INTRODUCTION
Marketing researchers increasingly recognize the fact that individuals may come to have perception- and affect-laden relationships with companies that may influence their personal behaviors and the behaviors of organizations they represent--behaviors that are relevant to those companies' business (Bengtsson 2003; Bhattacharya and Sen 2003; Fullerton 2005; Neville, Bell, and Mengilc 2005; Weiss, Anderson, and MacInnis 1999). It has been suggested that these relationships may even involve identification with the company (Ahearne, Bhattacharya, and Gruen 2005; Bhattacharya and Sen 2003; Cardador and Pratt 2006; Scott and Lane 2000), which is consistent with the increasingly popular notion that consumption, markets, and companies' products and brands are of significance in individuals' identity work, becoming entangled with their sense of self (e.g., Belk 1988; Fournier 1998: Holt 2002; McCracken 1986).
Moreover, there seems to be increasing appreciation among researchers in both marketing and management of the fact that all heterogeneous business-relevant actors, including organizations, are eventually represented by individuals (or individual agents) (Ahearne, Bhattacharya, and Gruen 2005; Freeman 1984; Rowley and Moldoveanu 2003). This is consistent with Jones's (1995) and Scott and Lane's (2000) view that the various stakeholders of a company basically consist of individuals and groups of individuals.
However, earlier marketing research focusing on individuals' perception- and affect-laden, identity-involving relationships to companies has not systematically examined how, specifically, such identity relationships to companies will stem from identity relationships to various other things that are relevant to people and associated with companies. Such relevant things include products (1) and brands, product categories, groups of people, communications and advertising, and even activities and areas of interest as well as ideas and ideals.
Besides, the earlier research has touched on the influence of such a company relationship on the individual's behaviors merely as a consumer or customer (Ahearne, Bhattacharya, and Gruen 2005; Bhattacharya and Sen 2003; Fournier 1998), or an employee (see Pratt 1998), or customer and employee (Cardador and Pratt 2006). Influences on the heterogeneous business-relevant behaviors of individuals, let alone the organizations they represent, have not been comprehensively examined--in a way that would refrain from assigning the individual to a certain pre-defined, exclusive stakeholder class, such as customers, employees, or investors. In other words, no such integrative theoretical conceptualization has so far emerged that would apply to the influence of individuals' perception- and affect-laden, identity-involving relationships to companies on their various behaviors, their stakeholder class notwithstanding.
In this article, we address these marketing research gaps by introducing an integrative theoretical framework that examines antecedents and behavioral consequences of a special kind of identity relationship that an individual may have with a company, which we term affective self-affinity (ASA). A further and even more fundamental contribution is to give a theoretical definition of ASA in the first place and to explicate how it works in principle. In so doing, we bring together certain existing theories in (social) psychologist consumer research that are characterized by roughly similar foci on the interactions between individuals' perceptions, affective evaluations, and identification with a variety of things but have not so far been integrated within one coherent theoretical conceptualization. The theories in question include those of self-concept congruence (S en and Bhattacharya 2001; Sirgy 1982), (social) identification (Ashforth and Mael 1989; Bergami and Bagozzi 2000; Bhattacharya and Sen 2003; Dukerich, Golden, and Shortell 2002; Hogg and Vaughan 2002; Scott and Lane 2000; Tajfel and Turner 1986), and enduring (or ego-) involvement (Celsi and Olson 1988; Higie and Feick 1989; Richins and Bloch 1986), as well as theories concerning the extended self (Ahuvia 2005; Belk 1988), (possession) attachment (Ball
In order to bring in the marketing perspective, we apply the ASA theory to a company and its business. We will first examine the antecedents of an individual's ASA relationship with a company, i.e., how it may stem from ASAs for other things such as products and brands, product categories, communications and advertising, activities and areas of interest, and ideas and ideals. Secondly, we will examine the business-relevant behavioral consequences of the individual's ASA for the company.
In this article, we address these marketing research gaps by introducing an integrative theoretical framework that examines antecedents and behavioral consequences of a special kind of identity relationship that an individual may have with a company, which we term affective self-affinity (ASA). A further and even more fundamental contribution is to give a theoretical definition of ASA in the first place and to explicate how it works in principle. In so doing, we bring together certain existing theories in (social) psychologist consumer research that are characterized by roughly similar foci on the interactions between individuals' perceptions, affective evaluations, and identification with a variety of things but have not so far been integrated within one coherent theoretical conceptualization. The theories in question include those of self-concept congruence (S en and Bhattacharya 2001; Sirgy 1982), (social) identification (Ashforth and Mael 1989; Bergami and Bagozzi 2000; Bhattacharya and Sen 2003; Dukerich, Golden, and Shortell 2002; Hogg and Vaughan 2002; Scott and Lane 2000; Tajfel and Turner 1986), and enduring (or ego-) involvement (Celsi and Olson 1988; Higie and Feick 1989; Richins and Bloch 1986), as well as theories concerning the extended self (Ahuvia 2005; Belk 1988), (possession) attachment (Ball
In order to bring in the marketing perspective, we apply the ASA theory to a company and its business. We will first examine the antecedents of an individual's ASA relationship with a company, i.e., how it may stem from ASAs for other things such as products and brands, product categories, communications and advertising, activities and areas of interest, and ideas and ideals. Secondly, we will examine the business-relevant behavioral consequences of the individual's ASA for the company.
We define an individual's ASA for a certain thing as the extent to which he/she perceives a positively affective congruence between the thing and his/her identity. This definition is mainly grounded on the widespread general notion that individuals have a reflective sense of self, and further derives especially from the above-mentioned theories about self-concept congruence and (social) identification, as well as those concerning effect and emotions (Damasio 1994, 2003; Slovic, Finucane, Peters, and MacGregor 2002a; Zajonc 1980). In building the theory further, we outline the principle of ASA transference, i.e., how an individual's ASA for one thing positively influences his/her ASA for another. We also consider principles of how ASA affects individual behavior: choice among similar alternative behaviors, committed and loyal behavior, and the adoption of supporting, interaction-seeking, and resource-awarding behavior. We apply these principles as theoretical grounds for a set of propositions concerning the antecedents of an individual's ASA for a company and its business-relevant behavioral consequences.
In providing these propositions we essentially deploy a piecemeal understanding of various ASA-related phenomena and concepts outlined in earlier marketing and consumer research. Thus, our contribution lies not only in theoretically defining ASA and explicating the principles of its transference and effects on behavior but also in explicitly framing and gathering together a coherent set of testable propositions concerning company-related ASAs as presented in earlier research implicitly or in a fragmented way. Accordingly, we also provide references to various streams of research that are supportive of or consistent with our propositions.
In providing these propositions we essentially deploy a piecemeal understanding of various ASA-related phenomena and concepts outlined in earlier marketing and consumer research. Thus, our contribution lies not only in theoretically defining ASA and explicating the principles of its transference and effects on behavior but also in explicitly framing and gathering together a coherent set of testable propositions concerning company-related ASAs as presented in earlier research implicitly or in a fragmented way. Accordingly, we also provide references to various streams of research that are supportive of or consistent with our propositions.
AFFECTIVE SELF-AFFINITIES
Definition. In this article, we define effective self-affinity (ASA) for a certain thing as the extent to which an individual perceives a positively affective congruence between the thing and his/her identity. Perceived congruence means that the individual perceives the thing to reflect his/her identity, i.e., overlap between the identity (attributes) of the thing and his/her own, personal identity (attributes) (cf. Dutton, Dukerich, and Harquail 1994; Marin and Ruiz 2007; Scott and Lane 2000). Note that below we use the terms "self' and "identity" interchangeably when referring to one's perceived image of oneself.
The definition presupposes the increasingly prevalent notion in consumer research that individuals have "a sense of self," being reflexively aware of and able to perceive, assess, and define themselves or their identities in relation or contrast to surrounding objects (Rosenberg 1979; for a recent review, see Reed 2002). Moreover, as Sirgy (1982: 291) has suggested, congruency assessments are not restricted to tangibles but also apply to intangibles such as services, persons, and abstract ideas, as well as to organizations. Accordingly, our notion is that an individual may have ASA for anything, including companies. Support for this is found in studies by Shimp and Madden (1988) and Ahuvia (2005), suggesting that an individual's relationship with any (consumption) object may involve positive feelings of effect, as well as identification. In any case, it is worth noting that in our view, congruency or overlap may be perceived between any attributed portion of one's identity and any attributed portion of the identity of the other thing--including, and often so, the totality or gestalt of the identities. This is to say that we advocate the "congruence as identity overlap" view (Dukerich, Golden, and Shortell 2002; Sen and Bhattacharya 2001; see also Bergami and Bagozzi 2000), rather than the far more constrained "congruence as similar personality adjectives" view (cf. Kressmann et al. 2006; Marin and Ruiz 2007; Sen and Bhattacharya 2001; Sirgy 1982). In the latter, congruence is seen narrowly as an individual's perception that similar human-like personality adjectives, traits, or characteristics, such as honest, enlightened, innovative, conservative, and funny apply to the other thing as well as to him/herself.
The definition presupposes the increasingly prevalent notion in consumer research that individuals have "a sense of self," being reflexively aware of and able to perceive, assess, and define themselves or their identities in relation or contrast to surrounding objects (Rosenberg 1979; for a recent review, see Reed 2002). Moreover, as Sirgy (1982: 291) has suggested, congruency assessments are not restricted to tangibles but also apply to intangibles such as services, persons, and abstract ideas, as well as to organizations. Accordingly, our notion is that an individual may have ASA for anything, including companies. Support for this is found in studies by Shimp and Madden (1988) and Ahuvia (2005), suggesting that an individual's relationship with any (consumption) object may involve positive feelings of effect, as well as identification. In any case, it is worth noting that in our view, congruency or overlap may be perceived between any attributed portion of one's identity and any attributed portion of the identity of the other thing--including, and often so, the totality or gestalt of the identities. This is to say that we advocate the "congruence as identity overlap" view (Dukerich, Golden, and Shortell 2002; Sen and Bhattacharya 2001; see also Bergami and Bagozzi 2000), rather than the far more constrained "congruence as similar personality adjectives" view (cf. Kressmann et al. 2006; Marin and Ruiz 2007; Sen and Bhattacharya 2001; Sirgy 1982). In the latter, congruence is seen narrowly as an individual's perception that similar human-like personality adjectives, traits, or characteristics, such as honest, enlightened, innovative, conservative, and funny apply to the other thing as well as to him/herself.
We anchor our conceptualization in the term "affinity" because possible alternatives such as "identification" and "attachment" are manifest in research traditions in which these terms and related theories apply mainly to specific types of objects (other persons, organizations, material objects) and/or are formed through certain kinds of mental processes not necessarily supported in our theoretical stance. Indeed, in terms of "identification," and according to social identity theory, individuals are seen to primarily identify with persons, groups, communities, and organizations (Ashforth and Mael 1989; Bhattacharya and Sen 2003; Dutton, Dukerich, and Harquail 1994; Hogg and Vaughan 2002; Pratt 1998; Scott and Lane 2000; Tajfel and Turner 1986). Moreover, most views of identification (e.g., Ashforth and Mael 1989; Bergami and Bagozzi 2000; Bhattacharya and Sen 2003; Dutton, Dukerich, and Harquail 1994) posit that underlying its formation is a cognitive process of self-categorization. For instance, organizational identification would arise from an individual's cognitive, active, and volitional assessment of the organization's identity in terms of its similarity to his/her own identity, its distinctiveness and prestige (Bhattacharya and Sen 2003). In contrast, the formation of ASA for a certain thing is not necessarily the result of an active and volitional act of (cognitive) assessment but is a partly unconscious and affective process (see the principle of ASA transference below). Thus, in line with Zajonc (1980), we claim that ASA-related affective identification does not necessarily need inferences. Zajonc argues convincingly that in order to arouse affect, objects do not necessarily have to be cognized much or with much, if any, information-processing effort, let alone on a level that is accessible to the subject's awareness: self-implicating affective reactions are often instantaneous and automatic. Thus, it is possible for us to have this affective reaction to something "before we know precisely what it is and perhaps even without knowing what it is" (p. 154, emphasis in the original).
As far as "attachment" is concerned, individuals are considered (a) in developmental psychology primarily to have attachments to persons (e.g., Bowlby 1979, 1980; Thomson, MacInnis, and Park 2005), and (b) in psychology and the philosophical view of possessions, to be attached to material objects and sometimes to brands (see Kleine and Baker 2004 for a review). Moreover, according to these two traditions, respectively, attachments are (a) formed in person-developmental or even biological processes, and/or (b) due to the natural possession-oriented mental nature or process of human beings. ASA does not call for either kind of process. Nor does it require strong feelings of love or passion towards the thing, which are often associated with "attachments" (Kleine and Baker 2004; Thomson, MacInnis, and Park 2005).
Thus, even if ASA has much in common with identification and attachment, in terms of the emotional significance attached to a thing and its relation to one's identity for example, we contend that it is not directed towards specific types of objects and that an individual may have ASA for anything, including people, groups of people, ideas and ideals, and activities and areas of interest, as well as product categories, products, brands, and companies. In terms of its formation, in turn, our stance emphasizes its partly unconscious and affective transference from one thing to another, as we discuss later.
We anchor our conceptualization to the term "affinity" because its intuitive meaning resonates well with our definition. This resonance is evident in the two meanings given to the word by The Free Dictionary: (1) natural attraction [to], liking [for], or feeling of kinship [with] [something], and (2) an inherent similarity between persons or things, or likeness deriving from kinship or from the possession of shared properties or sympathies. The positively affective perception suggested in the ASA definition resonates with (1), and the perception of congruence between a thing and one's identity with (2).
Convergence and divergence. ASA might be considered a two-dimensional construct. First, with regard to a thing for which an individual has ASA, his/her perception of the thing is tagged with a fair degree of positive affect (as opposed to negative affect). Note that research in the fields of psychology and even neurobiology increasingly demonstrates that all our perceptions of all things contain some effect, positive or negative (Damasio 1994, 2003; Slovic et al. 2002a; Zajonc 1980). Affect, in turn, manifests in positive vs. negative overall evaluations of things, i.e., liking vs. disliking (Berscheid 1983; MacGregor et al. 2000; Slovic et al. 2002a; Zajonc 1980). This first dimension converges somewhat with the concept of attitude. This holds insofar as attitude is considered an affective overall evaluation of a thing: an indication of the strength of a person's likes or dislikes (Ajzen and Fishbein 1980), or a summary evaluation of a psychological object along dimensions such as good-bad, harmful-beneficial, pleasant-unpleasant, and likable-dislikeable (Ajzen 2001).
Secondly, an individual perceives the thing as congruent with and thus personally relevant to his/her identity. This second dimension converges with certain dimensions of the concept of enduring (ego-) involvement. Perceived personal relevance is generally viewed as the essential characteristic of involvement (Celsi and Olson 1988; Higie and Feick 1989; Richins and Bloch 1986; Richins, Bloch, and McQuarrie 1992; Zaichkowsky 1985). In particular, the personal relevance implicated in enduring (or ego-) involvement (as opposed to situational involvement) is intrinsically (as opposed to situationally) motivated by the thing's being related to the individual's self-image (Higie and Feick 1989; Richins and Bloch 1986). Enduring involvement reflects relatively stable, enduring structures of personally relevant knowledge, derived from past experience and stored in the long-term memory (Celsi and Olson 1988).
However, ASA's second dimension is also distinct from enduring involvement in that ASA does not call for the hedonic component of fun derived from thinking about the thing in question (Higie and Feick 1989). On the one hand, an individual may have an ASA for a certain thing such as the idea of social responsibility, and a positive affective perception of it, and may consider it personally relevant or congruent, even if thinking about it would not yield any particular fun, enjoyment, or pleasure. On the other hand, even if thinking about a thing, such as an old joke, yields fun, enjoyment or pleasure, the individual does not necessarily have an ASA for it, and may not perceive it as personally relevant or congruent. Note also that an individual's ASA for a thing does not necessarily require the thing to be relevant to his/her identity in the strict self-defining sense that a perceived threat to the existence or welfare of the thing would lead to a sense of loss of self (cf. Ball and Tasaki 1992; Belk 1984, 1988). As far as ASA is concerned, it is sufficient that the thing is perceived to be congruent with one's identity, i.e. to reflect it: it does not have to be defining or relevant to the existence of one's self or identity.
Given the above conditions, ASA for a thing will converge, to some extent, with a confounded interaction of positive affect or attitude towards it and perception of it as being personally relevant to one's identity. As a confound, ASA may approximately converge with the product of positive attitude and personal relevance, yet diverge from these dimensions, respectively, if contrasted separately with only one of them. For instance, an individual may have a fairly positive attitude towards a thing--Toyota Corolla, for example--yet may not necessarily perceive it to be very relevant to his/her own identity. An individual may also perceive a thing--car-driving, for example--as being personally relevant and related to his/her identity, and yet not have a positive attitude towards it. In Sirgy's (1982) terms, the individual here could have a "self-image belief' ("I am a car-driving person"), which is nevertheless negatively valenced, i.e., the "self-image value" is negative ("I don't like being a car-driving person"). This may have something to do with the distinction between the actual vs. the ideal self (Belch and Landon 1977; Belch 1978; Dolich 1969; Sirgy 1982). For instance, car driving may be perceived as relevant to one's actual identity if one uses the car every day to go to work and/or feels a personal necessity to use it. Still, it may not be perceived as relevant to one's ideal identity, which, further, leaves it possible that one actually dislikes the very idea--and thus does not have ASA for it.
Finally, the idea of perceived congruency between a certain thing and one's self/identity gives ASA something in common with certain aspects of the rather vague notion of the thing being incorporated into one's extended self, as discussed by Belk (1988). However, as Cohen demonstrates (1989), Belk's notion is somewhat ill-defined and lacks internal coherence. Accordingly, we are content to state that ASA overlaps with the least controversial aspects of the extended-self notion, as pointed out by Cohen: the perception that the thing in question has an effect attached to it is assigned a special personal meaning and value and is perceived as personally relevant. However, as discussed above, it does not by definition necessitate that the thing has to be in any way (literally) part of one's self-definition (cf. Cohen 1989).
As far as "attachment" is concerned, individuals are considered (a) in developmental psychology primarily to have attachments to persons (e.g., Bowlby 1979, 1980; Thomson, MacInnis, and Park 2005), and (b) in psychology and the philosophical view of possessions, to be attached to material objects and sometimes to brands (see Kleine and Baker 2004 for a review). Moreover, according to these two traditions, respectively, attachments are (a) formed in person-developmental or even biological processes, and/or (b) due to the natural possession-oriented mental nature or process of human beings. ASA does not call for either kind of process. Nor does it require strong feelings of love or passion towards the thing, which are often associated with "attachments" (Kleine and Baker 2004; Thomson, MacInnis, and Park 2005).
Thus, even if ASA has much in common with identification and attachment, in terms of the emotional significance attached to a thing and its relation to one's identity for example, we contend that it is not directed towards specific types of objects and that an individual may have ASA for anything, including people, groups of people, ideas and ideals, and activities and areas of interest, as well as product categories, products, brands, and companies. In terms of its formation, in turn, our stance emphasizes its partly unconscious and affective transference from one thing to another, as we discuss later.
We anchor our conceptualization to the term "affinity" because its intuitive meaning resonates well with our definition. This resonance is evident in the two meanings given to the word by The Free Dictionary: (1) natural attraction [to], liking [for], or feeling of kinship [with] [something], and (2) an inherent similarity between persons or things, or likeness deriving from kinship or from the possession of shared properties or sympathies. The positively affective perception suggested in the ASA definition resonates with (1), and the perception of congruence between a thing and one's identity with (2).
Convergence and divergence. ASA might be considered a two-dimensional construct. First, with regard to a thing for which an individual has ASA, his/her perception of the thing is tagged with a fair degree of positive affect (as opposed to negative affect). Note that research in the fields of psychology and even neurobiology increasingly demonstrates that all our perceptions of all things contain some effect, positive or negative (Damasio 1994, 2003; Slovic et al. 2002a; Zajonc 1980). Affect, in turn, manifests in positive vs. negative overall evaluations of things, i.e., liking vs. disliking (Berscheid 1983; MacGregor et al. 2000; Slovic et al. 2002a; Zajonc 1980). This first dimension converges somewhat with the concept of attitude. This holds insofar as attitude is considered an affective overall evaluation of a thing: an indication of the strength of a person's likes or dislikes (Ajzen and Fishbein 1980), or a summary evaluation of a psychological object along dimensions such as good-bad, harmful-beneficial, pleasant-unpleasant, and likable-dislikeable (Ajzen 2001).
Secondly, an individual perceives the thing as congruent with and thus personally relevant to his/her identity. This second dimension converges with certain dimensions of the concept of enduring (ego-) involvement. Perceived personal relevance is generally viewed as the essential characteristic of involvement (Celsi and Olson 1988; Higie and Feick 1989; Richins and Bloch 1986; Richins, Bloch, and McQuarrie 1992; Zaichkowsky 1985). In particular, the personal relevance implicated in enduring (or ego-) involvement (as opposed to situational involvement) is intrinsically (as opposed to situationally) motivated by the thing's being related to the individual's self-image (Higie and Feick 1989; Richins and Bloch 1986). Enduring involvement reflects relatively stable, enduring structures of personally relevant knowledge, derived from past experience and stored in the long-term memory (Celsi and Olson 1988).
However, ASA's second dimension is also distinct from enduring involvement in that ASA does not call for the hedonic component of fun derived from thinking about the thing in question (Higie and Feick 1989). On the one hand, an individual may have an ASA for a certain thing such as the idea of social responsibility, and a positive affective perception of it, and may consider it personally relevant or congruent, even if thinking about it would not yield any particular fun, enjoyment, or pleasure. On the other hand, even if thinking about a thing, such as an old joke, yields fun, enjoyment or pleasure, the individual does not necessarily have an ASA for it, and may not perceive it as personally relevant or congruent. Note also that an individual's ASA for a thing does not necessarily require the thing to be relevant to his/her identity in the strict self-defining sense that a perceived threat to the existence or welfare of the thing would lead to a sense of loss of self (cf. Ball and Tasaki 1992; Belk 1984, 1988). As far as ASA is concerned, it is sufficient that the thing is perceived to be congruent with one's identity, i.e. to reflect it: it does not have to be defining or relevant to the existence of one's self or identity.
Given the above conditions, ASA for a thing will converge, to some extent, with a confounded interaction of positive affect or attitude towards it and perception of it as being personally relevant to one's identity. As a confound, ASA may approximately converge with the product of positive attitude and personal relevance, yet diverge from these dimensions, respectively, if contrasted separately with only one of them. For instance, an individual may have a fairly positive attitude towards a thing--Toyota Corolla, for example--yet may not necessarily perceive it to be very relevant to his/her own identity. An individual may also perceive a thing--car-driving, for example--as being personally relevant and related to his/her identity, and yet not have a positive attitude towards it. In Sirgy's (1982) terms, the individual here could have a "self-image belief' ("I am a car-driving person"), which is nevertheless negatively valenced, i.e., the "self-image value" is negative ("I don't like being a car-driving person"). This may have something to do with the distinction between the actual vs. the ideal self (Belch and Landon 1977; Belch 1978; Dolich 1969; Sirgy 1982). For instance, car driving may be perceived as relevant to one's actual identity if one uses the car every day to go to work and/or feels a personal necessity to use it. Still, it may not be perceived as relevant to one's ideal identity, which, further, leaves it possible that one actually dislikes the very idea--and thus does not have ASA for it.
Finally, the idea of perceived congruency between a certain thing and one's self/identity gives ASA something in common with certain aspects of the rather vague notion of the thing being incorporated into one's extended self, as discussed by Belk (1988). However, as Cohen demonstrates (1989), Belk's notion is somewhat ill-defined and lacks internal coherence. Accordingly, we are content to state that ASA overlaps with the least controversial aspects of the extended-self notion, as pointed out by Cohen: the perception that the thing in question has an effect attached to it is assigned a special personal meaning and value and is perceived as personally relevant. However, as discussed above, it does not by definition necessitate that the thing has to be in any way (literally) part of one's self-definition (cf. Cohen 1989).
How ASA works
We now move on to consider the principles governing ASAs. First, we discuss how an individual's ASA, for one thing, can contribute to his/her ASA for another, and this provides the basis for the later discussion on the antecedents of individuals' ASAs for companies. Secondly, we focus on how an individual's ASA can contribute to his/her behavior, which again provides the basis for the later discussion on the business-relevant behavioral consequences of individuals' ASAs for companies.
The principle of ASA transference--We contend that an individual's ASA for "thing x" positively influences the strength of his/her ASA for another "thingy", for the identity of which he/she perceives "thing x" to be essential and which he/she perceives to represent or support "thing x."
To the extent that an individual perceives a certain thing x (X) to be essential for the identity of thing y (Y), he/she will, by definition, perceive that Y somehow reflects the identity of X. Now, if this individual has ASA for X and thus perceives X to be congruent with his/her own identity, these are clearly the conditions under which he/she will perceive that Y reflecting X's identity, is also congruent with his/her own identity. Nevertheless, a necessary additional condition is that Y is perceived to represent (as opposed to misrepresent) or support (as opposed to suppressing) X: the mere perception that X is essential to Y's identity is not sufficient. In other words, if one perceives X to be congruent with one's identity on the one hand, and essential for the identity of Y on the other, but perceives Y to misrepresent or suppress X, it is likely that one will not perceive Y to be congruent with one's identity. Otherwise, one would perceive something (Y) that misrepresents or suppresses a reflection of one's own identity (X) as congruent with it, which is a contradictory (inconsistent) mental state that individuals will avoid (see Rosenberg 1979).
For example, an individual's ASA for the idea of the 'fight against cancer will positively influence his/her ASA for products or companies perceived to support it, such as biotechnology companies with a perceived identity related to the development of cancer drugs, it will not positively influence ASA for products (or companies) perceived to suppress the fight, such as tobacco (companies).
Furthermore, the positive effect attached to the perception of X will transfer to the perception of Y, because individuals are motivated and tend to exhibit consistency in their attitudinal evaluations (Abelson, Aronson, McGuire, Newcomb, Rosenberg, and Tannenbaum 1968; Festinger 1957; McGuire 1969)--and particularly consistency among their self-related attitudes (Epstein 1980; Rosenberg 1979; Sirgy 1982). Given people's positively affective evaluation of X, and since the support for or representation of X by Y is a positively associative link (see Heider 1946; Jacoby and Mazursky 1984), individuals tend towards consistency by assimilating their (neutral) evaluation of Y towards the positive evaluation of X. Note that the claim here is that perceived support, being a positively associative link, can indeed transfer the effect alternatively to perceived representation (representation that may also stem from perceiving Y to belong to a category defined by X, cf. Boush and Loken 1991; Wanke, Bless, and Schwartz 1998). Moreover, if an individual's perception and evaluation of X is positively affective and he/she perceives that not only X but also Y reflects his/her identity, it is mentally consistent for him/her to have a positive affective perception and evaluation of Y, whereas having a negatively affective evaluation would not be.
We maintain that this perceptual-evaluative process of ASAs for one thing "transferring" to another, including the case when a new thing is identified with associations to others, may be based not only or not necessarily on conscious cognitive processing but also on fast unconscious emotional reactions (see Damasio 1994, 2003; Slovic et al. 2002a; Zajonc 1980). Moreover, we also maintain that (repeated) perceptions of (previously) identified things result in rather stable, enduring structures of self-implicated knowledge about and ASAs for the different things stored in the long-term memory--as is the case with effect (Zajonc 1980), feelings or emotions (Damasio 1994; Damasio 2003), and involvement (Celsi and Olson 1988). This is also consistent with the theory and evidence related to the acquisition of a liking for vs. a disliking of objects through associative learning: specifically, the associative transfer of valence, which is often referred to as evaluative conditioning (EC) (Levey and Martin 1975; for a review, see De Houwer, Thomas, and Baeyens 2001). EC produces changes in the liking of a stimulus by pairing that stimulus with other positive or negative stimuli and has been shown to lead to object effects that are fairly resistant to "extinction," or diminution and elimination, after initial acquisition. In any case, we also maintain that the individual does not have to actively cognize, all the time, his/her ASA for a thing for which he/she has ASA: this may occur only during temporary states of activation or salience (see Forehand, Deshpande, and Reed 2002; Reed 2004). Yet, when trying to recall, recognize, or retrieve the object, one's ASA for it, as an effective quality of the original input, is likely to be the first element to emerge (see Zajonc 1980).
Finally, and to be more precise, we specify that the extent to which an individual's ASA for X positively influences the strength of his/her ASA for Y is moderated by the extent to which he/she perceives (i) X to be essential for Y's identity and (u) Y to represent or support X. This is important in our logic, since otherwise a strong ASA, for one thing, could embrace all things in the world for whose identity one perceives the thing to be marginally essential and to marginally represent or support the thing. Nevertheless, we still maintain that an individual may have ASAs--stronger or weaker--for a great variety and number of things, perceiving them, with positive affect, as congruent with his/her identity. As posited in symbolic interactionist identity theory (see Kleine, Kleine, and Kernan 1993; Reed 2004; Solomon 1983), ASAs for various things and the perceived associations between them and from them to other things, then, form a (consistent) self-structure in which the ASAs fit into an overall constellation and enact some parts of one's identity therein.
Finally, and to be more precise, we specify that the extent to which an individual's ASA for X positively influences the strength of his/her ASA for Y is moderated by the extent to which he/she perceives (i) X to be essential for Y's identity and (u) Y to represent or support X. This is important in our logic, since otherwise a strong ASA, for one thing, could embrace all things in the world for whose identity one perceives the thing to be marginally essential and to marginally represent or support the thing. Nevertheless, we still maintain that an individual may have ASAs--stronger or weaker--for a great variety and number of things, perceiving them, with positive affect, as congruent with his/her identity. As posited in symbolic interactionist identity theory (see Kleine, Kleine, and Kernan 1993; Reed 2004; Solomon 1983), ASAs for various things and the perceived associations between them and from them to other things, then, form a (consistent) self-structure in which the ASAs fit into an overall constellation and enact some parts of one's identity therein.