in Peabody’s tetradic model of trait\norganization. European Journal of Social Psychology, 38,\n1066–1072.\nPeeters, G., & Czapinski, J. (1990). Positive-negative asymmetry\nin evaluations: The distinction between affective and informa-\ntional negative effects. In W. Stroebe & M. Hewstone (Eds.),\nEuropean review of social psychology (Vol. 1, pp. 33–60).\nLondon, UK: Wiley.\nRusbult, C.E. (2000). Interdependence: Interdependence theory.\nIn A.E. Kazdin (Ed.), Encyclopedia of psychology (Vol. 4,\npp. 330–331). Washington, DC: American Psychological As-\nsociation.\nRusbult, C.E., & Van Lange, P.A.M. (1996). Interdependence\nprocesses. In E.T. Higgins & A.W. Kruglanski (Eds.), Social\npsychology: Handbook of basic principles (pp. 564–596).\nNew York, NY: Guilford.\nRusbult, C.E., & Van Lange, P.A.M. (2003). Interdependence,\ninteraction and relationships. Annual Review of Psychology,\n54, 351–375.\nScholer, A.A., & Higgins, E.T. (2008). People as resources: Ex-\nploring the functionality of warm and cold. European Journal\nof Social Psychology, 38, 1111–1120.\nSchwartz, S. H., & Sagie, G. (2000). Value consensus and impor-\ntance: A cross-national study. Journal of Cross-Cultural Psy-\nchology, 31, 465–497.\nSuitner, C., & Maas, A. (2008). The role of valence in the per-\nception of agency and communion. European Journal of So-\ncial Psychology, 38, 1073–1082.\nThibaut, J.W., & Kelley, H.H. (1959). The social psychology of\ngroups. New York, NY: Wiley.\nUchronski, M. (2008). Agency and communion in spontaneous\nself-descriptions: Occurrence and situational malleability. Eu-\nropean Journal of Social Psychology, 38, 1093–1102.\nWojciszke, B., & Abele, A.E. (2008). The primacy of communion\nover agency and its reversals in evaluation. European Journal\nof Social Psychology, 38, 1139–1147.\nWojciszke, B., Baryla, W., Parzuchowski, M., Szymkow, A., &\nAbele, A.E. (2011). Self-esteem is dominated by agency over\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\n93\n© 2013 Hogrefe Publishing\nSocial Psychology 2013; Vol. 44(2):84–94\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0ccommunion information. European Journal of Social Psychol-\nogy, 41, 617–627.\nWojciszke, B., Bazinska, R., & Jaworski, M. (1998). On the dom-\ninance of moral categories in impression formation. Personal-\nity and Social Psychology Bulletin, 24, 1245–1257.\nWortman, J., & Wood, D. (2011). The personality traits of liked\npeople. Journal of Research in Personality, 45, 519–528.\nYbarra, O., Chan, E., & Park, D. (2001). Young and old adults’\nconcerns about morality and competence. Motivation and\nEmotions, 25, 85–100.\nYbarra, O., Chan, E., Park, H., Burnstein, E., Monin, B., & Stanik,\nC. (2008). Life’s recurring challenges and the fundamental di-\nmensions: An integration and its implications for cultural dif-\nferences and similarities. European Journal of Social Psychol-\nogy, 38, 1083–1092.\nYbarra, O., Park, D., Stanik, C., & Lee, D.S. (2012). Self-judg-\nment and reputation monitoring as a function of the fundamen-\ntal dimensions, temporal perspective, and culture. European\nJournal of Social Psychology, 42, 200–209.\nAndrea E. Abele\nSocial Psychology Group\nUniversity of Erlangen-Nuremberg\nBismarckstr. 6\n91054 Erlangen\nGermany\nTel. +49 9131 8522307\nFax +49 9131 8524731\nE-mail abele@phil.uni-erlangen.de\nAppendix\nSituation Descriptions Study 1\nMutual Dependence Relationships\n– “Imagine that you have to prepare for an exam. The study\nmaterial is substantial, and you decide to ask another stu-\ndent, who has to pass the same exam, but whom you don’t\nknow yet, to prepare for this test together. Both of you\nwill prepare half of the study material and will then teach\nit to one another.”\n– “Imagine that you play tennis in a tennis team, and that\nyou had a teammate you played double with. This team-\nmate has left the team, so you will now play with a new\nteammate who just has entered the team.”\nUnilateral Dependent Relationships:\n– “Imagine that, due to time reasons, you were unable to\nattend a lecture. This lecture is highly relevant because\nyou have to take an exam on its topics. You post a mes-\nsage whether another student could lend you his/her\nnotes. Only one person, whom you did not know before,\nanswers and is willing to lend you his/her notes.”\n– “Imagine that you urgently need a lift to another city. On\nan internet platform you found someone who offers a lift\nto this city. You contact this person and arrange a meeting\npoint for the next day.”\nIndependent Relationship\n– “Imagine that you have applied for a voluntary course\nthat will last for two days. There are fixed seating ar-\nrangements, so that you will sit next to another student\nwhom you have not met yet.”\n– “Imagine that a new neighbor has moved into the flat next\ndoor to you. You do not yet know this person.”\nSituation Descriptions Study 2\nExchange Relationship\n– “Imagine you are applying for a three-month internship,\nduring which you will have to cooperate intensely with\nyour supervisor.”\nCommunal Relationship\n– “Imagine you are moving to a new city and looking for\na friend with whom you could perform a number of lei-\nsure activities.”\n94\nA.E. Abele & S. Brack: Kind of Relationship and Preference for Other Persons’ Traits\nSocial Psychology 2013; Vol. 44(2):84–94\n© 2013 Hogrefe Publishing\nThis document is copyrighted by the American Psychological Association or one of its allied publishers.  \nThis article is intended solely for the personal use of the individual user and is not to be disseminated broadly.\n\x0c'

import re

text = content

text = re.sub(r'[^\x00-\x7F]+', ' ', text)

text = re.sub(r'[\n\r\t]+', ' ', text)  

text = re.sub(r'\s+', ' ', text)  

text = re.sub(r'[^\w\s.,:;!?\'"-]', '', text)

text = text.strip() 

print(text)

A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits SocialP sychology 2013; Vol. 442:84 94 2013 Hogrefe Publishing Original Article Preference for Other Persons Traits Is Dependent on the Kind of Social Relationship Andrea E. Abele and Susanne Brack University of Erlangen-Nuremberg, Erlangen, Germany Abstract. Three studies test the proposition that people appreciate others traits relative to the kind of relationship they have. Hypotheses are derived from the dual-perspective model, according to which the fundamental content dimensions of traits, agency, and communion are linked differently to the perspectives of actor versus observer; and that the kind of relationship between both persons is a moderator of the perspective-trait link. Study 1 shows that people generally prefer communal traits in others, and that agency traits are chosen less frequently in independence relationships than in unilateral dependence relationships than in mutual dependence relationships. Study 2 replicates the communion finding and reveals that independent of respondents own agency and communion agency traits are chosen less frequently in communal relationships than in exchange relationships. Study 3 shows that in both kinds of relationship well-being goals enhance the selection of communion traits, whereas mastery goals enhance the selection of agency traits. Keywords: social relationships, dual-perspective model, fundamental dimensions of agency and communion, trait selection Imagine the people you interact with on an average day. Your interactions may include your partner, your child, a friend, a colleague, a supervisor, a teammate, a cashier in the super- market, an opponent in a negotiation, a server in a restaurant. You may, for instance, appreciate your supervisor s decisive- ness, your friend s trustworthiness, and the cashier s punctu- ality. Your relationship to these persons is different, ranging from long-time emotional bonds to chance encounters, from closeness to distance, from interdependence to independ- ence, or from exchange orientation to need orientation. The present paper is concerned with how we appreciate specific characteristics of other persons in dependence on the kind of relationship we have with them. Our general hypothesis states that one and the same characteristic may be highly appreciated in one kind of relationship and less so in another kind of relationship. We further assume that the distinction of traits into two broad content classes, the so-called fundamental dimensions or Big Two of agency and communion Abele, Uchronski, Suitner,  Wojciszke, 2008; Abele  Wojciszke, 2007; Fiske, Cuddy,  Glick, 2007; Judd, James-Hawkins, Yzerbyt,  Kashima, 2005; Ybarra et al., 2008, helps to more specifically predict which traits are appreciated in what kinds of relationships. We derive our hypotheses from the dual-perspective model of person perception DPM; Abele  Wojciszke, 2007; Wojciszke, Baryla, Parzuchowski, Szymkow,  Abele, 2011. It states that agency and communion are differen- tially linked to the perspectives of actor versus observer or recipient of an action. It also states that the link of agency and communion to the actor versus observer perspective is moderated by the specific relationship both persons have with each other. Whereas previous research has revealed supporting evidence for the differential link of agency and communion to the actor versus observer perspective Abele  Wojciszke, 2007; Wojciszke et al., 2011, research on the impact of different kinds of social relationships on trait ap- preciation is still scarce see Cislak, 2013; Wojciszke  Abele, 2008. The present study tests the latter prediction of the DPM. We apply reasoning from interdependence theory see Kelley  Thibaut, 1978; Rusbult, 2000; Rus- bult  Van Lange, 1996; Thibaut  Kelley, 1959 and from the differentiation of exchange versus communal relation- ships Clark, 1984; Clark  Mills, 1979, 1993. We also analyze whether individual differences in own agency and communion and individual differences in interaction goals moderate the impact of kind of relationship on preference for others traits. We present three studies that test our hy- potheses. Kinds of Social Relationships Thibaut and Kelley 1959 were among the first to argue that social behavior should be understood by studying the fabric of interdependence characterizing social situations. DOI: 10.10271864-9335a000138 Social Psychology 2013; Vol. 442:84 94 2013 Hogrefe Publishing This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. They defined interdependence as the strength and quality of the effects that interacting individuals exert on one an- other in terms of preferences, motives, behavior, and out- comes. Interdependence differs in level and mutuality of dependence, in the basis of the dependence, and in the co- variation of interests Kelley  Thibaut, 1978; Rusbult, 2000; Rusbult  Van Lange, 2003. We concentrate here on mutuality of dependence. Applied to two of our above examples, the level of in- terdependence with the cashier at a supermarket is low, since the person who goes shopping may choose one of several cashiers to pay the due amount. The mutuality of dependence is also low, as the cashier has many clients and is not dependent on a single one in order to, say, keep his or her job. On the other hand, the level of interdependence with a supervisor is high, since the employee has to fulfill the supervisor s orders and cannot simply decide to look for another supervisor. There is also a certain degree of mutuality of dependence because the supervisor wants the work to be done efficiently by hisher employee. Another relevant distinction of social relationships was introduced by Clark and Mills Clark, 1984; Clark  Mills, 1979, 1993, who argue that social relationships are either reward-based, something they call exchange relationships; or they are needs-based, something they call communal re- lationships. In an exchange relationship, benefits are given with the expectation of receiving a benefit in return. Each person is concerned with how much he or she receives in exchange for benefiting the other, and how much he or she owes the other for the received benefits. Members of a communal relationship assume that each is concerned about the welfare of each other. They have a positive atti- tude toward benefiting the other when a need for the benefit exists and the receipt of a benefit does not create a specific obligation to return a comparable benefit Clark  Mills, 1979. Exchange relationships typically exist between strangers and casual acquaintances as well as in certain long-term arrangements such as business partnerships. In contrast, strong communal relationships are usually limited to family members, friends, and romantic partners Clark  Mills, 1993. Applied to our above examples, the server is an example of an exchange relationship in which one benefit good ser- vice is expected to be reciprocated by another benefit a good tip. A parent-child relationship is, of course, a com- munal one in which the parent responds to the child s need and does not expect a direct compensation. The Fundamental Dimensions of Social Perception and the Dual-Perspective Model Research has revealed thatthere are two basiccontentdimen- sions in social perception the fundamental dimensions or Big Two thatunderlie judgmentsofvarioustargetslikethe self, other persons, groups, etc. Abele et al., 2008; Abele  Wojciszke,2007;Fiskeetal.,2007;Judd etal.,2005; Paulhus  Trapnell, 2008; Peeters, 1992; Ybarra et al., 2008. Agency also-called competence comprises competence, goal achievement, individuality, and self-assertion and focuses on own goals and their achievement. Communion also called warmth or morality, by contrast, emphasizes the individual as a social being and compromises cooperation, morality, warmth and trustworthiness. These two dimensions reflect the duality of human existence Bakan, 1966 and the basic goals of social information processing, namely, forming and maintaining social connections communion and pursuing goals and manifesting skills and accomplishments agency; Fiske et al., 2007; Ybarra et al., 2008. In addition to being a classification of content, the di- mensions of communion and agency can also be distin- guished with respect to the social value they have, i.e., their profitability for other people interacting with the trait possessor versus their profitability for the trait possessor him-herself Peeters, 2001, 2008; Peeters  Czapinski, 1990; see also Beauvois  Dubois, 2009. Communion traits tend to be other-profitable because they inform the perceiver about attributes of the target that first and fore- most have consequences positive or negative for other people interacting with the target. Agency traits tend to be self-profitable because they first and foremost have conse- quences positive or negative for the trait possessor and the pursuit of his or her goals. The DPM Abele  Wojciszke, 2007; Wojciszke et al., 2011 predicts that the interpretation of behavior in terms of traits as well as the appreciation and relevance of traits differ between perspectives. Communion generally is the primary of the two content classes, because a target s com- munal traits inform the observer about this person s inten- tions and whether it is safe to approach himher or not. The primacy of communion is revealed in processing speed, i.e., communal content is processed faster than agency con- tent Abele  Bruckm ller, 2011; Ybarra, Chan,  Park, 2001 and in the weighting of others traits because people place more weight on communion traits in forming overall impressions of others than on agency traits Abele  Woj- ciszke, 2007; Asch, 1946; Brambilla, Rusconi, Sacchi,  Cherubini, 2011; De Bruin  Van Lange, 1999, 2000; Kel- ley, 1950; Kenworthy  Tausch, 2008; Scholer  Higgins, 2008; Wojciszke  Abele, 2008; Wojciszke, Bazinska,  Jaworski, 1998. When asked to describe themselves, peo- ple also usually indicate higher communion than agency e.g., Abele, 2003; Uchronski, 2008; Wojciszke et al., 2011; Ybarra, Park, Stanik,  Lee, 2012 again indicating the primacy of communion. Apart from this general primacy of communion, howev- er, there are clear differences contingent on the actor-ver- sus-observer perspective. Communion content being oth- er-profitable is more relevant in the observer than in the actor perspective and agency content being self-profitable is more relevant in the actor than in the observer perspec- A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits 85 2013 Hogrefe Publishing Social Psychology 2013; Vol. 442:84 94 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. tive. Wojciszke and colleagues 2011, for instance, showed that people base their self-esteem more on their agency than their communion, whereas the evaluation of others is based more on their communion than their agency. Kind of Social Relationship and Preference for Others Traits There is only limited research concerning the moderating im- pact the kind of social relationship between actor and observ- errecipient mighthave inascribingor evaluating agencyver- sus communion traits. Wojciszke et al. 1998 found that re- spondents preferred others communal traits when they were seeking a person to share a secret with; however, they pre- ferred agentic traits when they were looking for a good nego- tiator. Abele and Wojciszke 2007, study 4 showed that agency traits were rated as more important for a close friend than for a more distant acquaintance. Wojciszke and Abele 2008 found that liking a supervisor was dependent on this supervisor s communion qualities when the supervisor had only limited influenceonthe respondent soutcomes.If,how- ever,the supervisor s influenceontherespondent soutcomes was strong, then liking the supervisor was more influenced by this person s agency qualities. Cislak 2013 also analyzed observers evaluations of other persons agency versus com- munion traits. The respondents were primed with roles or memories that were always connected with having or not having power. The findings revealed that having power led to an increased interest in others agency traits, whereas not having power led to an increased interest in others commu- nion traits. Relationship researchers not concerned with the funda- mental dimensions of agency and communion have also found evidence that fits the present reasoning. Cottrell, Neuberg, and Li 2007, for instance, analyzed the impor- tance of different traits for different kinds of targets gen- eralized ideal person ; ideal persons with respect to dif- ferent social relationships; persons differently interrelated with the judging person. They hypothesized and found that traits like trustworthiness and cooperativeness are general- ly highly valued a finding that can be covered under the present heading of primacy of communion. Characteris- tics such as intelligence were differently valued across tasks, goals, and functions. Assuming that intelligence belongs to the agency dimensions, this finding is also well in line with our prediction that the importance of agency traits varies with the kind of relationship. Individual Differences and Preference for Others Traits Preference for specific traits of other people we interact with is, of course, influenced not only by the kind of rela- tionship we have with these persons, but also by individual differences. From the well-known similarity attraction paradigm see Byrne, 1971 we know that people like oth- ers who are similar to themselves. It may, hence, be asked whether a person s own traits related to agency and com- munion influence their preference for others agency and communion traits. If the kind of relationship is in fact an important moderator of the preference for others traits, then this effect should remain even when individual differ- ences in agency and communion are controlled for. Another variable relevant to individual differences is the goal people pursue in an interaction. Apart from general differences in interaction goals contingent on the kind of relationship, such as exchanging rewards in an exchange relationship versus fulfilling needs in a communal relation- ship, every kind of relationship may serve further goals see Hogan, 1982. More specifically, an exchange relationship might, in addition to exchanging rewards, also serve well- being in the interaction situation, and people may differ in the weight they assign this goal in an interaction. On the other hand, a communal relationship may, in addition to fulfilling needs, also serve intellectual goals of interperson- al stimulation and learning from each other. Again people may differ in the weight they assign this goal in an inter- action. Regarding our present research questions, it is im- portant to see whether the influence of the kind of relation- ship upon the preference for others traits holds, even when these differences in interaction goals are controlled for. Present Research The present research tests the moderating impact that the kind of relationship between actor and observer might have on the preference for other persons agency and commu- nion traits. Whereas previous research studied specific kinds of relationships person who keeps a secret versus person who negotiates well; friend versus acquaintance; su- pervisor on whom one is dependent or not; having power versus not having power, here we base our analyses on a broader conceptualization of the kind of relationship. Starting from interdependence theory Kelley  Thi- baut, 1978; Rusbult  Van Lange, 1996; Thibaut  Kel- ley, 1959, we distinguish between interdependent rela- tionships, where both partners are mutually dependent, and unilateral dependence relationships, where one part- ner is dependent on the other, but the other is not or at least less dependent on this person; and independent re- lationships, where both persons pursue their goals inde- pendently of the other. Study 1 is based on this distinc- tion. Our first hypothesis states that averaged across all relationships there will be a primacy of communion, i.e., respondents will choose more communion traits than agency traits. We base this prediction on the above-men- tioned findings that people weight others communal traits more heavily in forming overall impressions Abele 86 A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits Social Psychology 2013; Vol. 442:84 94 2013 Hogrefe Publishing This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.  Bruckm ller, 2011; Abele  Wojciszke, 2007; Asch, 1946; Brambilla et al., 2011; De Bruin  Van Lange, 1999, 2000; Kelley, 1950; Kenworthy  Tausch, 2008; Scholer  Higgins, 2008; Wojciszke  Abele, 2008; Wojciszke et al., 1998; Ybarra et al., 2001 and on the reasoning that a target s communal traits reveal that per- son s intentions better than agency traits. Hypothesis 2 predicts that apart from the general primacy of commu- nion the partner s agency traits are chosen more fre- quently in mutual dependence relationships than in uni- lateral dependent relationships than in independent rela- tionships. If a person is in a mutual dependence relationship and to a lesser degree also in a unilateral dependence relationship then besides the other s com- munal traits also that person s ability to achieve mutual goals, i.e., hisher agency traits, become important as well. In independent relationships, in contrast, it is main- ly important that the other is friendly and trustworthy, i.e., communal, and that hisher agentic traits are of minor relevance for the observer. Studies 2 and 3 are based on the distinction between exchange relationships and communal relationships Clark, 1984; Clark  Mills, 1979, 1993: Hypothesis 1 again says that people generally choose more communion than agency traits. Hypothesis 3 says that the share of agency traits cho- sen is higher in exchange relationships than in communal relationships. The reasoning for Hypothesis 3 is that ex- change relationships are characterized by interests of pur- suing certain common goals, and that, in addition to trust- worthiness and friendliness examples for communal traits, competence and decisiveness examples for agency traits are positively valued too. Besides these hypotheses on the impact of the kind of relationship we are also concerned with individual differ- ences as outlined above. Study 2 includes a measure of own agency and communion. We test whether a partici- pant s own agency and communion influence the prefer- ence for the agency and communion traits of others in the sense of similarity-attraction Byrne, 1971, and whether this holds for both types of relationship. In Study 3 we include measures of two interaction goals that may be relevant in all kinds of interactions. These are the goals of well-being in a relationship and of learning and mas- tery in a relationship. As communion traits are other- profitable Peeters, 2008, the well-being goal should be especially fulfilled when the person one interacts with possesses positive communion traits. We therefore pre- dict that goals of well-being should lead to a higher choice of others communion traits Hypothesis 4. In the case of learning and mastery goals, a person should be especially interested in how competent and efficient the interactant is in pursuing hisher goals. Hence, the agency traits of others should become more relevant. We predict that goals of mastery and learning lead to a higher num- ber of agency traits selected Hypothesis 5. These indi- vidual differences, however, should not moderate the im- pact of the kind of relationship. Study 1 Overview We tested Hypothesis 1 and Hypothesis 2 by presenting our respondents six situations. Two situations were always con- cerned a mutual dependence relationship, a unilateral de- pendence relationship, and an independence relationship. We asked our respondents to imagine the respective situa- tion and then to choose out of a list of 16 traits those eight traits that they would like their partner to have in the re- spective situation. Pretest We first performed a pretest with 30 participants equal number of men and women; mean age 26 years. We ana- lyzed whether the situations we had constructed for the main study were valid operationalizations of the intended three types of relationship. Participants read the situation descriptions order balanced between participants and an- swered questions on dependency. The mutual dependence situations concerned a joint preparation of an exam and teammates in a tennis team. The unilateral dependence sit- uations concerned needing the lecture notes from another student and taking a lift for a longer trip from a driver one does not yet know. The independence situations concerned attending a weekend seminar together with another student and having a new neighbor in one s apartment building. The exact wording of the situations is listed in the Appen- dix. As an example, participants read the following situa- tion mutual dependence: Imagine that you have to prepare for an exam. The study ma- terial is substantial and you decide to ask another student, who has to pass the same exam, but whom you don t know yet to prepare for this test together. Both of you will prepare half of the study material and will then teach it to one another. Participants then rated own dependence  How dependent are you upon the other person in this situation?  and in case of interdependence the other s dependence  How much is the other person dependent on you?  on 7-point response scales each Not at all to Very much. Supporting our intended manipulations, the ratings for own depend- ence were higher in the mutually dependent M  5.48, SD  1.14 and unilateral dependent M  5.88, SD  0.94 situations than in the independent situation M  2.38, SD  1.18, F2, 58  130.41, p  .001, 2partial  .82. The rat- ings for the other person s dependence also followed the intended pattern with higher means for the interdependent situations M  5.33, SD  1.18 than for the unilateral dependence situation M  2.18, SD  1.27, t 29  8.70, p  .001. A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits 87 2013 Hogrefe Publishing Social Psychology 2013; Vol. 442:84 94 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Main Study Participants and Procedure A total of 65 female and 65 male students mean age  22 years, SD  1.9 participated voluntarily in a study on traits we regard as important in others. They received sweets for their participation. Participants were approached at various locations on the campus of a large German university. After having agreedtoparticipate,they received a booklet thatcon- tained the six situation descriptions as outlined above. Differ- ent versions of the booklet with different orders of the situa- tion descriptions were provided. The descriptions were al- ways presented in a same-sex version male participants: male target; female participants: female target. Participants were asked to imagine the respective situation. They then received a list of 16 traits different orders per respondent and chose eight of them. Finally, they provided their demo- graphics sex, age. Answering all questions took about 10 minutes. Design and Measures The design was 3 Relationship: mutual dependence, uni- lateral dependence, independence 2 Dimension: agen- cy, communion factorial with repeated measures on both factors. Agency and Communion We selected eight agency and eight communion traits from an international item pool Abele et al., 2008. These items were carefully matched for valence agency traits M  1.86; communion traits M  1.79; scales from 3  very negative to 3  very positive because valence must be controlled when trait selection is based on content dimension see Suit- ner  Maas, 2008. The traits were also carefully matchedfor typicality regarding the respective dimension agency traits M  1.93, communion traits M  1.96; scales from 3  not typical to 3  very typical. The agency traits were active, assertive, creative, intelligent, knowledgeable, rational, self- confident, and autonomous. The communion traits were sen- sitive, caring, helpful, polite, empathic, loyal, understanding, and trustworthy. Previous research has shown that the selec- tion of these two trait dimensions forms reliable scales  s from .78 to .83. Dependent measures were the number of agency possible range from 0 to 8 and communion traits possible range from 0 to 8 selected per situation. Results and Discussion Preliminary Analyses The data of 8 participants 5 men, 3 women had to be dis- carded because they had selected less than the required eight traits per situation. We also checked for possible differences in the answers of female and male participants. We found some main effects that are irrelevant here,butno interactions. Hence, we did not consider participants sex further. Hypotheses Testing We first averaged the number of agency and communion traits selected over all situations. The average number of communion traits was higher M  4.12, SD  0.55 than the number of agency traits M  3.88, SD  0.55, t121  2.46, p  .02. We then averaged the number of agency and communion traits chosen over the always two mutual dependence, unilat- eraldependenceand independencesituations.Sincethe num- ber of agency traits selected is 8 minus the number of com- munion traits selected it was sufficient to analyze the number of agency traits chosen by means of a three kind of relation- ship factorialMANOVAwith repeatedmeasures.Itrevealed a significant repeated measures effect, F2, 242  52.42, p  .001, 2partial  .30, with a higher number of agency traits chosen in the mutual dependence relationship M  4.44, SD  0.82 than in the unilateral dependence relationship M  3.70, SD  0.84, t121  6.94, p  .001, than in the inde- pendence relationship M  3.49, SD  0.83, t121  2.25, p  .03. Figure 1 shows the findings. The number of agency traits selected is higher than the number of communion traits selected when respondents imagine a mutual dependence sit- uation, but the number of communion traits selected is higher when respondents imagine a unilateral dependence situation and especially when they imagine an independent relation- ship situation. Specific Traits Selected We also analyzed which specific traits were chosen most often. These were trustworthy selected in 91 of the cases, helpful 85, independent 75, polite and intel- ligent both 72. Figure 1. Number of agency and communion traits selected in the mutual dependence, unilateral dependence, and in- dependence situations Study 1. 88 A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits Social Psychology 2013; Vol. 442:84 94 2013 Hogrefe Publishing This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Discussion Study 1 supported our hypotheses. In line with Hypothesis 1, participants generally selected more communion traits than agency traits. Most importantly, the data supported Hypothesis 2 regarding the moderating impact of kind of relationship. Respondents selected more agency traits, when in a mutual dependence relationship, than when in a unilateral dependence relationship than when in an inde- pendent relationship. Study 2 Overview Study 2 addressed the distinction between exchange rela- tionships versus communal relationships Clark, 1984; Clark  Mills, 1979, 1993. Participants were presented with one situation in which they are working with a super- visor exchange relationship and another situation in which they are seeking a friend communal relationship. Hypothesis 1 again says that communion traits are gener- ally chosen more often than agency traits; and Hypothesis 3 says that the share of agency traits chosen is larger for the exchange relationship than for the communal relation- ship. We also assessed our participants self-assessed agen- cy and communion and tested whether the above hypothe- ses are supported even when these individual differences are controlled for. Method Participants and Procedure A total of 72 female and 79 male students mean age  23 years, SD  2.3 participated voluntarily in a study on traits we regard as important in others. They received sweets for their participation. Recruitment was the same as in Study 1. After having agreed to participate, participants received a booklet that contained two situation descrip- tions. Half of the participants first answered the supervisor situation and the other half first answered the friend situa- tion. Both supervisor and friend description were formulat- ed in a gender-neutral language exact wording see Appen- dix. Participants were asked to imagine being in this situ- ation and to choose 8 traits their interacting partner should have. They then received the same list of 16 traits as in Study 1. At the end of the questionnaire they provided their demographics sex, age and rated their own agency and communion. It took about 5 minutes to answer the ques- tions. Design and Measures The design was 2 Exchange Relationship: supervisor; communal relationship: friend 2 Dimension: agency, communion factorial with repeated measures on both fac- tors. Agency and Communion Traits The agency and communion traits provided were the same as in Study 1. Dependent measures were the number of agency and communion traits selected per situation range from 0 to 8. Own Agency and Communion Participants rated their own agency and communion by means of the same traits as provided in the trait choice task. The response format was from 1 not at all to 7 very much. The ratings were averaged across the always 8 agency and communion traits. The internal consistency of both scales was acceptable agency  .73; communion:  .78. Results and Discussion Preliminary Analyses The data of four participants 3 men, 1 woman had to be discarded because they had selected too little 2 persons or too many 2 persons traits. We again found no differ- ences between female and male participants. Figure 2. Number of agency and communion traits selected in the exchange versus communal relationship left: Study 2, right: Study 3. A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits 89 2013 Hogrefe Publishing Social Psychology 2013; Vol. 442:84 94 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Hypotheses Testing Supporting Hypothesis 1, participants overall chose more communion averaged over both situations M  4.44, SD  1.03 than agency traits M  3.56, SD  1.03, t146  5.12, p  .001. Supporting Hypothesis 3, participants chose more agency traits for the exchange relationship M  3.94, SD  1.27 than for the communal relationship M  3.18, SD  1.34, F1, 146  32.27, p  .001, 2 partial  .18. This means that the number of agency and communion traits selected was the same in the exchange relationship and dif- fered with more communion traits selected in the commu- nal relationship Figure 2, left. Impact of Own Agency and Communion Participants rated their own communion as higher M  5.59, SD  0.74 than their agency M  5.04, SD  0.73, t145  7.15, p  .001. We computed a linear multiple regression with the number of agency traits chosen re- gressed on own agency and communion. This regression resulted in a significant positive  .26, p  .001, for agen- cy; and a significant negative  .27, p  .001, for com- munion. This means that persons high in own agency se- lected more agency traits than persons low in own agency. Conversely, persons high in own communion selected less agency traits and hence more communion traits, as the number of communion traits chosen is 8 minus the number of agency traits chosen than persons low in own commu- nion. We also computed this regression separately for both kinds of relationship and found the same results exchange relationship: agency,  .22, p  .01, communion,  .18, p  .03; communal relationship: agency,  .20, p  .01; communion,  .24, p  .01. Specific Traits Selected Finally we analyzed which specific traits were chosen most often. These were again trustworthy selected in 87 of the cases and helpful 81, then polite 71, and then in- telligent 62. Discussion Again supporting Hypothesis 1, participants generally se- lected more communion traits than agency traits; and sup- porting Hypothesis 3, they chose more agency traits in the exchange relationship than in the communal relationship. Participants own agency had a positive influence on the selection of agency traits, and participants communion had a positive influence on the selection of communion traits and this influence was the same for both kinds of relationship. Study 3 Overview Study 3 was concerned with processes underlying the choice of agency vs. communion traits in others. We used the same material as in Study 2 and we included measures for the importance respondents assign to two kinds of goals in a social relationship, i.e., goals for well-being and goals for learning and mastery. Hypothesis 1 again says that com- munion traits are generally chosen more often than agency traits; Hypothesis 3 says that the share of agency traits cho- sen is larger for exchange relationships than for communal relationships. Hypothesis 4 says that independent of the kind of social relationship goals for well-being influence the choice of others communion traits; and Hypothesis 5 says that goals for mastery and learning influence the choice of others agency traits. Method Participants and Procedure A total of 58 female and 23 male students mean age  21 years, SD  2.9 participated in a group-testing session vol- untarily and without payment. They received a booklet that contained the two situation descriptions used in Study 2. After having provided their demographics sex, age half of the participants first answered the supervisor situation and the other half first answered the friend situation. Design and Dependent Measures The design was 2 Exchange Relationship: supervisor; communal relationship: friend 2 Dimension: agency, communion factorial with repeated measures on both fac- tors. Dependent measures were the number of agency and communion traits selected per situation range from 0 to 8. Interaction Goals Participants rated the importance of 4 interaction goals per situation on a scale from 1 not important to 7 very im- portant. Two ratings pertained to well-being goals  that heshe takes care of my well-being ; that I think that we go well together  and two ratings pertained to mastery and learning goals  that heshe can teach me something ; that I can learn something from himher . The ratings were averaged across the two situations. The internal consisten- cy of the learningmastery scale was,  .82 well-being scale  .68. 90 A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits Social Psychology 2013; Vol. 442:84 94 2013 Hogrefe Publishing This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Results and Discussion Preliminary Analyses We had to discard the data of 6 participants all women because they had selected more than 8 traits per situation. We found no gender differences in the participants an- swers. Hypotheses Testing Supporting Hypothesis 1, participants overall chose more communion averaged over both situations M  4.56, SD  1.08 than agency traits M  3.44, t74  4.51, p  .001. Supporting Hypothesis 3, participants chose more agency traits for the supervisor M  3.96, SD  1.16 than for the friend M  2.92, SD  1.44, t74  6.02, p  .001 Figure 2 right. Impact of Interaction Goals The overall means of the two interaction goals did not differ learning M  4.82, SD  1.01; well-being M  4.82, SD  0.98, t  1. However, mastery goals were rated as more important M  6.11, SD  1.01 than well-being goals M  4.24, SD  1.10, t74  12.33, p  .001, in the exchange relationship. And mastery goals were rated as less impor- tant M  3.53, SD  1.41 than well-being goals M  5.39, SD  1.16, t74  8.98, p  .001, in communal relation- ships. We computed a linear multiple regression with the num- ber of agency traits chosen regressed on the importance of both goals. This regression resulted in a significant positive  .23, p  .05, for learningmastery goals; and a signifi- cant negative  .30, p  .01, for well-being goals. This means that persons with higher learningmastery goals se- lected more agency traits, whereas persons with higher well-being goals selected more communion traits. We also computed this regression separately for both kinds of relationship and found comparable effects. Specific Traits Selected Trustworthy was again the trait selected most often se- lected in 93 of the cases, then understanding 84, then helpful 81, and then active 61. Discussion The results of Study 3 replicate those of Study 2. Partici- pants generally selected more communion than agency traits Hypothesis 1; and they selected less agency traits for the communal relationship than for the exchange rela- tionship Hypothesis 3. Moreover, participants selected traits according to their interaction goals: Well-being goals led to the selection of more communion traits, and mastery goals led to the selection of more agency traits. The direc- tion of influence of these two interaction goals was the same across both types of relationship. However, the effect size was smaller for the exchange relationship than for the communal relationship. We interpret this slight divergence with methodological reasons. We conclude that our Hy- potheses 4 and 5 were supported by the present findings. On the single-trait level trustworthy was again selected most often. General Discussion The present research was concerned with the influence dif- ferent kinds of relationships have on the preference for oth- ers traits. Starting from the assumptions of the DPM Abele  Wojciszke, 2007; Wojciszke et al., 2011, we pre- dicted and found that the kind of relationship between two persons moderates the preference for traits the other person should have. In support of Hypothesis 1, we replicated the primacy of communion effect. Whereas previous studies revealed this primacy of communion with respect to information processing Abele  Bruckm ller, 2011; Ybarra et al., 2001 and social judgments Wojciszke et al., 1998, the present findings extend it to trait selection. Across all three studies participants selected more communion traits than agency traits. It should be noted that these results emerged with traits that were carefully balanced for valence and di- mension typicality. Hence, differential favorability or dif- ferential typicality of the agency versus communion trait lists are no confounds of the present findings. The most frequently selected single traits were trust- worthy and helpful. This finding is well in line with other areas of research, such as research on social values Schwartz  Sagie, 2000 as well as a sociofunctional anal- ysis of relationships Cottrell et al., 2007. Trustworthiness and cooperation are universal values, and interdependent humans are in need of trustworthy and cooperative rela- tionships. Trustworthy and helpful also belong to the morality subdomain of communion traits, which has been shown to influence judgments of others more strongly than traits belonging to the warmth subdomain of com- munion like friendly or sensitive Brambilla et al., 2011. In support of Hypothesis 2, we found that the choice of traits varies with interdependence in the relationship. Peo- ple select more agency traits in mutual dependent relation- ships than in relationships with unilateral dependence than in independent relationships. Stated differently, in accor- dance with the DPM, the more other people pursue com- mon goals, the more we prefer agency traits for them. Un- der these circumstances agency traits, which are self-prof- itable Peeters, 1992, 2001 because they serve the pursuit A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits 91 2013 Hogrefe Publishing Social Psychology 2013; Vol. 442:84 94 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. of own goals also become other-profitable traits, because they serve the pursuit of common goals. It might be argued that the present interdependence rela- tionship operationalizations, such as jointly preparing for an exam, were mainly agency -related interdependence rela- tionships in which traits like competence or decisiveness are especially important whereas other interdependence situa- tions, such as sharing a flat with somebody else in order to reduce expenses, are more communion interdependence situations. In more communion interdependence situations people may also chose more communion than agency traits for the other. We, however, think that in every interdepen- dence situation both the other s communal traits as well as hisher agency traits are important. A flatmate, for instance, should be trustworthy, friendly, and helpful, but that person should also be competent in pursuing household duties or assertive when talking to the flatowner. Hence, we argue that interdependence should always lead to more preference for another s agency traits than independence, in which the oth- er s communion is the sole focus. We also found that people select more agency traits in exchange-oriented relationships than in communal relation- ships, thus supporting Hypothesis 3. Exchange-oriented rela- tionships serve the pursuit of clearly defined common goals and, hence, self-profitable agency traits become other-profit- able, as well. Whereas previous research see Abele  Woj- ciszke, 2007; Cislak, 2013; Wojciszke  Abele, 2008; Woj- ciszke etal., 1998 wasconcerned withspecificrelationships, we supported the DPM reasoning for broad classes of rela- tionships. The present findings extend the DPM as they reveal some further variables that underlie the preference for others com- munal versus agentic traits. First, we found that people prefer similar others, because own agency and own communion are correlated with the selection of respective traits. People high in own agency selected more agency traits in others, and peo- ple high in own communion selected more communion traits in others. This finding is in line with research in the context of the similarity attraction paradigm Byrne, 1971. Second, it extends this research by showing that similarity attraction also works with respect to the fundamental dimensions of person perception. Finally, since own agency and commu- nion did not moderate the impact of kind of relationship on trait selection, it shows that this impact is a strong one that remains significant when individual differences in agency and communion are controlled for. Still more important for the extension of the DPM are the data of Study 3: They show that interaction goals vary with the kind of relationship higher mastery goals in interactions with a supervisor; higher well-being goals in interactions with a friend. Irrespective of their different levels, these in- teraction goals determine the preference for others traits in different kinds of relationship. People who endorse more well-being goals are more interested in others communal traits both in case of a supervisor and in case of a friend. And people who endorse more mastery goals are more interested in others agency traits again both in case of a supervisor and in case of a friend. Thus, on a very broad scale, our findings show that both the situation kind of relationship and related level of interaction goals and the person individual differ- ences in traits and goals influence the preference for certain characteristics in others. Our theorizing and findings support and extend other ap- proaches in relationship research, such as the sociofunctional analysis Cottrell et al., 2007 according to which people foremostseek trust and cooperation  communion inothers and only thereafter evaluate these others further traits in de- pendence on the kind of relationship they have with them. Based on the present analyses, the distinction between type of relationship, perspective, and the two fundamental dimen- sions would allow even more precise predictions. The present findings also have some implications for ap- plied issues. Knowing which traits are preferred in others in which kinds of relationships might help to place and match people, for instance, in work contexts. Knowledge about the upmost importance of certain communion traits might help in self-presentation and impression monitoring see also Wortman  Wood, 2011. However, further research is need- ed to analyze these preferences not only on the abstract level of traits, but also on more concrete levels of behavior selec- tion. These behaviors should be clearly interpretable as com- munal or agentic, and we are confident that the findings would be comparable to the present ones. Conclusion To sum up, the present research supported the notion that the preference for another person s traits is dependent on the kind of relationship we have with this person as well as on individual differences related to our similarity with the other trait similarity and related to interaction goals well- being versus mastery. The combination of interdepen- dence theory Kelley  Thibaut, 1978; Rusbult, 2000; Rus- bult  Van Lange, 2003, of the distinction of exchange relationships versus communal relationships Clark, 1984; Clark  Mills, 1979, 1993, of the conceptualization of the fundamental dimensions of social judgment Abele et al., 2008; Abele  Wojciszke, 2007; Fiske et al., 2007; Judd et al., 2005; Paulhus  Trapnell, 2008; Peeters, 1992; Ybarra et al., 2008, and of the dual-perspective model Abele  Wojciszke, 2007; Wojciszke et al., 2011 proved fruitful for analyzing these issues. Acknowledgments The present research was supported by a grant from the German Research Council to the first author DFG Ab 4510-2. We thank Ulla Hacker and Dilara Y ksel for their help in collecting the data for Study 2. Susanne Bruckm l- ler, Aleksandra Cislak, and Daniela Bernhardt gave helpful comments on an earlier version of this paper. 92 A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits Social Psychology 2013; Vol. 442:84 94 2013 Hogrefe Publishing This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. References Abele, A.E. 2003. The dynamics of masculine-agency and fem- inine-communion traits: Findings from a prospective study. Journal of Personality and Social Psychology, 85, 768 776. Abele, A.E.,  Bruckm ller, S. 2011. The bigger one of the Big Two ? Preferential processing of communion informa- tion. Journal of Experimental Social Psychology, 47, 935 948. Abele, A.E., Uchronski, M., Suitner, C.,  Wojciszke, B. 2008. Toward an operationalization of the fundamental dimensions of agency and communion: Trait content ratings in five coun- tries considering valence and frequency of word occurrence. European Journal of Social Psychology, 38, 1202 1217. Abele, A.E.,  Wojciszke, B. 2007. Agency and communion from the perspective of self versus others. Journal of Person- ality and Social Psychology, 93, 751 763. Asch, S.E. 1946. Forming impressions of personality. Journal of Abnormal and Social Psychology, 41, 258 290. Bakan, D. 1966. The duality of human existence. An essay on psychology and religion. Chicago, IL: Rand McNally. Beauvois, J.-L.,  Dubois, N. 2009. Lay psychology and the social value of persons. Social and Personality Compass, 3, 1082 1095. Brambilla, M., Rusconi, P., Sacchi, S.,  Cherubini, P. 2011. Looking for honesty: The primary role of morality versus so- ciability and competence in information gathering. European Journal of Social Psychology, 41, 135 143. Byrne, D. 1971. The attraction paradigm. New York, NY: Ac- ademic Press. Cislak, A. 2013. Effects of power on social perception: All your boss can see is agency. Social Psychology, 44, 139 147. Clark, M.S. 1984. Record keeping in two types of relationships. Journal of Personality and Social Psychology, 47, 549 557. Clark, M.S.,  Mills, J. 1979. Interpersonal attraction in ex- change and communal relationships. Journal of Personality and Social Psychology, 37, 12 24. Clark, M.S.,  Mills, J. 1993. The difference between commu- nion and exchange relationships: What it is and is not. Person- ality and Social Psychology Bulletin, 19, 684 691. Cottrell, C.A., Neuberg, S.L.,  Li, N.P. 2007. What do people desire in others? A sociofunctional perspective on the impor- tance of different valued characteristics. Journal of Personality and Social Psychology, 92, 208 231. De Bruin, E.,  Van Lange, P. 1999. Impression formation and cooperative behavior. European Journal of Social Psychology, 29, 305 328. De Bruin, E.,  Van Lange, P. 2000. What people look for in others: Inferences of the perceiver and the perceived on infor- mation selection. Personality and Social Psychology Bulletin, 26, 206 219. Fiske, S.T., Cuddy, A.,  Glick, P. 2007. Universal dimensions of social cognition: Warmth and competence. Trends in Cog- nitive Science, 11, 77 83. Hogan, R. 1982. A socioanalytic theory of personality. Nebras- ka Symposium on Motivation, 55 89. Judd, C.M., James-Hawkins, L., Yzerbyt, V.,  Kashima, Y. 2005. Fundamental dimensions of social judgment: Under- standing the relations between judgments of competence and warmth. Journal of Personality and Social Psychology, 89, 899 913. Kelley, H.H. 1950. The warm cold variable in first impressions of persons. Journal of Personality, 18, 431 439. Kelley, H.H.,  Thibaut, J.W. 1978. Interpersonal relations: A theory of interdependence. New York, NY: Wiley-Intersci- ence. Kenworthy, J.,  Tausch, N. 2008. Beliefs about the utility and stability of trait attributions in an intergroup context: Differ- ences between warmth and competence. European Journal of Social Psychology, 38, 1121 1129. Paulhus, D.L.,  Trapnell, P.D. 2008. Self-presentation of per- sonality: An agency-communion framework. In O.P. John, R.W. Robins,  L.A. Pervin Eds., Handbook of personality psychology: Theory and research 3rd ed., pp. 492 517. New York, NY: Guilford. Peeters, G. 1992. Evaluative meanings of adjectives in vitro and in context: Some theoretical implications and practical conse- quences of positive-negative asymmetry and behavioral-adap- tive concepts of evaluation. Psychologica Belgica, 32, 211 231. Peeters, G. 2001. In search for a social-behavioral approach- avoidance dimension associated with evaluative trait mean- ings. Psychologica Belgica, 41, 187 203. Peeters, G. 2008. The evaluative face of a descriptive model: Communion and agency in Peabody s tetradic model of trait organization. European Journal of Social Psychology, 38, 1066 1072. Peeters, G.,  Czapinski, J. 1990. Positive-negative asymmetry in evaluations: The distinction between affective and informa- tional negative effects. In W. Stroebe  M. Hewstone Eds., European review of social psychology Vol. 1, pp. 33 60. London, UK: Wiley. Rusbult, C.E. 2000. Interdependence: Interdependence theory. In A.E. Kazdin Ed., Encyclopedia of psychology Vol. 4, pp. 330 331. Washington, DC: American Psychological As- sociation. Rusbult, C.E.,  Van Lange, P.A.M. 1996. Interdependence processes. In E.T. Higgins  A.W. Kruglanski Eds., Social psychology: Handbook of basic principles pp. 564 596. New York, NY: Guilford. Rusbult, C.E.,  Van Lange, P.A.M. 2003. Interdependence, interaction and relationships. Annual Review of Psychology, 54, 351 375. Scholer, A.A.,  Higgins, E.T. 2008. People as resources: Ex- ploring the functionality of warm and cold. European Journal of Social Psychology, 38, 1111 1120. Schwartz, S. H.,  Sagie, G. 2000. Value consensus and impor- tance: A cross-national study. Journal of Cross-Cultural Psy- chology, 31, 465 497. Suitner, C.,  Maas, A. 2008. The role of valence in the per- ception of agency and communion. European Journal of So- cial Psychology, 38, 1073 1082. Thibaut, J.W.,  Kelley, H.H. 1959. The social psychology of groups. New York, NY: Wiley. Uchronski, M. 2008. Agency and communion in spontaneous self-descriptions: Occurrence and situational malleability. Eu- ropean Journal of Social Psychology, 38, 1093 1102. Wojciszke, B.,  Abele, A.E. 2008. The primacy of communion over agency and its reversals in evaluation. European Journal of Social Psychology, 38, 1139 1147. Wojciszke, B., Baryla, W., Parzuchowski, M., Szymkow, A.,  Abele, A.E. 2011. Self-esteem is dominated by agency over A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits 93 2013 Hogrefe Publishing Social Psychology 2013; Vol. 442:84 94 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. communion information. European Journal of Social Psychol- ogy, 41, 617 627. Wojciszke, B., Bazinska, R.,  Jaworski, M. 1998. On the dom- inance of moral categories in impression formation. Personal- ity and Social Psychology Bulletin, 24, 1245 1257. Wortman, J.,  Wood, D. 2011. The personality traits of liked people. Journal of Research in Personality, 45, 519 528. Ybarra, O., Chan, E.,  Park, D. 2001. Young and old adults concerns about morality and competence. Motivation and Emotions, 25, 85 100. Ybarra, O., Chan, E., Park, H., Burnstein, E., Monin, B.,  Stanik, C. 2008. Life s recurring challenges and the fundamental di- mensions: An integration and its implications for cultural dif- ferences and similarities. European Journal of Social Psychol- ogy, 38, 1083 1092. Ybarra, O., Park, D., Stanik, C.,  Lee, D.S. 2012. Self-judg- ment and reputation monitoring as a function of the fundamen- tal dimensions, temporal perspective, and culture. European Journal of Social Psychology, 42, 200 209. Andrea E. Abele Social Psychology Group University of Erlangen-Nuremberg Bismarckstr. 6 91054 Erlangen Germany Tel. 49 9131 8522307 Fax 49 9131 8524731 E-mail abelephil.uni-erlangen.de Appendix Situation Descriptions Study 1 Mutual Dependence Relationships Imagine that you have to prepare for an exam. The study material is substantial, and you decide to ask another stu- dent, who has to pass the same exam, but whom you don t know yet, to prepare for this test together. Both of you will prepare half of the study material and will then teach it to one another. Imagine that you play tennis in a tennis team, and that you had a teammate you played double with. This team- mate has left the team, so you will now play with a new teammate who just has entered the team. Unilateral Dependent Relationships: Imagine that, due to time reasons, you were unable to attend a lecture. This lecture is highly relevant because you have to take an exam on its topics. You post a mes- sage whether another student could lend you hisher notes. Only one person, whom you did not know before, answers and is willing to lend you hisher notes. Imagine that you urgently need a lift to another city. On an internet platform you found someone who offers a lift to this city. You contact this person and arrange a meeting point for the next day. Independent Relationship Imagine that you have applied for a voluntary course that will last for two days. There are fixed seating ar- rangements, so that you will sit next to another student whom you have not met yet. Imagine that a new neighbor has moved into the flat next door to you. You do not yet know this person. Situation Descriptions Study 2 Exchange Relationship Imagine you are applying for a three-month internship, during which you will have to cooperate intensely with your supervisor. Communal Relationship Imagine you are moving to a new city and looking for a friend with whom you could perform a number of lei- sure activities. 94 A.E. Abele  S. Brack: Kind of Relationship and Preference for Other Persons Traits Social Psychology 2013; Vol. 442:84 94 2013 Hogrefe Publishing This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

text = content

keyword = 'finding'

pattern = r'\b' + re.escape(keyword) + r'\b(?:[\s\S]+?\b\w+\b){0,30}'

matches = re.findall(pattern, text)

for match in matches:
    words_after_result = match.split()[1:31]  
    print(f"Words after '{keyword}':", " ".join(words_after_result))

Words after 'finding': and reveals that – independent of respondents’ own agency and communion – agency traits are chosen less frequently in communal relationships than in exchange relationships. Study 3 shows that in
Words after 'finding': that can be covered under the present heading of “primacy of communion.” Characteris- tics such as intelligence were differently valued across tasks, goals, and functions. Assuming that “intelligence” belongs to
Words after 'finding': is also well in line with our prediction that the importance of agency traits varies with the kind of relationship. Individual Differences and Preference for Others’ Traits Preference for specific
Words after 'finding': is well in line with other areas of research, such as research on social values (Schwartz & Sagie, 2000) as well as a sociofunctional anal- ysis of relationships (Cottrell et
Words after 'finding': is in line with research in the context of the similarity–attraction paradigm (Byrne, 1971). Second, it extends this research by showing that similarity–attraction also works with respect to