Identities: A developmental social-psychological perspective

ABSTRACT

In this contribution, we review research that uses a cross-fertilisation approach to integrate developmental and social-psychological perspectives on how identities are formed and changed over time and how identity processes are genuinely social, being embedded in social contexts and fed by social contents. First, we outline the three-factor identity model as a parsimonious approach to understanding the dynamics of identity development. Second, we review empirical studies with longitudinal approaches to shed light on how identity processes are embedded in key contexts such as family, friendships and society at large through behaviours such as civic engagement. Third, we discuss the interplay between personal and social identities. We conclude by highlighting how adopting a cross-fertilisation approach that combines social-psychological and developmental perspective can significantly advance the theoretical understanding of identity dynamics. Finally, we address similarities and differences between personal identity and social identity approaches, and we provide an agenda for future research.

Wondering about one’s own identities is one of the most human endeavours at almost at any stage of life (Crocetti et al., 2018). Identity questions are posed at the individual level (Who am I? Who are you?) and at the collective level (Who are we? Who are they?). Given the centrality of identity questions to the human experience it is not surprising that “identity is one of the most commonly constructs in the social sciences” (Vignoles et al., 2011, p. 1). Two main research traditions have devoted considerable attention to the study of identity. Identity theorisations and research advanced in developmental and social psychology have paved the way to studying personal and social identities, respectively. These two traditions have largely followed different “streams”, focusing on different identity facets, using different assessment methods, and addressing different implications of identity, as briefly outlined below.

In developmental psychology, the Eriksonian and neo-Eriksonian approaches (for reviews, see, Schwartz, 2001; Schwartz et al., 2014) have focused on personal identity. Within this tradition, identity is mainly conceptualised in terms of the processes by which individuals commit to meaningful life domains and, thus, arrive at a personal synthesis of different identifications and experiences (Erikson, 1968). Research conducted within this framework has relied heavily on longitudinal methods to tackle how individuals develop their identity over time, which factors affect the development of identity, and what the implications of different developmental trajectories are (for reviews, see, e.g., Kroger et al., 2010; Meeus, 2011). In terms of implication, most attention has been paid to the effects of identity processes on mental health and adjustment at the individual level (e.g., anxiety; Lillevoll et al., 2013).

In social psychology, drawing on the seminal contribution of Tajfel and Turner (1979), social identity is conceptualised as ” … the part of an individual’s self-concept which derives from his membership of a social group (or groups) together with the value and emotional significance attached to that membership” (Tajfel, 1978, p. 63). Within this tradition, experimental methods have primarily been used to examine how social identity can provide a heuristic framework for understanding intra-group (e.g., sense of loyalty to one’s group) and inter-group processes (Ashmore et al., 2004; Brewer & Gardner, 1996; Postmes et al., 2005). In terms of implications, social identity has mainly been related to social and collective outcomes, such as intergroup attitudes and discrimination (e.g., Brown, 2000; Spears, 2011).

In this contribution, we review a novel line of research that fruitfully integrates these developmental and social-psychological perspectives by adopting a cross-fertilisation approach. We start from two main reference points: the first is that, as emphasised by the developmental tradition, identity is a life-long endeavour, in the sense that identity is never achieved once and for all, but is subjected to continual changes. Hence, a theoretical developmental framework and longitudinal methods are necessary to understanding such plasticity. The second anchor point is that, as underscored by the social-psychological perspective, identity is never constructed in a vacuum; it is fed by social memberships and experiences of individuals. Thus, identity investigation needs to be the research into “identity development in context” (Bosma & Kunnen, 2008; Branje, 2022). By bridging developmental and social-psychological “streams” upon these two pillars, we sought to unravel how identities are formed and changed over time and how identity processes are genuinely social by being embedded in social contexts.

In this review, we address the steps taken to reach this goal. First, we outline the three-factor identity model (Crocetti et al., 2008) as a parsimonious approach to understanding the dynamics by which identity develops over time and discuss how identity processes are intertwined with socio-cognitive identity strategies (Berzonsky, 2011) and other components of the self-concept, such as self-concept clarity (Campbell et al., 1996). The second section reviews empirical studies that use a longitudinal approach to shed light on how self and identity processes are embedded in social reality, considering both proximal (family and friendships) and distal (society) contexts. Third, we address the interplay of personal and social identities and review empirical evidence. We conclude by highlighting the theoretical and practical implications of this integrative line of research.

A parsimonious model to study iterative identity cycles

This section reviews the literature on personal identity that, inspired by Erikson’s (1950, 1968) seminal work, has progressively shed light on the iterative dynamic by which identity is formed and changed over time. Theoretical advancements have led to the development of process-oriented models (Meeus, 2011) that have provided a fertile ground for studying “identity-in-context” and tackling the interplay of personal and social identities (as further discussed in sections two and three). We also introduce the importance of longitudinal methods as the most appropriate approach to trace identity developmental trajectories.

From Erikson’s psychosocial theory to identity process-oriented models

Erikson’s (1950, 1968) psychosocial theory has inspired research on personal identity up to the present time (Schachter & Galliher, 2018). A fundamental tenet of Erikson’s view is that identity formation is a core task that individuals strive to address throughout their entire life span, especially when they cope with important changes, which might undermine their sense of stability (Erikson, 1950, 1968). This happens especially in adolescence when rapid multiple changes taking place at the biological, cognitive, emotional and social levels make identity questions particularly salient. For this reason, Erikson (1968) conceptualised the conflict between identity versus identity confusion as the primary developmental task of adolescence. Adolescents who adequately perform this task reach a condition of identity achievement, combining and integrating relevant earlier identifications in a unique and personal guise. In contrast, young people who fail in this task remain in a state of identity confusion, in which they miss meaningful commitments that could provide them with a sense of direction.

Marcia’s (1966) identity status paradigm is probably the most well-known elaboration of Erikson’s views on identity formation. Marcia conceptualised identity statuses as an individual’s style of coping with the identity crisis described by Erikson. The author proposed that, in addition to Erikson’s two poles (i.e., identity versus identity confusion), other statuses should be differentiated based on the extent to which individuals have committed to significant life domains, after having explored, or otherwise, the available options. Thus, Marcia introduced commitment and exploration as the two key processes to differentiate four identity statuses: achievement (a commitment is made after active exploration of different alternatives); foreclosure (a commitment is made without exploring other options); moratorium (a commitment has not been made yet, the exploration phase is still ongoing); and diffusion (in this status both commitment and exploration are absent).

Consistent with Marcia’s (1966, 1980) view, the identity status paradigm has mainly been applied to the study of inter-individual differences among youth classified into various identity statuses. On the one hand, a large corpus of evidence highlighted that adolescents in different identity statuses could be further differentiated in terms of personality characteristics and psychosocial problems (for a review see, Kroger & Marcia, 2011). On the other hand, longitudinal studies started to provide preliminary evidence showing that although during adolescence progressive (e.g., from foreclosure to achievement) are more common than regressive changes (e.g., from achievement to moratorium), only a small percentage of youth (about 20%) reaches the status of identity achievement by the end of adolescence (Kroger et al., 2010). This indicates that identity is not achieved once and for all, especially in post-modern societies, but is a life-long task, as Erikson (1950) originally theorised.

Marcia’s (1966) identity status paradigm has been criticised for not fully capturing the process by which people can question and change their identity over time (e.g., Côté & Levine, 1988). To address this problem, neo-Eriksonian models and conceptualisations have been proposed to uncover further identity processes and dynamics (for reviews cf., McLean & Syed, 2015; Schwartz et al., 2011). In this fertile ground, the three-factor model has been proposed.

The three-factor identity model: its origins and assumptions

Starting from the 1980s, awareness of the importance of studying the process of identity formation increased. In this context, Bosma (1985) and Meeus (1996) reflected on the meaning of commitment and exploration, and underlined the importance of (a) studying their intensity, rather than their presence or absence, as initially done in the identity status paradigm, and (b) considering different functions they might serve. In line with these considerations, Bosma (1985) differentiated between commitment making and identification with commitment, underscoring that making a choice (e.g., voting for a specific political party) does not necessarily mean that individuals identify strongly with it. Meeus (1996) distinguished past exploration, needed to find new commitments, and present exploration, necessary to validate existing commitments. This differentiation implies that if a person has weighed up different alternatives before making a choice, he/she does not necessarily continue to explore the meaning of commitment in the present.

Within this renewed conceptual framework, we have proposed the three-factor identity model (Crocetti et al., 2008). Our model assumes that identity is formed in a process of continuous interplay between commitment, in-depth exploration, and reconsideration (cf., Table 1).

Table 1.

Overview of the identity and self constructs examined in this review

Construct	Definition	Measurement
Identity processes (Crocetti et al., 2008)		Utrecht-Management of Identity Commitments Scale (U-MICS; Crocetti et al., 2008,Dimitrova et al., 2016), 13 items that can be repeated for multiple identity domains
Commitment	Choices individuals make in various developmental domains and the self–confidence they derive from these choices	5 items (e.g., “My education allows me to face the future with optimism”)
In-depth exploration	Extent to which individuals think actively about the commitments they have made, reflecting on their choices, and getting others’ perspectives about them	5 items (e.g., “I often reflect on my education”)
Reconsideration of commitment	Comparison of present commitments with possible alternative commitments because the former are no longer satisfactory	3 items (e.g., “I often think it would be better to try to find a different education”)
Identity socio-cognitive strategies (Berzonsky, 1989)		Identity Style Inventory (ISI; Berzonsky, 1992; Berzonsky et al., 2013; Crocetti et al., 2009), 30 items
Information-oriented	Self-reflective strategy, implying actively seeking out and evaluating self-relevant information	11 items (e.g., “When making important decisions I like to have as much information as possible”)
Normative	Tendency to adopting prescriptions and values from significant others and conforming to these others’ expectations	9 items (e.g., “I prefer to deal with situations where I can rely on social norms and standards”)
Diffuse-avoidant	Tendency to procrastinate and delay dealing with identity issues for as long as possible	10 items (e.g., “I’m not really thinking about my future now; it’s still a long way off”)
Self-concept clarity (Campbell et al., 1996)	Extent to which beliefs about the self are clearly and confidently defined, internally consistent, and stable over time	Self-Concept Clarity scale (SCC; Campbell et al., 1996), 12 items (e.g., “In general, I have a clear sense of who I am and what I am”)
Social identification (Postmes et al., 2013)	The subjective aspects of group membership, including the sense of identity and self-definition provided by feeling subjectively attached to a group	Social Identification scale (Thomas et al., 2017), 6 items that can be repeated for multiple groups (e.g., “I identify with the group of my classmates/friends”)

Commitment refers to enduring choices individuals have made about various developmental domains and to the self-confidence they derive from these choices (akin to the concept of identification with commitment proposed by Bosma, 1985). In-depth exploration indicates the extent to which individuals think actively about the commitments they have made, reflecting on their choices, searching for additional information, talking with others about their commitments (referring to the concept of present exploration proposed by Meeus, 1996). Reconsideration of commitment refers to comparing current commitments with possible alternatives because the current ones are no longer satisfactory. This new process was added to express a way of searching for new commitments (as it was the concept of exploration proposed by Marcia) that stems from current experience and is fuelled by dissatisfaction with existing choices.

The model assumes that interplay between these processes underlines a dual-cycle (Meeus, 2011, 2018), exemplified in Figure 1. In the identity maintenance cycle, individuals can explore their commitments in-depth and verify whether they provide a good fit with their goals, overall talents, and potential. When this does not happen, and individuals start questioning their identity, they move to the identity formation cycle, where they search for new alternatives because their current commitments are not satisfying or do not provide a good fit. Overall, these two iterative cycles provide a parsimonious and heuristic framework for understanding how individuals may, on the one hand, consolidate their identity and, on the other, go through identity crises and transitions.

Figure 1.

The three-factor model: The dual cycle.

Hence, the three-factor model provides a parsimonious approach to understanding the dynamic through which identity develops over time (Crocetti, 2017, 2018). Extensive longitudinal studies conducted with this model have highlighted systematic evidence of maturing identity in adolescence. This means that it is possible to document increases in commitment and in-depth exploration throughout adolescence and decreases in reconsideration (for reviews, see, Meeus, 2011, 2016). Thus, while identity certainty (as indicated by high levels of commitment and in-depth exploration) increases, identity uncertainty (indicated by high reconsideration of commitment) tends to diminish. A similar pattern is documented when the focus is on the identity cycles; adolescents tend to transit from the identity formation cycle to the identity maintenance cycle (for a review, see, Meeus, 2018). This is a general pattern in which it is possible to find considerable heterogeneity among individuals’ trajectories and non-linear processes. It is not uncommon to shift from the identity maintenance cycle to searching for a new one (Meeus, 2016).

So, on the one hand, the three-factor model is conceived as a framework to capture changes in identity, in line with the key tenet of Erikson’s theory. On the other hand, it contends that identity development does not occur in a vacuum but is firmly rooted in the social context. This latter applies to both identity cycles.

First, it is assumed that adolescents develop their identity by re-questioning their preliminary commitments, rooted in childhood identifications. Thus, the commitment formation cycle is based on comparing current choices with available alternatives, offered by the specific contexts in which adolescents grow up. For instance, young adolescents might be interested in arts because they are inspired by their parents’ passion and then become more interested in science when participating in a school programme aimed at increasing early adolescents’ interest in STEM disciplines. The school major they will choose would be the result of the dynamic interplay between their former commitment and the alternatives available in their school context, as happens in the identity formation cycle.

Second, the social context’s role is also prominent in the identity maintenance cycle. Here, through in-depth exploration of their commitments, adolescents can verify them. This active evaluation includes both intra-personal and interpersonal processes. For instance, young adolescents can reflect on the extent to which their current education matches their interests and future aspirations of becoming a journalist. To understand this, they can gather information from significant others (e.g., asking what their parents and their friends think about this). Thus, validating existing commitments also implies a process of social verification and social comparison with others (Crocetti et al., 2018).

Identity processes and socio-cognitive identity strategies

Given this iterative and social nature of identity dynamics proposed by the three-factor model, it is of utmost importance to provide a fine-grained understanding of how individuals manage their commitments and change their identity over time in interactions with their social experiences. In this respect, research connecting the three-factor model (Crocetti et al., 2008) with the social-cognitive perspective on identity construction (Berzonsky, 1989, 2004, 2011) has shown systematic relations with the strategies that individuals adopt in processing, structuring, utilising, and revising self-relevant information.

More specifically, individuals relying on an information-oriented strategy are self-reflective, actively seek out and explore self-relevant information, and are likely to define themselves using personal attributes, like “my values,” “my goals,” and “my standards” (Berzonsky, 1989). Individuals focusing on a normative strategy tend to enact commitments in a more automatic fashion, by adopting prescriptive behaviours and values from significant others and conforming to their expectations; in this vein, they mainly define themselves on the basis of collective self-attributes, such as “my family,” “my religion,” and “my ethnicity” (Berzonsky, 2004). By contrast, individuals with a diffuse-avoidant strategy procrastinate and delay dealing with identity issues for as long as possible, and have a propensity to emphasise contingent social aspects of their self-elements, such as reputation, popularity, and impression management (Berzonsky, 2011).

Consistent empirical research (Crocetti et al., 2009, 2013; Negru-Subtirica et al., 2017; Zimmermann et al., 2012) has highlighted that identity processes at the basis of the identity maintenance cycle (commitment and in-depth exploration) are positively associated with the information-oriented and normative strategies. This indicates that, by taking a more personal stance (such as the information-oriented strategy) or a more social one (as in the case of the normative one) when approaching identity issues, young people can actively consolidate their sense of identity. Notably, when young people show that they combine the information-oriented and normative strategies, positive implications of this plasticity are evident, as they express higher commitment and in-depth exploration than their peers who rely predominantly either on the information-oriented or the normative strategy (Crocetti, Berzonsky et al., 2012).

By contrast, the diffuse-avoidant strategy is intertwined with high reconsideration of commitment (Crocetti et al., 2009, 2013; Negru-Subtirica et al., 2017; Zimmermann et al., 2012). This points to a dark side of reconsideration of commitment (Beyers & Luyckx, 2016) that, especially when very prolonged, can hamper engagement with meaningful choices. In other words, individuals who continue to reconsider their commitments without identifying options that might suit their identity search might get trapped into a condition of identity uncertainty.

The self-regulatory function of identity

In the Eriksonian tradition, substantial attention has been paid to how identity contributes to individuals’ adaptation and well-being. Drawing on Erikson’s (1968) principle that identity fulfils a self-regulatory function (Serafini & Adams, 2002), it has been highlighted (Crocetti et al., 2013) that the identity processes of commitment and in-depth exploration, together with normative and information-oriented strategies, provide individuals with a framework enabling them to process and understand self-relevant information and achieve a sense of consistency and coherence among their chosen values and beliefs. Furthermore, making meaningful commitments provides individuals with a sense of direction, future orientation, and continuity between past, present, and future. Finally, the more individuals consolidate their identity, the more they perceive a sense of personal control, free will, or agency that enables active self-regulation in the process of setting and attaining goals and moving towards future plans. These considerations are of the utmost importance, as they provide a theoretical framework for understanding why identity commitment and, to a certain extent, in-depth exploration (as opposed to reconsideration of commitment), and information-oriented and normative strategies (as opposed to diffuse-avoidant strategy), are related to multiple positive correlates, including several components of adjustment and mental health (e.g., Berzonsky & Kinney, 2019; Hatano et al., 2020; Karaś et al., 2015; Mercer et al., 2017; for reviews, see, Berzonsky, 2011; Crocetti, 2017; Meeus, 2011).

Identity dynamics and self-concept clarity

After discussing how identity processes and identity strategies matter for individuals’ well-being, it is essential to go a step further and consider how they are related to the content of the self-concept. In this respect, it is worth considering self-concept clarity as an overall index of how well-organised the contents of the self-concept are (Lodi-Smith & DeMarree, 2017). More specifically, self-concept clarity refers to “the extent to which the contents of an individual’s self-concept (e.g., perceived personal attributes) are clearly and confidently defined, internally consistent, and temporally stable” (Campbell et al., 1996, p. 141). Thus, while identity processes and identity strategies tackle how the self-concept develops and changes, “self-concept clarity might indicate how well the process of developing an own identity is going” (Crocetti & Van Dijk, 2017, p. 1). In other words, “personal identity is framed as an active agent, the ‘I’, that sorts through and organises self-relevant information. Self-concept clarity is framed as the object, the ‘me’, that represents the self-conception being constructed” (Schwartz et al., 2017, p. 145).

In her original work, Campbell (1990) introduced self-concept clarity as a key concept that could be used to explain differences between individuals high and low in self-esteem. She demonstrated that individuals with low self-esteem were more susceptible to situational influences, such as false feedback or social pressures, because they had lower clarity or certainty in their self-conceptions. Since her seminal work, the concept of self-concept clarity has been applied in several domains, and it has been used to account for individual differences in mental health and well-being (for extensive reviews, see, Lodi-Smith & DeMarree, 2017). For instance, it has recently been found that self-concept clarity even plays an important protective role even in the stress process related to the adaptation to the COVID-19 outbreak and lockdown conditions (Alessandri et al., 2021). The concept has also been discussed in relation to important theorisations, such as the identity-uncertainty theory (Hogg, 2007, 2012). In this regard, self-concept clarity may serve the function of reducing self-uncertainty.

Moreover, a large corpus of evidence has highlighted that self-concept clarity is positively associated with endorsement of identity commitments, whereas it is negatively related to identity crises driven by reconsidering and discarding current commitments (Crocetti et al., 2008, 2010; Morsünbül et al., 2014; Schwartz et al., 2011, 2012). Similarly, self-concept clarity is negatively related to diffuse-avoidant identity strategy (Szabo & Ward, 2015). Thus, individuals’ striving for a meaningful identity helps to enhance self-concept clarity. But this is not a merely unidirectional process: when individuals have low self-concept clarity, they can engage in the identity formation cycle (Schwartz et al., 2011, 2012), and question their current identity to search for more satisfying alternatives.

Summary

This first section reviewed how the three-factor identity model was developed. Rooted in the Eriksonian tradition, this model offers a parsimonious framework to express the iterative nature of identity, with its oscillations between identity formation and maintenance cycles. We reviewed how the three pivotal identity processes (i.e., commitment, in-depth exploration, and reconsideration of commitment) are meaningfully related to the socio-cognitive strategies (i.e., information-oriented, normative, and diffuse avoidant) that individuals can use to process self-relevant information. We also discussed why identity processes and strategies have important implications for individuals’ well-being. Finally, we examined the importance of self-concept clarity as an overall estimate of how clearly defined the contents of the self-concept are. In the next section of this review, we go a step further to examine how identity processes, socio-cognitive strategies, and self-concept clarity (for an overview of these constructs their respective measures, see, Table 1) are embedded in the social contexts in which adolescents come of age.

Identity development in social contexts

In this second section, we review empirical studies that highlight how social contexts can affect the development of adolescents’ identity and, at the same time, how adolescents are active agents and can affect their own contexts to the extent that they achieve a clear view of themselves. To accomplish this, longitudinal methods are the most appropriate to capture the dynamics by which social contexts affect the development of adolescents, and adolescents, in turn, introduce changes in their social contexts. Thus, transactional and reciprocal processes are examined across a variety of contexts that range from the first socialisation context, the family (Crocetti et al., 2017; Crocetti, Rubini et al., 2016), to other proximal groups that are highly significant for the experience of adolescents (i.e., friendships; Van Doeselaar et al., 2016). Finally, we consider the broader societal context and examine how the way in which individuals define themselves is related to their willingness to help other people (Crocetti, Moscatelli et al., 2016) and to becoming civically engaged (Crocetti et al., 2014). In the studies presented below, we focus on cross-lagged models (and their cross-lagged results schematised in Figures 2–11), which reveal the predominant direction of effects by highlighting whether a certain variable influences a related variable or vice-versa.

Figure 2.

Model linking parents’ and adolescents’ self-concept clarity.

Note. SCC = Self-Concept Clarity; Ado = Adolescent. ** p < .01. N = 497. The time lag between waves was one year. Cross-lagged paths were tested controlling for one-year and two-year stability paths and within-time correlations. For the sake of clarity, only significant standardised cross-lagged paths are reported. Reprinted with permission and adapted from Crocetti et al. (2016) in Journal of Personality, 84(5), 580–593. © John Wiley & Sons.

Figure 3.

Model linking family relationships and identity commitment.

Note. Negative = Negative interaction. *p < .05, **p < .01, ***p < .001. N = 497. The time lag between waves was one year. Cross-lagged paths were tested controlling for one-year and two-year stability paths and within-time correlations. For the sake of clarity, only significant standardised cross-lagged paths are reported. Since the model with time-invariant coefficients was retained as the final one, we present only two time points (T and T + 1), and all coefficients displayed represent the averaged standardised coefficients path coefficients over the five-time intervals. Reprinted with permission and adapted from Crocetti et al. (2017) in Child Development, 88, 210–228. © John Wiley & Sons.

Figure 4.

Model linking family relationships and identity in-depth exploration.

Note. Negative = Negative interaction. *p < .05, **p < .01, ***p < .001. N = 497. The time lag between waves was one year. Cross-lagged paths were tested controlling for one-year and two-year stability paths and within-time correlations. For the sake of clarity, only significant standardised cross-lagged paths are reported. Since the model with time-invariant coefficients was retained as the final one, we present only two time points (T and T + 1), and all coefficients displayed represent the averaged standardised coefficients path coefficients over the five-time intervals. Reprinted with permission and adapted from Crocetti et al. (2017) in Child Development, 88, 210–228. © John Wiley & Sons.

Figure 5.

Model linking family relationships and identity reconsideration of commitment.

Note. Negative = Negative interaction. *p < .05, **p < .01, ***p < .001. N = 497. The time lag between waves was one year. Cross-lagged paths were tested controlling for one-year and two-year stability paths and within-time correlations. For the sake of clarity, only significant standardised cross-lagged paths are reported. Since the model with time-invariant coefficients was retained as the final one, we present only two time points (T and T + 1), and all coefficients displayed represent the averaged standardised coefficients path coefficients over the five-time intervals. Reprinted with permission and adapted from Crocetti et al. (2017) in Child Development, 88, 210–228. © John Wiley & Sons.

Figure 6.

Model linking identity processes, social responsibility, and volunteer and political engagement.

Note. * p < .05. ** p < .01. *** p < .001. N = 392. This study was cross-sectional. For the sake of clarity, only significant standardised cross-lagged paths are reported. Reprinted with permission and adapted from Crocetti, Jahromi et al. (2012) in Journal of Adolescence, 35, 521–532. © Elsevier

Figure 7.

Model linking identity socio-cognitive strategies and civic engagement.

Note. *p < .05, **p < .01, ***p < .001. N = 1,308. The time lag between waves was one year. Cross-lagged paths were tested controlling for stability paths and within-time correlations. For the sake of clarity, only significant standardised cross-lagged paths are reported. Reprinted with permission and adapted from Crocetti et al. (2014) in International Journal of Developmental Science, 8, 115–124. © IOS Press.

Figure 8.

Model linking self-concept clarity and prosociality.

Note. SCC = Self-concept clarity. *p < .05, **p < .01. N = 244. The time lag between waves was one year. Cross-lagged paths were tested controlling for one-year and two-year stability paths and within-time correlations. For the sake of clarity, only significant standardised cross-lagged paths are reported. Reprinted with permission and adapted from Crocetti et al. (2016) in European Journal of Personality, 30(6), 594–607. © John Wiley & Sons.

Figure 9.

Model linking personal and social identities.

Note. *p < .05, **p < .01, ***p < .001. N = 304. The time lag between waves was three months. Cross-lagged paths were tested controlling for all stability paths and within-time correlations. For the sake of clarity, only significant standardised cross-lagged paths are reported. Since the model with time-invariant coefficients was retained as the final one, we present only two time points (T1 and T2&T3). ross-lagged paths displayed represent the averaged standardised coefficients. Reprinted with permission and adapted from Albarello, Crocetti et al. (2018) in Journal of Youth and Adolescence, 47, 689–702. © SpringerNature.

Figure 10.

Model linking identification with proximal groups, identification with humanity, and social well-being.

Note. *p < .05, **p < .01, ***p < .001. N = 304. The time lag between waves was three months. Cross-lagged paths were tested controlling for all stability paths and within-time correlations. For the sake of clarity, only significant standardised cross-lagged paths are reported. Bold arrows refer to paths underlying indirect effects (Identification with classmates T1 → Identification with humanity T2 → Social well-being T3; Identification with classmates T1 → Identification with friends T2 → Social well-being T3). Reprinted with permission and adapted from Albarello et al. (2021) in Journal of Youth and Adolescence. © SpringerNature.

Figure 11.

Model linking multiple categorisation, social dominance orientation, prejudice, and human identification.

Note. *p < .05, **p < .01, ***p < .001. N = 304. The time lag between waves was three months. Cross-lagged paths were tested controlling for all stability paths and within-time correlations. For the sake of clarity, only significant standardised cross-lagged paths are reported. Bold arrows refer to paths underlying indirect effects (Multiple categorisation T1→ Prejudice T2→ Human identification T3; Social dominance orientation T1→ Prejudice T2→ Human identification T3). Reprinted with permission and adapted from Albarello et al. (2020) in Child Development, 91, 1183–1202. © John Wiley & Sons.

The family context

The family context represents the first micro-system in which individual development occurs (Bronfenbrenner & Morris, 2006) and greatly impacts successive experiences with other proximal (e.g., peer groups, school contexts) and more distal systems. How can parents positively influence and support their children’s self and identity formation? In our research programme (Crocetti et al., 2017; Crocetti, Rubini et al., 2016), we addressed this question by considering two core mechanisms. First, we examined whether parents can affect adolescents’ development by communicating who they are and acting as modelling agents (Crocetti, Rubini et al., 2016). Second, we considered the impact of the quality of family relationships on adolescents’ identity processes (Crocetti et al., 2017).

Parents as modelling agents: intergenerational transmission processes

Parents can deeply influence their adolescent children by acting as modelling agents (Wiese & Freund, 2011). These considerations are rooted in social learning theory (Bandura, 1977) that emphasises the centrality of the concept of modelling to understanding the socialisation process. Thus, parents who hold well-defined self-belief might represent a stronger point of reference for adolescents in search of their identity as compared to parents with uncertain self-beliefs.

We documented this effect in a longitudinal study in which we examined intergenerational transmission of self-concept clarity in families with adolescents (Crocetti, Rubini et al., 2016). Participants were 497 Dutch families, including fathers, mothers, and their adolescent children. The fathers, mothers, and adolescents reported their own self-concept clarity for six annual assessments when adolescents were 13, 14, 15, 16, 17, and 18 years old. The results provided consistent evidence of uni-directional transmission processes (Figure 2). The self-concept clarity of fathers and mothers had positive effects on that of adolescent children over time, while adolescents’ self-concept clarity did not influence that of their parents. Furthermore, the pattern of influence in same-sex dyads (i.e., father-son, mother-daughter) was similar to the pattern in opposite-sex dyads (i.e., father-daughter, mother-son). Overall, this evidence underscores that when adolescents can count on parents high in self-certainty, they are more likely to increase their self-concept clarity throughout adolescence.

It is worth noting that the evidence of intergenerational transmission of self-concept clarity is consistent with intergenerational transmission processes occurring in other domains of adolescent development (cf., Meeus, 2016, for a review). For instance, transmissions of cultural orientations and attitudes (Meeusen, 2014; Ter Bogt et al., 2005; Vollebergh et al., 2001) and conflict resolution styles (Van Doorn et al., 2007) are also uni-directional processes: While parents influence their children, children do not influence their parents. Thus, longitudinal studies document parental dominance in intergenerational transmission processes.

This dominance can be further understood by considering the higher stability of parents’ self-views. In this regard, our study highlighted that parents reported having greater self-concept clarity and also displayed higher rank-order stability compared to their adolescent children (this was especially true for males). Thus, as they have a relatively more stable self than adolescents, parents are more likely to influence them than the other way around. In this vein, the impact of parents’ self-concept clarity on that of adolescents is consistent with the theoretical principle that systems with a higher degree of stability are more likely to affect those with a lower degree (Asendorpf & Van Aken, 2003).

To conclude, longitudinal studies clearly show intergenerational transmission processes. Self-concept clarity, as well as attitudes, values, orientations, and interpersonal styles (e.g., for meta-analyses, see, Cemalcilar et al., 2018; Degner & Dalege, 2013) are transmitted from parents to offspring. Thus, by communicating who they are, parents work as modelling agents for their children and consistently impact their development.

The impact of family relationship quality

Another core mechanism through which parents can promote the identity development of their children is by forming high-quality relationships that provide adolescents with a “secure basis” to explore their own identity. This idea is rooted in the attachment theory (Bowlby, 1988), according to which a secure bond with parents is necessary for exploring identity with confidence and making autonomous choices and decisions. Cross-sectional studies have confirmed that identity commitment and in-depth exploration are positively related to warm and supportive family relationships, whereas reconsideration of commitment is related to poor quality relationships (Crocetti et al., 2008, 2010; Morsünbül et al., 2014).

Studies using longitudinal design can substantially advance understanding of the associations between identity processes and quality of family relationships by discovering the predominant direction of effects. On the one hand, family literature has widely theorised that family relationships influence adolescents’ identity formation (Årseth et al., 2009). On the other hand, building upon Erikson’s (1950, 1968) psychosocial theory, it can be argued that changes in adolescents’ identity formation might in turn impact the quality of family relationships. Quoting Erikson (1968, p. 167), “true engagement with others is the result and the test of firm self-definition”. In line with the assumption of Erikson’s psychosocial theory that optimal resolution of the identity formation task is a developmental precursor of intimate and caring (generative) relationships (Lawford et al., 2020), it is reasonable to expect that the more adolescents develop a clear sense of who they are, the more they can establish warm and balanced relationships with significant others.

We tested these two competing hypotheses in the longitudinal study mentioned before (Crocetti, Rubini et al., 2016) which involved 497 Dutch families. In this case, differently from the prior one, the sample included not only the target adolescents, their fathers and mothers, but also their siblings, and all participants took part in the study for five years (Crocetti et al., 2017). Thus, we examined the associations between adolescents’ identity processes (commitment, in-depth exploration, and reconsideration of commitment) and quality of relationship assessed by considering multiple indicators (support, negative interaction, and power; Furman & Buhrmester, 1985) reported directly by adolescents’ fathers, mothers, and siblings. A novelty of this study was the focus on siblings too. Although sibling relationships have received less attention, they are an essential component of family relationships (Buist et al., 2013), and siblings can influence adolescents’ identity formation by engaging in meaningful conversations regarding identity issues (e.g., Cardwell & Soliz, 2020).

The main results indicated that the quality of family relationships had an impact on adolescents’ identity formation (i.e., mothers’ levels of support negatively predicted changes in adolescents’ reconsideration of commitment throughout adolescence), but most of the effects were in the other direction (i.e., from adolescents’ identity to later quality of family relationships; Figures 3–5). More specifically, adolescents’ commitment had a positive effect on the relationships with the mother (that over time became more supportive and less conflictual) and with the sibling (that over time became more egalitarian; Figure 3). Over time adolescents’ in-depth exploration improved the relationship with all family members, leading to a more supportive and mutual relationship with the father, a more supportive and less conflictual relationship with the mother, and a more supportive relationship with the sibling (Figure 4). Adolescents’ reconsideration of commitment reduced paternal support (Figure 5). Notably, multi-group analyses indicated that results were not moderated by adolescent gender, sibling gender similarity, or sibling age.

Overall, this study indicates that adolescents’ identity can function as a developmental precursor of the quality of relationships with parents and siblings. This evidence is consistent with other longitudinal studies documenting bidirectional links between family relationships and identity in adolescents (Schwartz et al., 2009) and in emerging adults (Beyers & Goossens, 2008; Luyckx et al., 2007) and showing that identity reduced loneliness (Kaniušonytė et al., 2019).

In general terms, this evidence highlights that parents do not unilaterally influence the development of youth identity, but it is rather a process of reciprocal influence where adolescents play an active, transactional role (Sameroff, 2009). In this vein, adolescents’ identity development significantly impacts on interaction with family members. When parents and siblings perceive ongoing changes in adolescents’ identity leading to increased maturity and certainty, they relate better. By contrast, when adolescents reconsider their identity and show problematic behaviours (e.g., reporting high aggression or being very anxious), they enter into a negative loop that gives rise to a progressive erosion of the quality of family relationships (Crocetti, Moscatelli et al., 2016; Hale et al., 2016).

Friendships

In adolescence, young people increasingly spend time with their friends (De Goede et al., 2009), who become a primary source of intimacy and support (cf., Brown & Larson, 2009 for a review). According to the classical concept proposed by Sherif and Sherif (1964), when adolescents interact with their friends, they are in a “social laboratory”. In this context, they can experiment with different behaviours and self-presentation strategies, receive informative feedback from others, and benefit from social comparison processes.

Friendships, like family relationships, can influence the development of adolescents’ identity in different ways. In particular, two distinct mechanisms can operate. First of all, the way in which friends develop their own identity might affect the identity formation of their fellow mates. Second, friendships can form a safe interpersonal context in which adolescents test their identity choices through social comparison (McLean & Jennings, 2012). In this vein, the quality of relationships with friends may significantly impact adolescents’ identity development.

These two mechanisms were tested in a five-wave longitudinal study with 464 Dutch adolescents and their self-nominated best friends (Van Doeselaar et al., 2016). On the one hand, the study provided limited evidence of transmission processes: in fact, correlations between the educational identity of the target adolescents and of their best friends were small, and over time higher levels of adolescents’ commitment predicted a slight decrease in best friends’ commitment (but only in stable, as compared to unstable, friendships). However, on the other hand, the study highlighted reciprocal associations between the educational identity of the target adolescents and the quality of the relationship with the best friend, assessed with a measure of balanced relatedness (i.e., the extent to which adolescents increasingly accept the opinions and ideas of their friend, even when they differ from their own; Shulman & Knafo, 1997). More specifically, target adolescents’ commitment and in-depth exploration over time predicted a positive increase in balanced relatedness, whereas balanced relatedness predicted a relative decrease in reconsideration of commitment.

Overall, this evidence indicates the protective role of high-quality friendships. When adolescents feel accepted and supported by their friends, over time they show lower problematic reconsideration of commitment in the educational domain. Furthermore, educational identity processes of commitment and in-depth exploration were found to foster balanced relatedness in the relationship with the best friend. Thus, as was documented for family relationships (Crocetti et al., 2017), with friendships too, reciprocal effects are involved and reveal how adolescents developing their own identity can exert agency in interpersonal contexts.

Experiencing commitment in the society at large

So far, we have discussed how self and identity processes are embedded in the main adolescents’ proximal contexts of socialisation (i.e., family and friendships). We now consider how identity processes are intertwined with participation in broader civil society. Theoretically, the clearer a view adolescents have of who they are, the more they can care for others and express their agency in society. This hypothesis was tested in a cross-sectional study with 392 Italian adolescents (Crocetti, Jahromi et al., 2012). The findings confirmed that identity commitment and in-depth exploration were positively related to both volunteer and political engagement, and this effect was mediated by increased social responsibility (Figure 6). Although, on the one hand, this study highlighted meaningful associations between identity and different forms of participation in civil society, on the other hand, its cross-sectional design did not allow us to test the direction of these effects (Hardy et al., 2011). To fill this gap, we conducted two longitudinal studies, in which we examined long-term associations between identity socio-cognitive strategies and civic engagement (Crocetti et al., 2014) and between self-concept clarity and prosociality (Crocetti, Moscatelli et al., 2016), respectively.

In the first study, we tackled associations between identity socio-cognitive strategies and civic engagement (i.e., participation in school self-government activities, volunteering activities, youth political organisations, and non-political organisations) in a two-wave longitudinal study with 1,308 Lithuanian adolescents (Crocetti et al., 2014). The main results (Figure 7) revealed that identity strategies were a stronger predictor of civic engagement (i.e., each identity strategy affected civic engagement at a later time) rather than the opposite (i.e., civic engagement determined changes at later times only in one out of three identity strategies). We found that adolescents with a preference for an information-oriented strategy might make use of more psychosocial resources to approach civic engagement. However, defining themselves mainly on the basis of individual characteristics may prevent them from establishing a sense of connection and belongingness, which is an essential component of civic participation (Yates & Youniss, 1996). By contrast, adolescents using a normative strategy might be more driven by external motivations when approaching civic engagement (e.g., desire to make a good impression on others) but be more likely to appreciate their experience based on their tendency to form and maintain strong social bonds. Moreover, adolescents who delay and postpone identity issues as long as possible are less prone to care and become committed in the civic domain. In addition to these effects of identity on civic engagement, we found that civic engagement also predicted positive changes in identity one year later (i.e., high civic engagement lessened reliance on the diffuse-avoidant strategy).

In a second six-wave longitudinal study with 244 Dutch adolescents transitioning to emerging adulthood (Crocetti, Moscatelli et al., 2016), we found that participants’ self-concept clarity was related to their prosociality (i.e., the tendency to perform voluntary activities regarded as beneficial to others, including helping, sharing, comforting, guiding, rescuing, and defending others; Padilla-Walker & Carlo, 2014). The results indicated that, on the one hand, the clearer view adolescents had of themselves, the more willing they were to help other people; on the other hand, the more young people behaved prosocially, the more their self-concept clarity increased (Figure 8). Interestingly, the effect of prosociality on self-concept clarity was stronger than the reciprocal effect of self-concept clarity on prosociality. This suggests that the likelihood of being involved in prosocial activities can enhance self-understanding, leading to more stable and confident self-definition.

Overall, these longitudinal studies provide convergent evidence on how self and identity processes are related to participation in civil society. On the one hand, achieving greater self-certainty and a relatively stable identity lays the basis for establishing caring relationships with others, especially with people that might be in a condition of need (Hatano et al., 2022). As part of a virtuous circle, the multiple social experiences that adolescents have when they help others and are actively engaged in their communities provide them with a sense of industry and self-efficacy derived from being actively involved in activities of societal relevance; opportunities to strengthen social bonds with people from different in-groups (e.g., the volunteering group in which adolescents perform their prosocial activities) and out-groups (e.g., a marginalised group towards which volunteer activities are targeted) and to reflect on their core values (Yates & Youniss, 1996). These experiences increase self-understanding and a more mature identity (Erentaitė et al., 2019).

Summary

Taken together, the longitudinal studies reviewed in this second section illustrate how identity processes are deeply embedded in the interactions that adolescents have within multiple social contexts. Notably, two main conclusions can be drawn from this research programme (and the broader adolescent psychosocial literature). Significant others, especially parents, strongly contribute to shaping the identity development of their adolescent children by “showing who they are”. Indeed, intergenerational transmission of self-concept clarity, as well as of values and personality characteristics, are characterised by uni-directional effects (from parents to adolescents) that pervasively help to influence adolescents. In contrast, adolescents’ identity and interactions with significant others in both proximal (family and peers) and distal (civil society) contexts display bi-directional effects, showing that not only meaningful relationships and diversified experiences among significant others and adolescents influence how the latter define their identity, but also that adolescents are active agents in these contexts and have a significant impact on them. Building upon this evidence, in the third section of the review, we will explore associations with social identity processes, with a specific focus on social identification (Crocetti et al., 2018).

The interplay between personal and social identity

This third section reviews studies that considered both personal and social identities in adolescence. First, we outline how personal and social identity processes can be intertwined, even though they have mainly been studied separately. Second, we outline how identifications with proximal and abstract groups can lead to social well-being (Keyes, 1998), which is conceived as a “public phenomenon” (Rollero & De Piccoli, 2010) referring to the extent to which individuals are integrated into society and can actively contribute to its development (Keyes, 2006). Finally, we consider how different cognitions influence symbolic processes such as identification with the most inclusive group, that is, the human group (Albarello, Crisp et al., 2018; Albarello & Rubini, 2012; Turner et al., 1987).

The interplay between personal and social identification processes

How can personal and social identification processes be intertwined? In recent contributions, we tackled this issue by combining a developmental and a social psychological perspective (Albarello, Crocetti et al., 2018; Albarello et al., 2020, 2021). As a first step, we examined the longitudinal interplay between personal and social identity processes in adolescence (Albarello, Crocetti et al., 2018). In a three-wave longitudinal study with 304 Italian adolescents attending the 11^th and 12^th grades, we examined within-time and across time associations between personal identity processes (i.e., identity commitment, in-depth exploration, and reconsideration of commitment) in the educational and interpersonal domains and social identification with two salient social groups, experienced daily (i.e., classmates and friends met outside school).

The results indicated that personal and social identity processes were associated, both concurrently and longitudinally (Figure 9). Identification with classmates was positively related to educational commitment at each time point. In addition, identification with the group of friends was positively associated with interpersonal commitment and in-depth exploration and negatively linked to reconsideration of commitment. Most importantly, the results highlighted significant paths from identification with classmates to interpersonal commitment and interpersonal in-depth exploration and from identification with friends to interpersonal commitment and interpersonal reconsideration of commitment. This evidence reveals that social identifications primarily influenced personal identity formation and maintenance in the interpersonal identity domain. This implies that symbolic processes involving adolescents as members of social groups can affect their personal identity development. In this vein, it has to be underlined that such processes should not be examined in isolation if we aim to acquire a deep understanding of how adolescents deal with identity processes in their increasingly complex social context.

Social identifications with proximal and distal groups

Going a step further, in view of the increased variety of adolescents’ experiences with the social environment they are embedded in, we examined the developmental trajectory of social identification (Albarello et al., 2021). Differently from the studies reported above (e.g., Crocetti et al., 2017; Van Doeselaar et al., 2016) which considered as social variables, social factors that are external to participants (e.g., beliefs of parents and peers), in this further set of studies, we focused on participant’s identification with different social groups. In this way, we tapped into personal perceptions and meaning of social memberships, an aspect that is known to have crucial implications for intergroup behaviours (cf., Ellemers et al., 2002; Tajfel & Turner, 1979; Turner et al., 1987). More specifically, we analysed longitudinal association between identifications with proximal (i.e., classmates and friends) and more abstract groups, such as the human group (i.e., the most inclusive group that encompasses the whole variety of social groups and can be considered as the most abstract). We also focused on the impact of these identifications on social well-being, as an indicator of youth adaptation in their societies and communities, at a later time. Specifically, well-being as the “appraisal of one’s circumstance and functioning in society” (Keyes, 1998, p. 122) refers to the extent to which individuals “feel socially integrated and socially contributive” (Keyes, 1998, p. 133), that is, feel part of the groups and societies in which they live and perceive that they can contribute something valuable to the common good. It comprises multiple components that rely on the evaluation of the self with respect to the social context (social integration, social contribution), the evaluation of other people (social acceptance), and the evaluation of society (social coherence, social actualisation) (see, Cicognani et al., 2008; Rollero & De Piccoli, 2010). Reasoning that social well-being can be regarded as a crucial achievement of adolescence and an indicator of positive youth development (Cicognani et al., 2008), we analysed how social identification with proximal and more abstract groups can affect social well-being in a three-way longitudinal study with Italian adolescents (Albarello et al., 2021).

Our results highlighted that identification with proximal social groups (especially classmates) was positively associated with identification with the human group, and identifying with both proximal and abstract groups was related to adolescents’ social well-being over time (Figure 10). Furthermore, identification with the human group and identification with the group of friends mediated the longitudinal effects of identification with classmates on social well-being. Thus, such findings highlighted the importance of experiences with proximal social groups (especially classmates) in shaping more complex and abstract identification processes (i.e., human identification), which in turn can affect the way in which young people think and behave as society members at a later time. These results resonate with the idea that belonging and being identified with groups is a core need of human beings (Fiske, 2010). Interestingly, these results add that fulfilling this need can also produce a sense of social well-being, with all it entails in terms of functional adaptation of youth and active involvement in their community and society.

Individual cognitions and symbolic processes

Besides producing beneficial outcomes for in-group members such as self-esteem enhancement (Smith & Silva, 2011; Tajfel & Turner, 1979), social identity processes also have important implications for intergroup relationships. Along this line, we considered the impact of beneficial and detrimental factors on prejudice against the stigmatised out-group of migrants. With regard to beneficial factors, that is factors that can hinder or reduce prejudice, we focused on using multiple categorisations to define migrants. In this study, unlike what has generally been done with the multiple categorisation paradigm (Albarello & Rubini, 2012; Crisp et al., 2001; Crisp & Hewstone, 2007; for a recent review see, Prati et al., 2021), we assessed the extent to which individuals rated four combinations of multiple categorisation as definitions that can be applicable to the outgroup of migrants. A multiple categorisation mean score was then obtained by averaging the means of the four items. High scores indicate that multiple categorisation definitions are a suitable way of describing migrants. In contrast, low scores indicate that multiple categorisation definitions are less suitable to define them. In this vein, multiple categorisation of migrants can reflect individual cognition, which varies among individuals. The extent to which individuals consider multiple categorisation as applicable to the definition of migrants can affect prejudice towards them.

As regards detrimental factors (i.e., factors that are usually associated with severe prejudice), we considered social dominance orientation (Sidanius & Pratto, 2001). Social dominance orientation is an individual trait expressing “a generalized orientation towards and desire for unequal and dominant/subordinate relations among salient social groups, regardless of whether this implies ingroup domination or subordination” (Pratto et al., 2006, p. 282). It is often associated with great prejudice and legitimisation of social inequalities. In fact, people high in social dominance orientation are characterised by insensitivity to moral violations and the welfare of others. In contrast, people low in social dominance orientation are motivated by egalitarianism and altruistic social concern and prioritise fairness and harm-avoidance (Duckitt, 2001).

In a novel way, besides unravelling how multiple categorisations and social dominance orientation interact in explaining prejudice over time, this research also aimed to puzzle out the associations between prejudice and social inclusivity over time. The latter was measured as the extent to which individuals identify with the common human group (Albarello et al., 2020). Notably, prejudice and social inclusivity cannot be conceived merely as two faces of the same coin, but are most likely driven by different processes (e.g., tolerance appears at a different developmental stage; Van Zalk & Kerr, 2014).

Findings of a three-wave longitudinal study with Italian adolescents underlined both the positive role of multiple categorisations in reducing prejudice and the negative role of social dominance orientation in leading to high prejudice, at a later time (Figure 11). The extent to which adolescents used multiple categorisations was negatively related to prejudice against migrants later. By contrast, social dominance orientation was positively associated with later prejudice. Interestingly, the results also showed that the association between social dominance orientation and prejudice was bidirectional; that is, prejudice was positively associated with social dominance orientation at a later time, highlighting a “dark chain” through which prejudice can work as a legitimising myth of social inequalities. Moreover, and most importantly, we also found that the association between social dominance orientation and human identification (McFarland et al., 2019) is mediated by the view (i.e., prejudice) that individuals have about some very salient outgroups that can be included in the human ingroup, such as migrants. That is, prejudice against migrants played an active (i.e., mediating) role in explaining adolescents’ identification with humanity at a later time.

This finding highlights that prejudice should not only be considered an outcome of individual-level processes such as categorisation, but can also affect other phenomena, such as social inclusivity, as a symbolic outcome of an individual’s perception and experience of others. Such intriguing novel evidence underlines the close interplay between individual and symbolic processes that can explain young people’s inclusive identifications, with all it entails in terms of promoting harmonious relations in diverse societies (e.g., reduction of dehumanisation; Albarello & Rubini, 2012). It also suggests that it is important to consider developmental trajectories of more complex and abstract cognition about one’s memberships in order to thoroughly analyse what factors can foster this form of social inclusivity.

Summary

In this section, we highlighted that personal and social identification processes can be deeply intertwined. Social identification has been emphasised as a factor affecting the development of personal identity. For instance, this means that identity formation in adolescence must also be studied by considering the social bounds of young people with the groups to which they belong in order to achieve a deeper understanding of such developmental processes. A combined analysis of the interplay of various (proximal and more abstract) social identifications can help us to understand other developmental tasks such as fostering adolescents’ social well-being as a means of individuals’ adjustment to their communities. Other social level phenomena such as prejudice can also affect the way in which individuals define and feel interconnected with others through human identification. This suggests that identification, at the personal as well as at the social level, needs to be analysed by considering the (wider) social contexts in which individuals are embedded in, to be thoroughly understood. In other words, the cognitions that individuals develop about themselves as single persons, as group members, and as human beings can be deeply influenced by their views about self and others – for instance, in terms of stereotypes and prejudice held about outgroups – as well. Considering such intertwined social cognitive processes, rather than conceiving them as separate and independent, is thus crucial to augmenting our understanding of identity in its different facets.

Conclusions

The research programme reviewed in this article offers in-depth, convergent evidence showing that the process by which individuals define their own identity is intertwined with diversified and continuous experiences in multiple contexts. Thus, this review sheds light on the dynamic and transactional process by which adolescents’ identity is nourished by social interactions and, in a reciprocal loop, affects social experiences.

We reviewed a set of longitudinal studies conducted with Southern (i.e., Italian; Albarello, Crocetti et al., 2018; Albarello et al., 2020, 2021), Eastern (i.e., Lithuanian; Crocetti et al., 2014) and Western (i.e., Dutch; Crocetti et al., 2017; Crocetti, Moscatelli et al., 2016; Crocetti, Rubini et al., 2016; Van Doeselaar et al., 2016) European adolescents. Overall, they provided consistent evidence that individuals’ identities are defined in close interaction with proximal (family, friends) and distant (civic society) life contexts. At the same time, the more individuals achieve a more established sense of who they are, the more they can exert a powerful influence on their contexts as well.

Overall, this review underscores that adopting a cross-fertilisation approach, where a classic social-psychological phenomenon (i.e., self and identity) is examined from a developmental perspective, can significantly advance the understanding of identity development processes. Along this line, this contribution sheds light on the social nature of identity development by highlighting that, although identity processes reside and develop within the individual, they are nevertheless systematically influenced by the social contexts in which people live in terms of providing options for identity choices, memberships of social groups, contents of self and others’ definition.

It should be taken into account that whereas research on personal identity was initially pursued from an intra-individual perspective leaving in shadow the role of others and using longitudinal methods that make it possible to address how identity is formed and transformed over time, social identity research has focussed from the very beginning on the self-others interplay, using mainly experimental methods that made it possible to advance theoretical explanations in a controlled way. Although personal identity research has considered exploration as the fundamental process to acquire information on a given life domain and establish a related identity choice, social identity tradition recognises social categorisation as the initial process of identity formation. Once the self recognises that a certain categorisation provides a usable and significant self-definition (Turner et al., 1987) and others define the self accordingly, social identity is acquired to the extent that the self also identifies with that particular category/group (Brown, 2000). Postmes et al. (2005) provided a more finely grained explanation of social identity formation by referring to deductive and inductive processes rooted in the collective characteristics of group members.

In general terms, the notion that is common to personal and social identity is commitment. In the domain of personal identity, commitment is the process through which relatively stable choices are made in the various life domains. In the context of social identity, commitment has been addressed as an important component of group identification (Doosje et al., 1999). It is plausible to argue that commitment manifests itself at the cognitive level through the awareness of being a member of a certain group, at the emotional level through the affective commitment to one’s own ingroups, and at the behavioural level through the actions performed to maintain and advance one’s own groups.

Another communality between personal and social identity resides in the strategies enacted when one’s own personal or social identity is no longer satisfying. On the front of personal identity, through in-depth exploration, one can reconfirm commitment to a certain identity choice or decide to leave the current commitment after pondering reconsideration of commitment. As for social identity,Tajfel (1981) had already contended that one could enact mobility strategies if it is easy to move from one group to another or adopt collective social-change strategies if group boundaries are not very permeable. In this vein, what is important for both personal and social identity is that individuals can achieve relatively stable identity choices that can be submitted to change when altered conditions (whether personal, social or structural ones) jeopardise their identities.

In general terms, inspired by the advances of research on personal (Crocetti, 2017) and social identity (Prati et al., 2021) it is very important that adolescents and people in general can rely on relatively stable and yet flexible identities to be able to adapt to the multiple challenges of contemporary societies successfully and to be inclusive of diverse people and groups. Conversely, making identity choices in only a few domains and being strongly identified with only a few groups can lead to ethnocentrism, discrimination and any other kind of social prejudice (Crocetti et al., 2021).

Directions for future research

Drawing on the developmental social-psychological perspective of this review, future research might increase theoretical understanding of identity dynamics in several directions. First, future research should directly address the issue of social identity and group belongingness in longitudinal designs. This would make it possible to understand when and how youth start to be aware of their group memberships, how they maintain, enlarge or change their adherence to these important social psychological experiences, especially in the light of the renewed ethnic and cultural mosaic of contemporary societies that has dramatically changed due to migration processes (United Nations, 2017). In this vein, the increasing diversity of societies requires adolescents to manage their identities in a complex way, acknowledging that how they address the core question “who am I?” could be the result of a dynamic process based on multiple and diverse social interactions with individuals from ethnic and cultural groups different from their own in-group. For adolescents with a migrant background (e.g., refugees, first-generation, and second-generation immigrants, international students), the identity formation task can be particularly challenging, as they have to manage several (and sometimes even conflictual) alternatives proposed, on the one hand, by their culture and family of origin, and, on the other hand, by the host society (e.g., Karataş et al., 2020; Schwartz et al., 2013). As a result, they can be more uncertain about their identity choices, and can keep on considering and reconsidering their commitments in various life domains (Crocetti et al., 2011). Similarly, for adolescents without a migrant background growing up in societies with increasing diversity provides the opportunity to consider different cultural systems and explore diverse identity pathways. Thus, future research is needed to tackle how adolescents’ identity formation is related to the quality of intergroup contact experiences in multiple socialisation contexts.

Second, the current state-of-the-art indicates that identity is a multifaceted construct (Crocetti & Salmela-Aro, 2018) and that identity processes have important implications for adolescents’ well-being (Crocetti et al., 2018). However, longitudinal research in this domain is still fragmented and sparse. While there is evidence that the development of personal identity affects psychological well-being and mental health (e.g., Hatano et al., 2020), the implications of personal identity formation for other components of well-being, such as social integration and physical health, still need further investigation. Similarly, social and human identities have mainly been related to collective outcomes, such as intergroup attitudes and discrimination (e.g., Albarello & Rubini, 2012), while deserving more attention to shed light on how these identities affect multiple dimensions of adolescents’ well-being over time. Above all, the next frontier for identity research requires the development of integrative frameworks to uncover how the dynamic interplay of personal and social identity processes (instead of considering only one identity facet) affects multiple components of well-being in adolescence.

In this vein, it has been suggested that “personal and social identities form a fundamental symbolic tool that individuals use to adapt to the multiple domains of their lives. In a continuous flow of interaction with their social context, people can use and strengthen their personal and social identities to satisfy the main needs and motives underlying human behaviour” (Crocetti et al., 2018, p. 306). Notably, future research could explore the impact of multiple personal and social identity processes on adolescents’ well-being to test whether fulfilment of fundamental human needs and motives (such as need to belong and to be competent; cf., Fiske, 2010; Vignoles, 2011) are key factors (i.e., mediators) underlying these associations.

From a methodological perspective, in this review, we discussed studies that were mainly aimed at disentangling the predominant direction of effects, by uncovering whether a certain factor influences a related one or vice-versa. To do so, the results of cross-lagged models were discussed. However, increasing attention has been devoted to the importance of differentiating between-persons effects (as examined in traditional cross-lagged models) and within-person effects (that can be detected by applying a random intercept cross-lagged panel model; Hamaker et al., 2015). Studies (e.g., Crocetti et al., 2015; Mercer et al., 2017; Miklikowska, 2018; Negru-Subtirica et al., 2020) in which both models were applied and compared highlighted that, while some results might be replicated, others can be attributed either mainly to a difference in the relative standing of an individual within a group (i.e., between-person effects) or to deviations from individuals’ average scores (i.e., within-person effects). Thus, further research might further clarify which effects operate at which level and, in doing so, inform more precisely evidence-based interventions (Meeus, 2016; Orth et al., 2021).

To conclude, we do believe that this review has highlighted the value of studying identities from a developmental social-psychological perspective and, in doing so, has paved the way for future research aimed at further understanding how identities are formed and how they impact adolescents’ well-being in contemporary societies.

Funding Statement

This work was supported by a grant from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (ERC-CoG IDENTITIES Grant agreement N. [101002163]; Principal investigator: Elisabetta Crocetti).

Disclosure statement

No potential conflict of interest was reported by the author(s).