Abstract
This paper will review the evolution of Social Identity Mapping as a potential theoretical and analytical framework for interpreting complex online behaviours. It will commence with an exploration of the evolution and utility of social identity mapping, before critically examining the application of social identity mapping in empirical research, both online and in person. It will conclude by presenting an argument for the ability to, and need for, objective markers of the self to be developed as a mechanism for understanding multiple salient social identities, across multiple online platforms, in a manner that is generalisable to a range of Contexts.
Examining social identities through online data is a rapidly expanding field of study. In the last decade, as digital humanities and computational social sciences have emerged as disciplines in their own right, the increase in scale of social networking sites and their use by social groups ranging from the local mom’s group to extremist ideology-based groups, has created a plethora of data. A greater interest in online environment research has evolved alongside growing awareness of the ability for exploitation by extremist actors, conspiracy theorists, and groups of concern. Platforms ranging from topic-specific forums to social networking sites to online gaming have all provided data (Kowert et al., 2022). Particularly in a post-Covid world, the role of the internet cannot be removed from identity formation, whether direct personal contacts or indirect online contacts occur first, both play a role in supporting the development and maintenance of various social identities (KhosraviNik & Amer, 2020; KhosraviNik & Esposito, 2018). Whilst scholars have examined online environments through the theoretical lenses of SIDE (Social Identity Deindividuation Effect, see also: Klein et al., 2007; Postmes et al., 1998; Postmes et al., 2001; Spears, 2011; Vilanova et al., 2017) and SIMCA (Social Identity Model of Collective Action, see also: Thomas et al., 2015; Thomas et al., 2020, van Zomeren et al., 2008), the psychological mechanisms through which identities become active and govern online personal and group behavior, in particular cognitive engagement and efforts of expression, need further exploration (Kowert et al., 2022). Indeed, many studies of social identity online either present an atheoretical, narrative construction, or a theoretical framework which is not applied to the data analysis.
This paper will review the evolution of Social Identity Mapping as a potential theoretical and analytical framework for interpreting complex online behaviours. It will commence with an exploration of the evolution and utility of Social Identity Mapping, before critically examining the application of Social Identity Mapping in empirical research, both online and in person. It will conclude by presenting an argument for the ability to, and need for, objective markers of the self to be developed as a mechanism for understanding multiple salient social identities, across multiple online platforms, in a manner that is generalisable to a range of contexts. By “objective markers of the self” we mean measurable digital traces reflecting identity when there has been no self-report of identity relations and where a self-reported Social Identity Map provided by the owner of the identities is not available, possible or reliable. The aim to extend existing Social Identity Mapping methodologies to enable use of objective markers of self for building social identity maps when self-report is not available, possible or reliable, is the novel contribution of the approach advocated for in this paper and the suggested response to the empirical work reviewed below.
Theoretical Framing: The Social Identity Perspective
Identity is defined by Stets and Burke (2014) as: “a set of meanings that defines individuals in terms of the roles they occupy, the social categories or groups they belong to, and the individual characteristics that define them as unique persons” (Stets & Burke, 2014, p. 412). The social identity perspective (Tajfel, 1974, 1978; Tajfel & Turner, 1986) emphasizes motivation by individuals for positive and distinctive social identification, where belief systems which are valued by the individual are reflected in the norms of the groups to which they belong (Deaux & Burke, 2010; Tajfel & Turner, 1986; Turner et al., 1987; Turner & Oakes, 1986). The social identity perspective is, in its simplest terms, the theoretical framework explaining meaning seeking and meaning creation, from the perspective of both individuals and groups (Richards, 2022; Turner et al., 1987; Turner & Oakes, 1986). Reflecting on the social identity perspective as comprising both social identity theory and self-categorization theory, Turner and Oakes (1986) acknowledged that group membership is a dynamic and interactional process, subject to regular cognitive evaluation in order for the individual to assign meaning and identify representation from social structures or ideological belief systems. They posit that: “the raison d’etre of social psychology is to solve scientifically a paradox… the issue is can there be and how can there be a non-individualistic science of the individual, a social psychology in the fullest sense?” (Turner & Oakes, 1986, p. 237).
A social group identity is recognized as a cognitive entity that provides meaningful interaction and representation of belief systems at various stages of the individual’s lifespan (Tajfel, 1974). Group membership is most commonly expressed through the performance of normative behavior consistent with the salient groups with which an individual identifies (Douglas et al., 2005; Stryker & Burke, 2000). These groups may be traditional social categorizations, such as gender, race, ethnicity and nationality; or opinion-based groups or issues-based groups (Curtin & McGarty, 2016; Hogg & Turner, 1985; McGarty et al., 2014; Richards, 2022).
Acknowledging the cognitive effort in processes of social identification, Burke & Reitzes (1991) contend that identity stability and identity maintenance require the ability to learn and internalize functions of normative group behavior relevant to each identity (Burke & Reitzes, 1991). In contrast, Cast and Burke (2002) posit that if a social identity develops from a period of stress, the internalization of norms is likely to be too challenging to be authentic and sustainable (Cast & Burke, 2002). This position advances the theory that group identities developed during times of stress, regardless of when they are activated in terms of identity salience, are more likely to provoke emotions of anxiety and depression, rather than reinforce favorable belief systems (Cast & Burke, 2002; Hogg, 2012, 2014; Stets & Burke, 2014).
These psychologically meaningful connections, whether entirely positive, mixed, or indeed, negative, are cognitively processed and internalized by the individual. Under the social identity perspective, this is known as comparative fit, normative fit, and perceiver readiness (McGarty, 1999; Oakes et al., 1991; Turner et al., 1987; Turner et al., 1987, 1987).
Self-categorization theory explains social categorizations shaped by normative fit and comparative fit as individual perceptions of group-based stimuli (Nolan et al., 1999; Richards, 2022). Haslam and colleagues extend this definition in relation to national identity stating: normative fit refers to the content related aspects of the match category specifications and the instances being represented. For example, to categorize a group of people as Americans rather than Australians, they must not only differ (in attitudes, actions, etc.) from Australians, more than from each other (comparative fit), but the nature of this difference must be consistent with a perceiver’s content related expectations, beliefs, and theories about the categories (Haslam et al., 1999, p. 810).
Neither comparative fit nor normative fit can be achieved without the willingness of an individual to embrace a collective group identity (Haslam et al., 1999; McGarty, 1999; Oakes et al., 1991; Onorato & Turner, 2004; Turner et al., 1987; Turner & Oakes, 1986). Perceiver readiness is explained by Onorato & Turner (2004) as follows: Personal identity refers to ‘me’ versus ‘not me’ categorizations – all the attributes that come to the fore when the perceiver makes interpersonal comparisons with other ingroup members. Social identity, on the other hand, refers us versus them categorizations – all the attributes that come to the fore when the perceiver compares his or her group (as a collective) to a psychologically relevant outgroup (Onorato & Turner, 2004, p. 259).
How we choose to identify, self-categorize, and our readiness to do so, is core to the ability to map our social identities through times of stress, illness, challenge, identity threat, identity change, identity repression, rejection, or growth.
Self-categorization theory within the social identity perspective is described by Spears (2011) as follows: If social identity theory is primarily a theory of intergroup relations, self-categorization theory can be seen as a more general theory of the self, of intragroup as well as intergroup processes – possibly as close as we come in contemporary social psychology to a grand theory (Spears, 2011, p. 208).
Self-categorization allows the individual to divide the social world into groups which reflect their emotions, views and perceptions (Abrams et al., 1990; Richards, 2022). These categorizations, in turn, direct behaviors and norms, reinforce intra-group conformity, and polarize outgroup norms and behaviors (Abrams et al., 1990; Richards, 2022). Self-categorization theory additionally extends the concept of identity salience. Termed depersonalization, social identity salience is the cognitive, process-based predictor of when a social identity will take precedence over a personal identity (Abrams et al., 1990; Onorato & Turner, 2004; Richards, 2022; Spears, 2011).
Theoretical Antecedents to Social Identity Mapping
Initially developed to support clinical psychology programs for individuals experiencing health, illness, and significant lifestyle transitions, Social Identity Mapping draws on principles developed in the organisational psychology model, ASPIRe (Haslam et al., 2003), resulting from the application of theoretical notions from the social identity perspective.
ASPIRe recognised that when there is substantial change in a workplace, the success in retaining and supporting quality employees is anchored in how they view their identity and position within the organisation. The theory places the employee at the centre of the solution. It considers their personal and social capital as two separate, but equally valuable modes of support. Personal capital values the relationships and roles outside the organisation, which support the individual during times of transition and identity conflict; whereas social capital considers the importance of the groups with which to associate. Where individuals record a high level of personal and social capital, ASPIRe predicts a greater likelihood of flexibility, resilience and adaptation to a new work environment. However, when either dimension, be it interpersonal relationships or belonging to social groups, was lacking, an individual was more likely to struggle during a substantial restructuring. Early identification allows for greater support to be provided to these employees, and has been recommended by ASPIRe theorists as a strategy promoting retention of quality staff (Haslam et al., 2003).
Recognising the importance of work in an individual’s life from a dual perspective, namely the time one spends engaging in work, and the meaning work can give, ASPIRe recognised the impact on individual wellbeing and mental health, and sought mechanisms to maximise this during times of stress, transition and uncertainty. A decade later, Social Identity Mapping theorists sought to adapt this model to achieve the same maximisation of wellbeing and mental health in individuals experiencing stress, uncertainty, and undergoing change in their lives, and, to describe and map diagrammatically and relationally, the self-reported aspects of the identities key to achieving wellbeing for an individual as a social being with multiple possible social identifications (Cruwys et al., 2016).
Bentley et al. (2020) refer to ASPIRe as the ‘embryonic form’ (p. 216) of Social Identity Mapping. Absent a specific data retrieval methodology in ASPIRe, though, the priority of hierarchical identities and leveraging social capital centre to ASPIRe, however, was not immediately demonstrable. Drawing on the key principles of ASPIRe and a social identity approach, Social Identity Mapping in its most simple sense, emerged in clinical settings treating stroke recovery as a visual representation of social networks, social identities, and support mechanisms (Bentley et al., 2020; Cruwys et al., 2016).
In advancing Social Identity Mapping from a paper-based to online-based self-report response tool, Bentley and colleagues (2020), reflect on the influence of
Cruwys and colleagues, in the development of Social Identity Mapping, posit: It is well established that measuring social identity is highly reactive so that the process of asking whether we identify as Australian (say) is likely both to make the identity more salient and to increase reported identification. In this way, self-report measures that ask participants to reflect upon specific predetermined social identifications will themselves have an impact on those identifications. This, then, is not just a problem of measurement (e.g., see Haslam et al., 1999), but points to the need to develop and validate procedures that allow participants to report subjectively important emergent social identities, rather than forcing them to orient to social categories that may in some contexts be subjectively meaningless. We propose that the measurement of social identity can be transformed into an interactive, reflective, and reflexive process (Cruwys et al., 2016, p. 615).
Social Identity Mapping: An Overview
After brainstorming the different possible self-categorizations in the abstract and not in response to a particular intergroup context or issue, the individuals at the centre of their maps, “were asked to use the size of each Post-it note as an indicator of group importance, such that the names of the most important groups were written on the largest size, groups of moderate importance on the medium size, and groups of least importance on the smallest size (Cruwys et al., 2016, p. 619)”. Additionally, participants were asked to include “a measure of group positivity, which asked participants to rate, from 0 (not positive at all) to 10 (very positive), how positive they felt about being a member of each group in the top left-hand corner (Cruwys et al., p. 619).”
In the initial iteration of the pSIM (paper-based Social Identity Mapping) this was facilitated through sticky notes of various sizes and colors, with the larger sticky notes representing a higher degree of identity priority. In the second iteration of the process of constructing and recording a social identity map which became the oSIM (online-based social identity map), prioritisation of identities was achieved online via responses made on an app by assigning a rank between 1 and 4 (with 1 = being of high importance and 4 = being of lesser importance), instead of using the scaled sizes of sticky notes when the map was constructed in person using physical resources like sticky notes in pSIM. In both pSIM and oSIM, once each identity had been assigned a priority, participants were asked to draw (or select in the online oSIM tool), a straight line between relationships which were compatible or complimentary, and, a jagged line between those which were or had potential to be in conflict with one another (Bentley et al., 2020; Cruwys et al., 2016).
Considered in line with foundational social identity theory, the Social Identity Mapping approach invokes self-categorization and the relative salience of a self-categorized identity at the time of engaging the individual. Thus, the groups which are self-identified with greater weight in the hierarchy may be subject to variation in prominence depending on salient identities.
Whilst the social network – that is, which groups and by extension, individuals, are part of an individual’s network – is indeed a part of Social Identity Mapping, the evolved process goes beyond the notion of who knows whom and provides insights into the value of each group membership to the individual (Bentley et al., 2020; Cruwys et al., 2016). That is, whereas traditionally mapping has approached each point of a social network in isolation, Social Identity Mapping aims to present a cohesive, integrated approach showing types of relationships (eg. compatible or conflictual) between identities and points within a social network. Further, Social Identity Mapping seeks to quantify the priority of each group membership an individual associates with in the abstract; weight the impact of the group membership regarding positive or negative theoretical based implications; and evaluate the compatibility of varying group memberships (Cruwys et al., 2016).
Whilst not entirely dissimilar to the ASPIRe model in valuing both personal and social capital, described here as interpersonal relationships and group membership, Social Identity Mapping takes the process a step further in examining the hierarchy and interactivity of an individual’s identities. Again, the personal identity is placed at the centre, with terminology varying across the literature to label this position as the ‘core’ or ‘central node’ (Cruwys et al., 2016; Yang et al., 2021). From here, the individual is asked to list the roles, relationships and group memberships which they value. Consequently, for the purpose of developing objective markers of identity for Social Identity Mapping, as opposed to engaging self-report, understanding how multiple, self-categorized identities are expressed or described as important or salient at any given time is critical in objectively evaluating the priorities of those identities to the social and personal identities of the individual.
Additionally, the centering of the participant and their subjective self-report has been viewed in both earlier research (Jetten et al., 2010), and subsequent research (Best et al., 2017) as potentially advantageous in therapeutic settings where health developments have resulted in a substantial shift in identity, leveraging existing networks of identities to support transition and enhance wellbeing (Cruwys et al., 2016; Haslam et al., 2008; Jetten et al., 2010). Anchored in the advancement of the social cure theoretical framework, that highlights as a social determinant of health research that “draws attention to the ways in which a person’s social relationships, social networks, social support and other social identity-based resources feed into health outcomes” (Haslam et al., 2018, p. 14; also citing Jetten et al., 2012), the move from use of a paper-based Social Identity Mapping (pSIM) tool to an online Social Identity Mapping (oSIM) tool, aimed to develop epidemiological work in the space of health and wellbeing, by capturing a wider participant pool doing online self-reported mapping, and thus greater depth and breadth of data relevant to the exploration of identity benefits, identity change and identity intervention (Bentley et al., 2020).
The evolution of the oSIM tool further aims to reduce inevitable human error in data collection and data entry; reduce the necessity for in-person delivery (Bentley et al., 2020); and reduce time taken to collect what may be categorized as primary data and metadata. That is, the primary data may be the subjectively self-reported map of social group identification, and the metadata may be the value-based attributions of priority and importance, and the interactions between groups, and group compatibility. The discussion below contemplates how to achieve such mapping without subjective self-report and via analysis of objective markers of identity expression, priority and (in)compatibility.
The ability to harvest subjective self-reported data which informs complex debates in social psychology, such as group importance and group compatibility, is a strength of both the pSIM and oSIM models. From both a methodological and theoretical standpoint, they contribute a new process for holistic modelling of the individual and their social identities which in turn provides scalable data that was previously only obtained through pre-defined psychological testing metrics such as social identification scales (Bentley et al., 2020; Cameron, 2004; Cruwys et al., 2016). However, contrary to the novel approach argued for in this paper, both pSIM and oSIM still require self-report and do not allow mapping to occur with self-report is not available, possible or reliable, for example, in intelligence analysis of emerging extremists posting online.
The term ‘Social Identity Mapping’ is also used by Jacobson and Mustafa (2019), where an individual considers the priorities they assign to their socially constructed identities. These may or may not include group membership, and includes factors such as gender, sexuality, race and religion; before assigning value and assessing each in terms of positionality to research (Jacobson & Mustafa, 2019).
Documenting maps of group memberships, which facilitate meaning making for individuals as social group members in context, is a key strength of the social identity approach, embedded in Cruwys et al., (2016) and Bentley et al., (2019)’s models. However, the emphasis on mapping
The social identity perspective, which is employed for the development of Social Identity Mapping, posits that meaning is made through collective action. That is, whilst an individual exists with certain key features, the way they behave and understand the world is through group affiliations selected through the process of self-categorization (see also: Swann et al., 2012; Swann et al., 2012; Tajfel & Turner, 1986; Turner et al., 1987). Consequently, a salient personal identity is likely replaced by a salient group identity, one where the individual is depersonalized and members of the collective are largely interchangeable (see also: Lee et al., 2001; Postmes et al., 1998; Postmes et al., 2001; Spears, 2011, 2021).
However, identity fusion theorists such as Swann et al. (2012) argue that social psychological metrics should consider both the personal and social capital of an individual defining identity fusion as: “. . . a unique form of alignment with a group, one that entails a visceral feeling of oneness with the group. This feeling of oneness is associated with increased permeability of the boundary between the personal and social self. Such elevated permeability will increase the likelihood that the fused person’s personal identity will influence his or her group identity and vice versa” (p. 441).
This urging is not dissimilar to what ASPIRe (Haslam et al., 2003) set out to achieve. Swann and colleagues (2009) specifically note the prominence of highly fused identities in leadership roles; that is, individuals retain differentiation and advocate for new mechanisms for advancing the ideological beliefs which they both personally subscribe to, and which form the basis of the group. This is further reflected in their work determining that the actions of the individual may be against self-interest, and that even when the group identity is salient, individual differences and unique interpersonal relationships are noticeable (Swann et al., 2012).
Mavor and Ysseldyk (2020) illustrate the complexities of identity fusion through an examination of religiosity. They argue that religion is at the nexus of the personal and social self; with expression occurring at both individual and collective levels (Mavor & Ysseldyk, 2020). Positing that religion is ‘all-encompassing (in) nature’ (Mavor & Ysseldyk, 2020, p.2) they theorize that whilst potentially applicable across a range of identities, religion requires both the individual and the collective in order to make meaningful connections.
This concept of all-encompassing nature is applicable to a range of circumstances and identities which are consistently salient. The ASPIRe model, which recognised the importance of personal self-categorisation concurrent with social self-categorisation, could arguably be invoking the same principle, as for religiosity, due to the centrality of occupational identity in contemporary life and the fusion of both personal and social levels of expression of occupational identity. When considering using objective rather than subjective markers of self for the purpose of Social Identity Mapping, it is necessary to consider how this may be relevant particularly with the affordance of perceived or actual anonymity in the online environment. This perceived or actual anonymity permits an ideological belief that is salient, but not central, to be more easily expressed as it is yet to be comparatively positioned in line with other competing identities that may be compatible or conflictual. Therefore, the online environment may afford a greater sense of perceived acceptance through perceptions of anonymity for expressing views which are contextually normative but are in conflict to the offline self (see also: Postmes et al., 1998; Spears, 2021; Suler, 2004).
Social Identity Mapping in Contemporary Scholarship
The concept of Social Identity Mapping is both explicit and implicit in contemporary scholarship. Yang et al. (2021) consider comparative social identity maps of employees’ engagement with their colleagues in both online and offline environments, using a case study of hospitality workers. Whilst this research did not use the specific pSIM or oSIM tools, the use of weighted networks and questions of compatibility are relevant to this review.
Using the term “social networks” to refer to social media platforms, and
Perhaps most interesting in their work, was the analysis that social identifications were amplified online, when computational inference of identification via frequency of engagement and depth of content measures showed enhanced identity expression online vis-à-vis offline behaviours. Whilst online environments were strategically engaged with for work-related communications and career advancement, identity dissonance presented concurrently for individuals offline, particularly where a high degree of interdependence was required for success (Yang et al., 2021). That is, whilst the same group memberships and individual relationships may be highly compatible offline; in an online context they are more likely to be incompatible.
A further example of inferring social identification and relationships between identities via online behaviours occurred in work by Boduszek et al. (2021). Whilst still utilizing the term social network analysis with weighted compatibility metrics as opposed to the aforementioned pSIM and oSIM, Boduszek and colleagues (2021) advance the methodology by weighting nodes (variables within the study) and edges (data reflecting correlations and compatibilities between the variables). Not entirely dissimilar to the assigning of priority seen in Social Identity Mapping, the weighting allows researchers to consider the subjectively perceived priority of each connection in the individual’s network (Boduszek et al., 2021). Applying this process of mapping a social network with weighting, the researchers suggest the term “Criminal Social Identity” be used to reflect the importance of criminogenic behavior to an individual’s broader social identity, through conscious and cognitive engagement with a criminal group (Boduszek et al., 2021). Consistent with the social identity perspective, the higher the priority of a
Where a positive sense of self worth is recognized by some individuals in studies on criminal social identity, the negative impacts on social identity, inclusion and rejection of such identities are noted by Fox et al. (2024) and Kyprianides et al. (2019); particularly when dealing with marginalized populations. Often overrepresented in prison populations, the intersectional vulnerabilities of Indigenous youth offenders present not only layers of increased risk, but increased opportunity for social identity intervention. Using Social Identity Mapping alongside semi-structured narrative interviews in a sample of nine Maori youth in Aotearoa, Fox and colleagues found the weighting of social identity priorities to be largely in friends, family, and the immediate community where the individuals lived (Fox et al., 2024). These groups simultaneously presented opportunities and challenges, with participants expressing the negative impact of, and desire to remove some groups (identity transition), to an appreciation during rehabilitation programs of the need to sever connections built up as a consequence of offending behavior (identity reintegration) (Fox et al., 2024). Unlike the prior study on Criminal Social Identity, belonging to an offending social group was not expressed as having any positive impact on social identity. Additionally, the importance of ethnic group membership to these individuals raised the question of future research examining the embedding of cultural competencies and knowledge in the Social Identity Mapping process to extend understanding of centrality and identifying greater avenues for identity intervention in the context of desistance (Fox et al., 2024). This arguably links to the concept of highly fused personal and social identities, where centrality of that fused identification (eg. religiosity or occupational identity) is consistent and consistently salient, as an important future dimension of Social Identity Mapping.
Scholarship into addiction recovery has also employed both pSIM and oSIM methods in the development of the SIM-AR (Social Identity Mapping-Addiction Recovery) model. The SIM-AR model centres research on the importance of multiple group memberships, particularly group memberships which are compatible and of quality to individual wellbeing and mental health, including lower rates of depression (Best et al., 2017; Cruwys et al., 2014; Haslam et al., 2008, 2018; Jetten et al., 2015). For addiction recovery, the quality of connectivity with not only a plethora of groups, but also with those which focus on support and positive values, is of high priority to maintaining wellbeing (Best et al., 2017; Haslam et al., 2018).
Use of linguistic markers, sentiment analysis and engagement metrics relating to online posts in addiction recovery fora on Facebook allowed identity to be computationally inferred with a methodology influenced by Social Identity Mapping resulting in maps of social capital (Best et al., 2017). As a longitudinal study, the research measured identity change through continuation of, and variation to, engagement on the forum and between users over time. The development of objective markers of identity through linguistic patterns and sentiment analysis, allowed the researchers to quantify differentiations of recovery in the community, and weight the impact of interpersonal bonds in the process of wellbeing. It further considered salient events of a positive nature, recording the hypothesised ‘spike’ in engagement, interaction, and increased sentiments of success (Best et al., 2017). The findings of this study ultimately informed the development of the Social Identity Model-Addiction Recovery (SIM-AR) model, and application of Social Identity Mapping, in addiction recovery research. Of specific importance to the development of SIM-AR was the finding that whist participation in online communities aided addiction recovery, it was the interpersonal relationships and personal capital that were critical in the maintenance of sobriety (Beckwith et al., 2019; Best et al., 2017).
Building on the initial model of Social Identity Mapping, and objective markers of addiction recovery was evidenced in the work of Best and colleagues (2017), SIM-AR captures the normative content of groups, particularly levels of substance use as a group-based normative behavior (Beckwith et al., 2019). In this study, the data showed that dependence on group membership alone was associated with a greater risk or relapse. The ability to map and identify an absence of inter-personal relationships or capital, in this instance, informed the ability to intervene and promote greater levels of support to ensure wellbeing (Beckwith et al., 2019). The research reflected that increased understanding and application of social identity processes has potential to aid clinical and therapeutic treatment programs, as well as contributing to theoretical advances (Beckwith et al., 2019).
Additionally, whilst not aligned with social identity mapping practices, we recognize the broader contributions to computational inferences of social identity, particularly in assessing ideological based extremist identities. This work traditionally has involved analysis of corpus texts from online forums dedicated to ideological topics, which is both reliant on linguistics and provides contextual frameworks of analysis through self-categorisation by individuals choosing to engaging in such online spaces (Chee et al., 2023; Miró-Llinares et al., 2018). Scholars who have undertaken this identity focused research have noted the limitations of linguistics, or natural language processing (NLP) models. Indeed, noting the lack of integrated research (Ebner et al., 2023) and limited incorporation of theoretical framings, notably deeming the concept of “othering” as insufficient (Kowert et al., 2022); the need for novel computational methods which are anchored in existing theory, is well documented in existing scholarship.
Discussion and Future Directions: Beyond Self Report Using Objective Markers and Objective Mapping
The scholarship extending Social Identity Mapping into addiction recovery utilized LIWC sentiment and frequency analysis to establish objective markers of normative group behavior online, and to contribute to evaluations of the quality of interactions between group members. In our work we will be building on the praiseworthy and noted benefits that Best and colleagues (2017) outline when objective markers assessing real-time changes to social capital and networks are constructed in analyses, “without the intrusion or bias of self-report (Best et al., 2017, p. 172)”. The researchers note a concern relating to ethical considerations when using social media data in this space, and they express concerns over processes such as consent waivers to social media data (Richards & Nolan, 2024) when examining issues with clinical implications and with the potential for engaging with providers to undertake identity intervention as a result of observations of objective markers of identity (Best et al., 2017). However, they concur with the position of this paper that Social Identity Mapping utilizing objective markers of self can be developed from a more generalised social psychological research perspective, to capitalize on real-time presentations of social identity in a diverse and continually evolving space of online communities (Best et al., 2017).
The concept of objective markers of personal and social identity, particularly with the continued growth of social media and computer mediated communication-based research is underscored by logistical realities in engaging participants. Big data research, quantitative sampling, and mixed method Social Identity Mapping processes require access to data where either the sheer scale of data or invocation of markers of anonymity mean direct participation is not viable. Thus, whilst the data and the social identities of a presented online user can assist in advancing theoretical designs and implications of Social Identity Mapping in practice, identity intervention is not ethically appropriate in this model. Whilst the burden is then on the researcher to infer meaning, the development of largely objective markers of social identity also allows the mitigation of limitations of self-reporting in social psychological research including identity concealment, recall bias, and deception.
Self-concealment of social identities is linked to wellbeing and self-esteem in the literature. Of particular note, stigmatized or marginalized identities, when expressed and accepted, are likely to increase wellbeing and self-esteem (Chaudoir & Quinn, 2010; Mummendey et al., 1999). By contrast, rejection of a social identity by members of an in-group is evidenced to lead to a continual concealment of said identity, with decreased wellbeing and self-esteem (Chaudoir & Quinn, 2010; Rodriguez & Kelly, 2006). Identity concealment is also evidenced in research on experiences of relative deprivation, where the expectations of support or endorsement of a given identity are either not realistic, or are unmet, by the ingroup; particularly in the first instance of disclosure (Derks et al., 2011; Mummendey et al., 1999; Rodriguez & Kelly, 2006; van Veelen et al., 2020).
In evaluating the potential for the development of objective markers for Social Identity Mapping, the ecosystem through which social identity is expressed and data is retrieved, can heavily impact which identities are disclosed and to what degree. Rodriguez & Kelly (2006) present metrics of satisfaction with identity disclosure linked to perceived acceptance by the social group, with Chaudoir & Quinn (2010) emphasizing the expectations of reciprocity and positive reinforcement as precursors to social identity disclosure. This is of particular importance to potential developments of objective markers, as the unique ecosystem of each type of social media platform or online forum may facilitate the expression of social identities which are concealed in offline, interpersonal environments. The perceived anonymity of online engagement in this instance offers individuals a space to express, maintain, and perform social identities which may be deemed incompatible in their daily life (see also: Postmes et al., 1998; Postmes et al., 2001; Spears, 2021; Suler, 2004). In turn, naturally, the commonsense limitation to this process is any assumption that all aspects of life will be disclosed due to anonymity, perceived or actual, and must be considered in the development of object markers of identity for Social Identity Mapping, particularly regarding questions of identity compatibility.
At the same time as balancing questions about social identity concealment and disclosure, we cannot eliminate the possibility of attempts at deception in online environments, fuelled in part by the perception or actuality of anonymity. Future developments in the space of objective markers for Social Identity Mapping using digital data needs to give careful consideration to both behavioral indications and sentiment-based expressions which may allude to the motivation to disclose a disingenuous identity for malevolent purposes.
The theoretical development of Social Identity Mapping, and review of scholarship aligned with the core principles in research design, demonstrates the breadth of data and value that weighted compatibilities between identities offers to social psychological research. Concurrent with this is the continually evolving need to develop theoretical frameworks which can be employed with specific attributes relative to online ecosystems and social media platforms.
The positive review we have given of relevant literature above indicates substantial value of using objective markers of self to the building of social identity maps when type of participant, topic discussed online, the extremist nature of views expressed all means that researchers must move beyond self-reporting as self-report is not available, possible or reliable. This would further address logistical challenges of research (Richards & Nolan, 2025) in the space of computer mediated communications, and is thus simultaneously an opportunity to further apply Social Identity Mapping with the concept of objective markers of identity.
Whilst this review has identified some of the benefits of research which can identify and include real time changes in social networks, tone and sentiment of expressed identities, there is a need to consider what objective markers would look like to this end. For example, in a pilot study testing the feasibility of one researcher taking a digital or analog ethnographic approach (Forberg & Schlit, 2023), we were able to draw social identity maps at the group level from analysis of online posts by the same users found on both Telegram and Reddit in October to December of 2023 following Hamas’s invasion of Israel. At least nine different identities could be drawn objectively in a social identity map. The relationships between those identities as being either compatible, incompatible or mixed compatibility (sometimes allied but sometimes not) could be drawn even though not done with self-report methodology. The first author of this paper was able to track the posts of these nine social identity groups (with a minimum of 10 or more ingroup users per group) using digital ethnographic practices, for at least 2 hours per day over the two platforms in the study period running from October to December 2023. Longitudinal changes and redefinition of identity perspectives were also able to be noted: for example, both “Zionists” and “De-colonialists” both adopting a unifying misogynist stance over time in reaction to some of the media debates, despite initially expressing incompatible identities (Richards, 2025a; Richards et al., 2025). In this and related work, tools such as LIWC, Communalytic, and Linktree were highly useful in initial identification of users to track across platform.
Following this pilot study confirming the ability to identify individuals and map groups across platforms, absent of self-report and to a relative degree of certainty, the next stage is to extend the research to apply concepts of behavioral effort to individual social identity maps. We hypothesize that anchoring individual identification in periods of identity salience, such as exigent ideological events, will allow longitudinal analysis of identity presentation one month prior, and one month following the event, to develop matrices of cognitive engagement and map multiple social identities with changing degrees of centrality (Richards, 2025b).
Conducting this type of work includes an acknowledgement of the limitations of linguistics in a digital research space, when geographical location or language is not an inclusion criteria.
The internet is commonly associated with Western representations of identity, groups, and linguistics, but specific platforms can impact this presumption and exclude large volumes of potentially valuable data. Thus, it is proposed that future directions of research move beyond sentiment analysis and also consider the elements of cognitive engagement, or behavioral effort, demonstrated across various platforms, and by users across their online presentation of the personal and social self. Such behavioural matrices were used in the single-platform study of far-right extremism in the US by Richards (2022). The use of objective behavioural matrices operationalising cognitive engagement used in preference to, or in addition to, sentiment analysis, adds to the development of predictive indicators of psychological phenomena, such as trust, motivation, loyalty, and intent.
The theory presented within this review is critically reviewed as offering substantial value to processes where self-report is a viable option. The value of the methodology to understanding social identity networks to support processes of identity change in times of stress. The evidenced literature demonstrates a generalizability of the concept of social identity mapping across change in organizational settings as noted with the precursor model of ASPIRe; individual health journeys, as noted in addiction recovery; and re-integration of offenders as noted in studies on Maori youth and criminal social identity; which adds strength to the proposal of expanding the model to include objective markers of social identity. Absent of the ability to access reliable modes of self-report, this approach will enable reseaachers and practitioners across the above-mentioned fields, and into realms of security and intelligence modelling, to ethically and practically engage with volumes of data that would otherwise be impractical and inaccessible.
Footnotes
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr. Peita L. Richards is the recipient of an Office of National Intelligence National Intelligence Postdoctoral Grant (project number 202308) funded by the Commonwealth of Australia.
Declaration of Conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.