Social Interest and Differentiation of Self

Patrick Johnson
Portland State University

Adina J. Smith
Montana State University

Author Note

Patrick Johnson, Ph.D., is associate professor and department chair in the Department of Counselor Education, Portland State University. Adina Smith, Ph.D., is associate professor and Clinic Director in the Department of Health and Human Development, Montana State University.

Correspondence concerning this article should be addressed to: Patrick Johnson at the Department of Counselor Education, Portland State University, P.O. Box 751, Portland, OR 97207-0751.


In this study, we assessed the relationship between differentiation of self and social interest, providing a family systems conceptualization of the Adlerian concept. Participants were 813 college student volunteers who completed measures of differentiation and social interest. Results indicate that various dimensions of differentiation have unique effects on social interest. Implications of these results are discussed.

Keywords: social interest, differentiation, Adler, Bowen, family systems theory

Social Interest and Differentiation of Self 

     Social interest is the hallmark concept of Adlerian theory. Early translations from German defined social interest as “communal feeling” or “communal sense” (Ansbacher, 1968/1991). Adler theorized social interest as “a feeling of community, an orientation to live cooperatively with others, and a lifestyle that values the common good above one’s own interests and desires” (Guzick, Dorman, Groff, Altermatt, & Forsyth, 2004; p. 362). It has also been conceptualized as an active interest in the welfare of humankind, and identification and empathy with others (Ansbacher & Ansbacher, 1956).

     Social interest is typically viewed as being predictive of individuals’ adjustment and psychological health (Ansbacher, 1968/1991; Manaster, Zeynep, & Knill, 2003). According to Adler, social interest protects individuals against feelings of inferiority and promotes better coping and a healthier attitude toward stressful situations (Crandall & Putman, 1980). Richardson and Manaster (2003) contend that “the ideal of social interest is intended to inspire a way of life that counters such disconnectedness and helplessness with a sense of wider belonging and purpose” (p. 124). Adler emphasized that inadequate development of social interest in children and adolescents leads to psychological difficulties in adulthood (Guzick et. al., 2004; Maltby, Macaskill, Day, & Garner, 1999; Nikelly, 2005; Sweitzer, 2005).           

     According to Adler, parent-child relationships and family life affect the development of social interest, a contention which has been supported by research. For example, Johnson, Smith, and Nelson (2003) found that higher levels of family cohesion and expressiveness and lower levels of conflict were associated with higher levels of social interest in young adults. These results are in-line with prior research that indicate social interest is higher in individuals whose families were helpful and supportive, emotionally and behaviorally expressive, and did not experience excessive conflict (Amerikaner, Monks, Wolfe, & Thomas, 1994; Leak & Williams, 1989).

     The relationship between family variables and social interest seems to suggest an optimal level for cohesion and closeness of family members (Johnson et al., 2003). Amerikaner et al. (1994) suggested that when there is not an adequate level of family cohesion, members may feel their sense of security is threatened. Conversely, very high levels of family cohesion may be problematic (Johnson, Thorngren, & Smith, 2001), as “overly cohesive” families may be considered synonymous with the concept of fusion (Bowen, 1976, 1978). Fusion is marked by the tendency for family members to exhibit discomfort with separateness in their relationships. Such dependency could extend to a need to be fused with the larger community and may look like social interest, but may not be psychologically healthy for individuals. Thus, a Bowen conceptualization of relationships may enhance the understanding of social interest. 

     A principle concept of intergenerational family systems theory is differentiation, which is a balance between the life energies of togetherness and individuality (Bowen, 1976, 1978; Kerr & Bowen, 1988). Level of differentiation is determined by how well individuals are able to separate their emotional reactivity from logical reasoning (Skowron & Dendy, 2004; Titelman, 1998b). Greater differentiation is characterized by higher levels of emotional and intellectual functioning, including greater flexibility, adaptability, and ability to deal with stress (Schwartz, Thigpen, & Montgomery, 2006). Differentiation is also associated with the Bowen concept of the I Position, which reflects the degree to which individuals have a clearly defined sense of self, stick to their convictions, and take action based on those convictions even in the face of criticism from others (Bowen, 1978; Johnson, Buboltz, & Seemann, 2003).

     The ability to differentiate is largely determined by how an individual’s nuclear family manages anxiety related to balancing separateness and togetherness. Projection of anxiety onto children typically produces lower levels of differentiation.

When individuals are overwhelmed by emotionality and anxiety in their family, they are subject to fusion and emotional cutoff (Kerr & Bowen, 1988). Fusion can be defined as the extent to which family members borrow or lend their “self” to one another. Emotional oneness between family members occurs such that individuals have not formed a separate self (Titelman, 1998a).

     Emotional cutoff relates to how individuals manage unresolved emotional attachments with their parents (Bowen, 1978). When fusion between parents and children is extreme, anxiety increases and individuals react by reactively distancing from their parents or family (Skowron, Holmes, & Sabatelli, 2003; Titelman, 1998a). Cutoff serves to decrease anxiety when intimacy is perceived as threatening.

     In this study, we assessed the relationship between differentiation and social interest. Our main research question was: Does level of differentiation predict level of social interest? We hypothesized that the various dimensions of differentiation would be predictive of level of social interest. Specifically, we hypothesized that higher levels of emotional cutoff and I position would predict less social interest while higher levels of fusion would predict more social interest. We also hypothesized that higher levels of emotional reactivity would not significantly predict social interest, because individuals with high levels of reactivity tend to both seek close relationships and to reactively create distance in their relationships (Johnson & Waldo, 1998).


Procedures and Participants

     Participants were 813 volunteers who were enrolled in classes at a large U.S. university. Researchers collected data at nineteen undergraduate classes across two semesters. After providing written consent, participants completed standardized instruments and demographic questions. After collecting the instruments, researchers provided a brief lecture about the theoretical concepts involved in the study. Participants included 563 women (69%), 222 men (27%), and 28 who did not indicate gender (4%). Of the participants, 795 provided information on age (= 21.26 years, SD = 5.91). Of those responding, there were 727 Caucasian/Anglo (89%) Americans, 29 Native Americans (4%), 14 Asian Americans (2%), 11 Hispanic/Latino Americans (1%), 7 African Americans (1%), and 17 who indicated “other” (2%). 


     Differentiation of Self Inventory (DSI). The DSI (Skowron & Friedlander, 1998) is comprised of 43 items, which are rated on a 6-point scale, that generate four subscale scores. The first subscale, Emotional Reactivity, reflects the degree to which a person responds to environmental stimuli with emotional flooding, lability, or hypersensitivity. The second subscale, I Position, reflects a clearly defined sense of self and the ability to thoughtfully adhere to one’s convictions. The third subscale, Emotional Cutoff, reflects feeling threatened by intimacy, fears of engulfment, and use of defensive behaviors such as distancing. The fourth subscale, Fusion with Others, reflects emotional overinvolvement with others.

     Skowron & Friedlander (1998) provided information about the psychometric properties of the DSI. Construct validity was supported as the DSI correlated highly with a measure of chronic anxiety and with symptomatic distress. Across several studies, internal consistency coefficients supported moderate to high reliabilities for each of the four subscales (Emotional Reactivity = .88, Emotional Cutoff = .79, Fusion with Others = .70, and I Position = .85).

     The Social Interest Scale (SIS). The SIS (Crandall, 1981) consists of 15 items and 9 “buffer” items, which reduce the transparency of the scale. Crandall indicated that the SIS “was designed to assess the degree of a person’s interest in and concern for others” (p. 24). 

     Crandall (1981) provided psychometric information about the SIS, including test-retest reliability scores of .82 over 5 weeks and .65 over 14 months. Internal reliability coefficients were .73 and .71. Corrected odd-even reliabilities produced reliability coefficients of .77 and .73. Construct validity was supported by the SIS being positively related to concepts such as cooperation, altruism, and empathy and negatively related to criminal behavior and self-centeredness. Further support of the SIS was reported by Watkins (1994) who concluded “the SIS has been found to possess good reliability and validity across a number of studies” (p. 90). However, a more recent meta-analysis by Bass, Curlette, Kern, and McWilliams (2002) pointed out psychometric challenges of the SIS, which will be discussed in the Limitations section. 


     To assess the impact of differentiation and the unique effects of each component of differentiation on levels on social interest in young adults, we used simultaneous multiple regression analyses. The full model significantly predicted social interest, F (4, 764) = 29.22, p < .001, R2 = .13. More specifically, higher levels of emotional cut off significantly predicted lower levels of social interest, F (1, 767) = 42.17, p < .001. Higher levels of fusion significantly predicted higher levels of social interest, F (1, 767) = 27.77, p < .001. Higher levels of I Position significantly predicted lower levels of social interest, F (1, 767) = 5.72, p < .05. Emotional reactivity did not significantly predict levels of social interest. 


      Results of this investigation support our hypotheses that differentiation variables are related to the development of social interest. Results indicate that higher levels of fusion are associated with higher levels of social interest. Fused individuals tend to invest excessive amounts of energy into togetherness and to be uncomfortable with separateness. Such fusion could extend to the larger community. As was discussed by Johnson et al. (2003), excessive need for togetherness could mimic high levels of social interest but not be psychologically healthy for individuals.

     Results indicate that high levels of emotional cutoff are associated with lower levels of social interest. This suggests that individuals who are uncomfortable with closeness in relationships may not be actively involved in the larger community, indicating disconnection from a broader life-purpose as well as lack of empathy and altruism.

     In addition, higher levels of I position were predictive of lower levels of social interest. Theoretically, I position designates a “healthy independence,” or an individual’s ability to hold on to one’s own thoughts and beliefs despite pressure to do otherwise. Strong personal boundaries and inner convictions, however, may make it difficult for these individuals to work within the goals of a larger community. It may also be that the concept of social interest needs to be re-considered to take into account the idea of finding a healthy balance between separateness and togetherness in relationships.

     Lastly, results did not support an association between emotional reactivity and social interest. It may be that highly reactive individuals have difficulty exhibiting interest in others and the larger community because, although drawn to relationships, they become overwhelmed by them. They tend to want to be close to others, but reactively create distance in their relationships (Johnson & Waldo, 1998).

     Taken together, these results suggest that very high levels of social interest may be as unhealthy as low levels. Very high levels of social interest may suggest an addiction to relationships or “co-dependency” whereas low levels may suggest disconnectedness, “too much” independence, or decreased sense of belonging. Thus, rather than assuming that social interest at any level is positive, there may be an optimal level of social interest that facilitates psychological health through connectedness with others while still maintaining the ability to be separate.

     It may be useful for counselors to assess the underlying motivations of their clients’ social interest, especially if the social involvement seems excessive. By using a differentiation lens, counselors could assess if their clients community-oriented activities are a form of fusion, for example. Conversely, clients exhibiting emotional cutoff could be supported in increasing their social interest. An overarching, therapeutic goal could be to encourage clients in their efforts to achieve a balance of social involvement and self-directed behaviors, which is the essence of healthy differentiation.

Limitations and Future Directions

     Possible limitations of this investigation are related to the definition and function of some of the theoretical constructs. Bass and colleagues (2002) noted that there is no operational definition for social interest; the various measures appear to assess different dimensions of social interest. The SIS defines social interest as “the degree of a person’s interest and concern for others” (Crandall, 1981; p. 24). As explored in the introduction, this definition does not appear to fully encompass such a complex construct as social interest. Further, as data from this investigation suggests, when social interest as defined by the SIS is examined through the lens of Bowen theory, extremely high and low levels of social interest do not appear to be psychologically healthy.

     It is also important to acknowledge that the results of this investigation support an inherently Western cultural ideology of individuality and togetherness. The concept of differentiation has been challenged in its relevance to other cultures in that it appears to draw upon an individualistic moral vision that possibly limits its potential universal applicability (Christopher & Smith, 2006). As stated by Guisinger and Blatt (1994) “Western psychologies have traditionally given greater importance to self-development than to interpersonal relatedness, stressing the development of autonomy, independence and identity as central factors to the mature personality.” (p. 104)

     In furthering this challenge to the individualistic bias of Western psychology, Mozdzierz and Krauss (1996) link this notion to Adlerian theory and particularly the relationship between a wider social interest and psychological health. Schumaker (2001) acknowledges that depression in Western culture is related to a focus on consumption and lack of social interest. Carlson and Carlson (2000) suggest that Adlerian psychotherapists tend to focus on cooperation and socially oriented values as opposed to competitive and individualistic values, which characterize Western culture. 

     Based on this conceptualization, several directions for future research emerge. First, social interest should be examined to determine an optimum healthy range for individuals, including determining the levels at which individuals are overly self-involved or self-sacrificing. Second, as suggested by Johnson et al. (2003), it will be important to further examine the relationship between the different aspects of differentiation and social interest “to determine a healthy balance between autonomy and togetherness.” (p. 289) Third, the sample for this investigation was drawn from a college student population with an ethnic background that is primarily Anglo/Caucasian American. Further cross-cultural examinations of differentiation and social interest seem warranted. As suggested above, differentiation may be a construct based mainly on Western cultural values. The concept of the individual (and the self) likely holds different meanings, values, and visions in different cultures.


Amerikaner, M., Monks, G., Wolfe, P., & Thomas, S. (1994). Family interaction and individual psychological health. Journal of Counseling and Development, 72.

Ansbacher, H. L. (1968/1991). The concept of social interest. The Journal of Individual Psychology, 24(2), 28-46.

Ansbacher, H. L., & Ansbacher, R. R. (1956). The Individual Psychology of Alfred Adler. New York: Basic Books. 

Bass, M. L., Curlette, W. L., Kern, R. M., & McWilliams, A. E., Jr. (2002). Social interest: A meta-analysis of a multidimensional construct.  The Journal of Individual Psychology, 58(1), 4-34.

Bowen, M. (1976). Theory in the practice of psychotherapy. In P. J. Guerin Jr. (Ed.), Family Therapy: Theory and Practice. New York: Garner.

Bowen, M. (1978). Family therapy in clinical practice. Northvale, NJ: Jason Aronson.

Carlson, J. M., & Carlson, J. D. (2000). The application of Adlerian psychotherapy with Asian-American clients. The Journal of Individual Psychology, 56(2), 214-225.

Christopher, J. C., & Smith, A. J. (2006). A hermeneutic approach to culture and psychotherapy. In R. Moodley & S. Palmer (Eds.), Race, Culture and Psychotherapy: Critical Perspectives in Multicultural Practice. Philadelphia, PA: Routledge/Taylor & Francis Group. 

Crandall, J. E. (1981). Theory and measurement of social interest: Empirical tests of Alfred Adler's concept. New York: Columbia University Press.

Crandall, J. E., & Putman, E. (1980). Relations between measures of social interest and psychological well-being. The Journal of Individual Psychology, 36(2), 156-168.

Guisinger, S., & Blatt, S. J. (1994). Individuality and relatedness: Evolution of a fundamental dialectic. American Psychologist, 49, 104-111.

Guzick, D. T., Dorman, W. J., Groff, T. S., Altermatt, E. R., & Forsyth, G. A. (2004). Fostering social interest in schools for long-term and short-term outcomes. The Journal of Individual Psychology, 60(4), 361-366.

Johnson, P., Buboltz Jr., W. C., & Seemann, E. (2003). Ego identity status: A step in the differentiation process. Journal of Counseling and Development, 81 191-195.

Johnson, P., Smith, A. J., & Nelson, M. D. (2003). Predictors of social interest in young adults. The Journal of Individual Psychology, 59(3), 281-292.

Johnson, P., Thorngren, J., & Smith, A. (2001). Parental divorce and family functioning: Effects on differentiation levels of young adults. The Family Journal, 9(3), 265-272.

Johnson, P., & Waldo, M. (1998). Integrating Minuchin's boundary continuum and Bowen's differentiation scale: A curvilinear representation. Contemporary Family Therapy, 20(3), 403-413.

Kerr, M. E., & Bowen, M. (1988). Family evaluation: An approach based on Bowen theory. New York: Norton.

Leak, G. K., & Williams, D. E. (1989). Relationship between social interest and perceived family environment. The Journal of Individual Psychology, 45(3), 363-368.

Maltby, J., Macaskill, A., Day, L., & Garner, I. (1999). Social interest and Eysenck's personality dimensions. Psychological Reports, 85, 197-200.

Manaster, G., J., Zeynep, C., & Knill, M. (2003). Social interest, the individual and society: Practical and theoretical considerations. The Journal of Individual Psychology, 59(2), 109-122.

Mozdzierz, G. J., & Krauss, H. H. (1996). The embeddedness of Alfred Adler in modern psychology: Social policy and planning implications. The Journal of Individual Psychology, 52(3), 224-236.

Nikelly, A. G. (2005). Positive health outcomes of social interest. The Journal of Individual Psychology, 61(4), 329-342.

Richardson, F. C., & Manaster, G., J. (2003). Social interest, emotional well-being, and the quest for civil society. The Journal of Individual Psychology, 59(2), 123-135.

Schumaker, J. (2001). The age of insanity:Modernity and mental health. Westport, CT: Praeger.

Schwartz, J. P., Thigpen, S. E., & Montgomery, J. K. (2006). Examination of parenting styles of processing emotions and differentiation of self. The Family Journal, 14(1), 41-48.

Skowron, E., & Dendy, A. K. (2004). Differentiation of self and attachment in adulthood:  Relational correlates of effortful control. Contemporary Family Therapy, 26(3), 337-357.

Skowron, E., & Friedlander, M. L. (1998). The Differentiation of Self Inventory:  Development and initial validation. Journal of Counseling Psychology, 45, 235-246.

Skowron, E., Holmes, S. E., & Sabatelli, R. M. (2003). Deconstructing differentiation: Self regulation, interdependent relating, and well being in adulthood. Contemporary Family Therapy, 25(1), 111-129.

Sweitzer, E. M. (2005). The relationship between social interest and self-concept in conduct disordered adolescents. The Journal of Individual Psychology, 61(1), 55-79.

Titelman, P. (1998a). Emotional cutoff in Bowen family systems theory: An overview. In P. Titelman (Ed.), Emotional cutoff: Bowen family systems theory perspectives (pp. 9-65). New York: The Haworth Clinical Practice Press.

Titelman, P. (1998b). Overview of the Bowen theoretical-therapeutic system. In P. Titelman (Ed.), Clinical applications of Bowen family systems theory. (pp. 7-48) New York: The Haworth Press.

Watkins, C. E. (1994). Measuring social interest. The Journal of Individual Psychology, 50(1), 69-96.