Understanding actor engagement through psychological ownership: a multi-sport brand perspective

Contemporary service markets are characterised by multi-brand interactions where consumers allocate their time, attention and resources across multiple touchpoints rather than exclusive relationships (Sharma et al., 2020). Engagement scholarship increasingly acknowledges that individuals can engage with multiple touchpoints (Dessart et al., 2016) and invest in interconnected relationships with other actors (forming brand repertoires; Dawes and Nenycz-Thiel, 2013). Psychological ownership refers to the state in which individuals feel as though an object (i.e. a brand) is theirs, even in the absence of legal rights or claims (Pierce et al., 2001; Van Dyne and Pierce, 2004). Within marketing and service contexts, psychological ownership as a concept has emerged as an insightful explanatory foundation for engagement (Conduit and Chen, 2017), as feelings of ownership over a brand can facilitate engagement (Kumar and Nayak, 2019a) by influencing product demand (Fuchs et al., 2010), consumer-product affect (Shu and Peck, 2011) and positive word-of-mouth (Hair et al., 2016). These feelings are critical for scholars and managers alike to consider, as they arise across cultures, countries and contexts (Pierce et al., 2003) and connect the focal object to the self, supporting self-expression and motivating investment and protection (Belk, 1988); subsequent engagement enhances competitive performance (Hollebeek et al., 2022).

However, there is limited understanding of how an individual's engagement with one brand is shaped by simultaneous interactions with other brands within the same competitive field (Habachi et al., 2023; Hollebeek et al., 2022). Most empirical research to date focuses on examining engagement as a dyadic relationship between a consumer and brand (Alexander et al., 2025). As such, scholars have called for research into relevant psychological mechanisms that influence subsequent engagement (Uude et al., 2025). Psychological ownership has been investigated within the engagement literature (Harmeling et al., 2017) but largely studied in relation to single focal targets, implicitly assuming that ownership operates independently of other potential objects (Kim et al., 2024). Contrarily, in multi-brand service contexts, consumers likely hold varying degrees of psychological ownership across multiple brands simultaneously, creating distinct configurations of ownership within their broader engagement field (i.e. one brand may be strongly experienced as being psychologically owned, while another is regarded as peripheral, neutral, or even oppositional; Schreuder et al., 2023). Limited attention has been given to how these dimensions are structured in competitive, multi-brand contexts.

This research seeks to respond to this gap in understanding by undertaking an empirical investigation into how psychological ownership unfolds in multi-actor settings and shapes engagement (Kim et al., 2024; Uude et al., 2025). This not only includes favourable outcomes for brands, but also understanding adverse outcomes of engagement when varying levels of possession over multiple brands is felt (Baker et al., 2021; Kim et al., 2024; Koklic et al., 2025), and even passive/disengaged implications concerning those who are still present in a service ecosystem and may still invest their resources (e.g. money; Kuchmaner et al., 2025; Kumar and Nayak, 2019b). To examine the research gap at hand, we adopt a conjoined ownership perspective (referring to the degree to which psychological ownership is simultaneously high (or low) across both brands; operationalised as the multiplicative product of ownership towards brand x and ownership towards brand y [POx × POy]) in which engagement outcomes are theorised to arise from the joint presence and interaction of focal and rival psychological ownership, rather than from their independent effects. That is, the effects of ownership are not merely additive; rather, different ownership combinations represent distinct psychological states with potentially divergent implications for engagement. Consistent with this conjoined perspective, we argue that psychological ownership influences engagement through two distinct mechanisms that are particularly salient when multiple brands coexist within a consumer's engagement field, namely territoriality and apathy. Together, these mechanisms suggest that psychological ownership in multi-brand contexts does not simply increase engagement in a linear fashion. Instead, it structures the direction and magnitude of resource investment across brands. By conceptualising psychological ownership as shaping engagement through territoriality and apathy, this study extends psychological ownership theory beyond single target settings and contributes to engagement research by clarifying how ownership dynamics operate within competitive service ecosystems.

We examine these relationships in the context of sport, which represents a high involvement service domain in which brands are emotionally salient, socially embedded and positioned in direct competition (McDonald et al., 2022; Yoshida et al., 2014). Supporters may develop strong psychological ownership towards teams, experience rivalry dynamics or remain casual participants within the broader system (Palau-Saumell et al., 2022; Tyler and Cobbs, 2015). This context provides an appropriate setting to examine how psychological ownership across multiple brands relates to favourable and unfavourable engagement, as well as variations in engagement intensity. Our study addresses the following question:

To answer this question, we survey and conduct PROCESS mediation analysis on 379 football supporters, linking psychological ownership felt across two brands to favourable (positively valenced) and unfavourable (negatively valenced) engagement outcomes, mediated by territoriality and apathy. The research highlights how dynamic levels of psychological ownership felt towards multiple, salient brands (e.g. high/low towards brand x; high/low towards brand y) shapes both engagement valence and intensity; this can amplify positively valenced engagement yet also induce unfavourable engagement outcomes. In doing so, the study advances understanding of psychological ownership in multi-brand contexts and clarifies how ownership-related mechanisms influence engagement within competitive service settings. The next section provides an examination and critique of extant literature on the engagement concept and psychological ownership.

Engagement research in marketing is typically associated with customers and focussed on the dyadic customer/firm relationship; this can be seen as neglecting the complexity of multiple engagement objects (Jaakkola et al., 2018). Scholars now advocate for a broader view, advancing research on “dual foci engagement” (Dessart et al., 2016), “triadic engagement” (Fehrer et al., 2018) and now “(multi) actor engagement” (Brodie et al., 2019), defined as a dynamic process where multiple actors in service systems integrate and invest resources in interactions within one another in a service system (Li et al., 2017). This shift also includes “actors” within networks (Alexander et al., 2025) whose engagement varies in valence and intensity, reflecting affective, cognitive and behavioural resource investments (Dessart et al., 2015). The research herein adopts the spectrum view of valence from Li et al. (2018), encompassing positively and negatively valenced resource investments. Similarly, intensity is widely used and defined variously; we define it as the degree of actors' engagement dispositions, and the extent of their resource investments in interactions with other actors and/or brands (Wang et al., 2023).

A decade of research has recognised engagement with multiple objects, including between brand communities and affiliated brands (Bowden et al., 2017), online groups and storytelling (Kaczorowska et al., 2023) and interactions between positively and negatively valenced engagement behaviours (Azer and Alexander, 2020). Nonetheless, engagement research continues to focus predominantly on dyadic customer–seller relationships (Alexander et al., 2025). Scholars call for deeper insight into actors' interactions beyond dyadic relationships (Barrera Rodríguez et al., 2023; Pang et al., 2026) and subsequent influence(s) on one's capacity for (un)favourable engagement (Hollebeek et al., 2022). In particular, it remains unclear how an individual's engagement with a brand is influenced by the presence and/or interaction(s) with another brand(s) (Habachi et al., 2023). Additional actors (i.e. another brand) may facilitate engagement through underlying psychological mechanisms central to decision-making (Uude et al., 2025). Consequently, research is needed on the dynamics of multi-brand interactions and the implications for how firms can leverage or mitigate subsequent implications (Jaakkola and Alexander, 2024; Wang et al., 2023). The next section introduces and discusses psychological ownership as a theoretical lens capable of capturing and structuring resource investment across multiple brands.

Defined as the “state in which individuals feel as though the target of ownership is theirs” (i.e. “It is MINE!”; Pierce et al., 2001, p. 299), psychological ownership is comprised of a cognitive-affective core encompassing an individual's evaluation of and emotional attachment to the target of possession (Jussila et al., 2015). An individual's need to own is rooted in four fundamental human needs (Avey et al., 2009); self-efficacy (one's belief that they can favourably influence their environment, such as sports fans chanting and cheering during matches to motivate their team or put off the opposition; Palau-Saumell et al., 2022); self-identity (defining oneself in relation to cultural meanings, individual experiences and club values; Sumida et al., 2015); having a place (i.e. feeling a sense of belonging with a club and fellow supporters; McDonald et al., 2022) and accountability (calling on others to justify failures and/or propagate successes, given winning is the most important thing to fans, those within the organisation and stakeholders; Sumida et al., 2015). Once any or all these needs are met, feelings of ownership can develop over (in)tangible objects (e.g. ideas, places or brands) by controlling (participating in a firm's management decisions via “one member, one vote” or voicing opinions over team management), coming to intimately know (repeated viewership of matches; learning about the organisation, players, history) and self-investment with the target of possession (investing time, money and personal efforts to support the team; Sumida et al., 2015).

Objects that are psychologically owned are perceived as an extension of the self; the stronger degree of psychological ownership felt over the target of possession, the greater it feels a part of a consumer's extended self (Pierce et al., 2001, 2003). Its application within marketing however is still emerging (Baker et al., 2021), with limited research integrating psychological ownership into the nomological network of engagement since its identified role in the engagement process and subsequent outcomes (Harmeling et al., 2017). Prior research on psychological ownership in marketing has primarily adopted a unidimensional focus (Kumar, 2022), such as considering its influence on consumer preferences (Feng et al., 2023), purchase intention (Khandeparkar et al., 2025) and positive word-of-mouth (Tseng, 2020). Given the complex and often crowded nature of modern service marketplaces, consumers often hold membership to more than one branded community (Kumar and Nayak, 2019b). This has produced a pertinent research gap for academia and practice alike that has remained neglected; what are the outcomes of psychological ownership that manifests in multi-actor relationships upon engagement (Baker et al., 2021; Kim et al., 2024)?

Scholars have identified not only needing to empirically examine favourable outcomes (Baker et al., 2021), but also unfavourable (Khandeparkar et al., 2025; Lyu et al., 2023) and even low/passive manifestations (Bhandari et al., 2024) that pose considerable implications for brands given how easy consumers can harm brands (Do and Bowden, 2025), as well as for firm managers' understanding on how best to allocate, leverage and mitigate resources in crowded, psychologically complex markets (Sharma et al., 2020). The (multi) actor engagement literature emerges as a suitable lens to explore the complexities and dynamics of multi-faceted environments (Behl et al., 2024), given its focus on multiple parties and multi-dimensional (un)favourable outcomes that advances the aforementioned unidimensional exploration. Building on this foundation, we next develop hypotheses explaining how conjoined psychological ownership across competing brands shapes engagement outcomes through territoriality and apathy. The next section provides conceptual elaboration with hypotheses.

Psychological ownership has been shown to stimulate favourable engagement outcomes, including purchase intent (Fuchs et al., 2010), positive affect (Koklic et al., 2025) and extra-role behaviours such as word-of-mouth (Jaakkola and Alexander, 2014) that align with positively valenced engagement. However, gaps exist regarding how dynamic levels of psychological ownership manifest within multi-faceted environments (i.e. multiple brands; Kim et al., 2024), with research primarily examining positive organisational behaviour (Mayhew et al., 2007), brand community spillover effects (Kumar, 2021) and product customisation (Kou et al., 2021).

Psychological ownership is grounded in self-extension processes, whereby objects perceived as owned become incorporated into one's identity and serve as vehicles for satisfying and expressing belonging, efficacy and distinctiveness (Belk, 1988; Pierce et al., 2001). When individuals voluntarily extend their possessions in ways that complement or reinforce their existing sense of self, they tend to evaluate such additions more positively, particularly when these objects contribute to identity affirmation or self-enhancement (Hulland et al., 2015; Pierce et al., 2003).

In dynamic consumption contexts, individuals may integrate multiple category-relevant objects into their self-concept simultaneously, particularly when those objects satisfy distinct but compatible psychological needs (Jussila et al., 2015; Pierce and Jussila, 2011). In sport settings, for example, supporters may retain psychological ownership of a hometown team while also developing ties towards another team due to relocation (Collins et al., 2016), familial ties, participation in fantasy competitions (McDonald et al., 2010), or resonance with symbolic and heritage elements of a club (Palau-Saumell et al., 2022). Although such teams may compete at the market level, they need not be mutually exclusive within the self-concept.

When psychological ownership is simultaneously strong across multiple brands within the same category, both objects may become identity-relevant, increasing category salience and strengthening motivation to invest cognitive, emotional and behavioural resources. For example, English football supporters often display flags during matches that support “club and country” in order to convey pride and allegiance to national and local regional clubs (Hart, 2017). Accordingly, we conceptualise simultaneous psychological ownership across brands as a configuration of self-relevant attachments that may shape engagement towards each focal object. Because consumers may hold psychological ownership towards more than one brand within the same category, we conceptualise psychological ownership as a multi-brand configuration rather than a single-target state.

To capture the extent to which psychological ownership is simultaneously felt across both focal brands, we operationalise the conjoined level of psychological ownership as the multiplicative product of ownership towards brand x and ownership towards brand y (POx * POy). Higher values reflect higher simultaneous ownership across both brands, whereas lower values reflect weaker ownership towards one or both brands. We use the conjoined term as an operational indicator of simultaneous ownership intensity across both brands. In line with this, we hypothesise:

Psychological ownership and engagement are frequently associated with favourable outcomes (Peck et al., 2020), but both constructs have a dark side that has been neglected in favour of studying their positive counterparts rather than exploring interrelationships between them (Kim et al., 2024). This has primarily been explored via adverse organisational behaviours, such as knowledge hiding, exclusion or Machiavellianism (Wang et al., 2019). However, its link with negatively valenced engagement outcomes towards brands and within services/marketing contexts remains less developed (Khandeparkar et al., 2025). Some scholars (i.e. Koklic et al., 2025) have begun to explore psychological ownership's direct relationship with unfavourable engagement outcomes. Given individuals can exhibit simultaneous engagement towards multiple objects that co-exists in varying valences and intensities (Alexander et al., 2018), the bidirectional nature of psychological ownership and engagement (i.e. whether negatively valenced engagement recedes as positively valenced engagement increases, and vice versa) remains an important gap to understand (Doyle et al., 2024; Kumar, 2021).

Psychological ownership is conceptually distinct from negatively valenced engagement. The former reflects an affective and cognitive sense of “what is mine” (Pierce et al., 2001), whilst the latter is differentiated by its more complex and multi-dimensional structure, involving unfavourable affections (e.g. hatred; Rodrigues et al., 2021), cognitions (e.g. rumination; Kähr et al., 2016) and behaviours (negative word-of-mouth; Azer and Alexander, 2020) towards a focal object or other actors. Accordingly, positively and negatively valenced engagement are treated herein as conceptually independent outcomes (Naumann et al., 2020) that may vary in intensity and may coexist towards multiple engagement objects (Alexander et al., 2018).

Stronger feelings of ownership result in individuals being less likely to relinquish access to psychologically owned objects tied to the self (Pierce et al., 2009). Instead, they strive to maintain and protect these, which can increase the perceived specialness of those objects (Liu et al., 2025). Engagement with psychologically owned objects may contribute positively to well-being (Kemppainen and Paananen, 2024) and build familiarity that increases one's sense of satisfaction (Shen et al., 2022). Conversely, perceived loss or threat to owned objects can produce discomfort or a sense of “shrinkage” of the self (Pierce et al., 2001). Thus, psychological ownership not only shapes one's proclivity to engage adversely but simultaneously promotes well-being (Koklic et al., 2025). Stronger conjoined psychological ownership across multiple brands may therefore reduce one's propensity to engage in ways that directly undermine or detract from those brands in which stronger levels of ownership are felt. In line with this, we hypothesise:

Whilst we have accounted for “promotive” forms of psychological ownership (i.e. positively oriented manifestations), research has conceptualised “preventative” forms that are more defensive and conservative (Avey et al., 2009). Actors within networks may have differing goals which can create tension and lead to diminishing others in reaction (Hollebeek et al., 2020). The greater degree of psychological ownership an individual feels towards a target, the greater likelihood they will mark or defend what is perceived as belonging specifically to them at the expense of pro-social behaviour (e.g. “this is MY team, not yours!”; Avey et al., 2009). This phenomenon is defined as “territoriality”, an inherent human instinct involving establishing, communicating, maintaining and restoring behavioural control over the object (Brown et al., 2005).

Unlike psychological ownership, which can be considered a psychological state, territoriality differs in that it requires an external factor that a consumer perceives to (or is about to) infringe upon or threaten one's access to the needs satisfied from the object(s) of possession (Kirk et al., 2018). Negatively valenced engagement encapsulates adverse attitudes and behaviours towards a focal object(s) during interactive experiences (Li et al., 2018), which may arise due to a “critical event” whereby an actor transgresses in some capacity (Juric et al., 2015), such as perceived wrongdoing (Hollebeek and Chen, 2014), service failure (Do et al., 2020) and spillover effects (i.e. a brand community influencing a focal actor's engagement; Bowden et al., 2017).

Scholars have found territoriality to trigger adverse consequences that align with multi-dimensional negative forms of engagement outcomes, where its valence lies in individuals' emotions and cognitions determining behaviour (Wang et al., 2023). Infringement (whether actual or perceived) can evoke immediate affective responses such as anger (Brown and Robinson, 2011), feelings of loss or stress (Pierce et al., 2003) and frustration or outrage (Brown et al., 2005), proceeded by cognitive evaluations of unjustness, the size and consequences of the perceived threat and who can be held responsible (Brown and Robinson, 2011). Behavioural responses are motivated by preventing future infringements (Brown and Robinson, 2011); controlling or limiting access to the target(s) of possession (Avey et al., 2009), negative word-of-mouth (Wang et al., 2019), boycotts (Kirk et al., 2018) and in extreme cases, sabotage, destruction or inflicting physical damage (Pierce et al., 2003).

In competitive settings, rival brands may be construed as encroaching upon the symbolic (e.g. intellectual property) or physical (e.g. geographic location) domain of the owned brand becoming the target of territorially motivated negatively valenced engagement. For example, rivalries such as between Manchester United and Manchester City (Premier League) often infringe on the opponent brand via claims of exclusive ownership (e.g. “this is OUR city”), leading to the antagonising brand becoming a conduit for adverse fan engagement (Edensor and Millington, 2008). Micu and Ashley (2022) likewise find that higher levels of psychological ownership increase one's likelihood to engage in territorial behaviours detrimental to brands on social media when one's sense of possession to an object is threatened by others (or perceived to be). As such, within multi-actor settings where individuals interact with multiple brands, stronger conjoined psychological ownership may heighten sensitivity to perceived infringement. When ownership is felt towards a particular brand within the same competitive field, boundary distinctions between owned and competing objects become more salient, increasing the likelihood of territorially motivated defensive responses. Stronger conjoined ownership may therefore heighten sensitivity to perceived infringement and increase the likelihood that actors respond defensively in order to protect self-relevant attachments. In line with this, we propose:

Despite its unfavourable effects, scholars have found preventative forms of psychological ownership (i.e. territoriality) to simultaneously promote positive outcomes towards objects of possession (Koklic et al., 2025), such as increasing commitment to one's organisation (Brown et al., 2005) and creating/strengthening positive attitudes (Brown and Zhu, 2016). Further research is required to examine when perceived infringement induces favourable outcomes towards brands (Kuchmaner et al., 2025). In addition to its potential to generate antagonistic responses towards rival brands due to anticipated or perceived infringement, territoriality may also intensify attachment-consistent behaviours towards the owned brand by reinforcing in-group solidarity and engagement towards the owned object (Kumar, 2021; Shu and Peck, 2011). Being on the receiving end of hostile treatment from an out-group(s) can lead individuals to strengthen identification and engagement towards in-group members as a counter-negativity coping mechanism (Branscombe et al., 1999); for example, when opposition fans of team y deride a player of team x, fans of team x often cheer louder to drown out boos and support both their team and players as an act of brand defence (Wong and Hung, 2022). Given individuals strive to maintain and take care of objects which they feel psychological ownership towards (Avey et al., 2009), we anticipate an additional exhibition of positively valenced engagement towards possessed object(s) in response to (perceived) territorial infringement to defend and affirm one's ownership and status received from affiliation, and propose:

Within multi-actor settings, there will almost always be individuals who are unfamiliar or uninterested in engaging or interacting with other actors within the service ecosystem (be it unfeasible or lacking the requirements to; Chandler and Lusch, 2014). This has been documented within the fan engagement literature (“non-fans”; see McDonald et al., 2023), and scholars have called for greater attention to consumers who may exhibit passive or disengaged forms of engagement while remaining present within the broader ecosystem (McDonald et al., 2023; Ng et al., 2020). Examining lower levels of psychological ownership is necessary to understand withdrawal-based engagement dynamics in multi-brand contexts, given research lacks perspective on those who hold lower levels of psychological ownership, but still offer tangible benefits to brands (e.g. purchase; Kumar, 2019, 2022; Kumar and Nayak, 2019a).

Engagement and disengagement are not separate constructs; rather, they are connected (Saravade et al., 2025). Disengaged individuals usually have a negative disposition (Wang et al., 2023) that includes behaviours like detachment and neglect (Do and Bowden, 2025; Dolan et al., 2016). This has been previously conceptualised as “actor disengagement behaviours” which occur when an individual(s) lacks the capacity to balance their engagement with an ecosystem (Alexander et al., 2018; Katsifaraki and Theodosiou, 2020). If an individual perceives engagement demands (such as coordinating involvement across multiple brands) as irrelevant or burdensome, they may withdraw from interaction or reduce investment altogether (Li et al., 2017). If consumers likewise feel that an object does not belong to them, they are inclined to distance themselves from it (Kumar and Khandeparkar, 2025).

In multi-brand contexts, lower conjoined psychological ownership across brands may signal reduced category-level attachment, increasing the likelihood of apathy. Apathy, defined as diminished interest, emotion or concern towards a focal object (Marin, 1990), reflects motivational withdrawal rather than defensive activation. Unlike territoriality, which arises under perceived threat to strongly owned objects, apathy emerges when ownership levels are insufficient for brands to become self-relevant. As constraints such as time, money or knowledge accumulate, actors may further diminish their engagement intentions (Lee and Park, 2025). Accordingly, lower conjoined psychological ownership may increase apathy, which in turn reduces positively valenced engagement towards focal brands. In line with this, we hypothesise:

Figure 1 summarises the hypothesised relationships.

We employed a quantitative, cross-sectional survey to acquire data on individuals' engagement with brands within the Australian sporting market. With over 120 professional sporting brands despite a population of 28 million people, “no other nation can claim … to be home to as diverse a sport ecosystem” (Fujak, 2021, p. 9). We utilised the premier football competitions, the Australian Football League, A-League, National Rugby League and Super Rugby Pacific to contextualise the research. Participants were sourced via panel provider Qualtrics, with all statistical analyses carried out using SPSS Statistics v28. Upon being recruited to the survey and meeting the eligibility criteria (>18 years old, must support a team and viewed at least one match in any professional football league, thereby indicating some degree of familiarity with the research context), respondents were first instructed to select the football code they engaged with the most, which then displayed brands therein to select.

Psychological ownership has been shown to act as a predictor of engagement outcomes (Kumar and Nayak, 2019b). To acquire varying levels of psychological ownership felt towards both engagement objects without artificially prejudicing the sample, we ran a within-subjects experimental design in which respondents evaluated two distinct brands, allowing assessment of psychological ownership configurations across competing brands within the same category. To achieve this, a survey randomisation function for the team selection questions was created. When respondents were provided the opportunity to select brand x from their preferred football code, the question had a 50/50 chance of either displaying the instructions “Please select a team that you support” or “Please select a team that you DO NOT support”. Following their selection, respondents completed a set of questions relating to psychological ownership, positively and negatively valenced engagement, territoriality and apathy with the brand they selected prior to populating the research items. Upon completion of these questions, respondents were then redirected to select a second team (brand y; with the same 50/50 chance of either set of instructions being displayed) from their preferred competition, then repeating with the same items. Following data collection, 397 responses were harvested, reduced to 379 following data cleaning (i.e. straight-lining, response duration, attention checks, conflicting answers).

To capture the extent to which psychological ownership was simultaneously high (or low) across both focal brands, we computed a conjoined ownership index as the multiplicative product of ownership towards brand x and ownership towards brand y (POx * POy; Baron and Kenny, 1986). This operationalisation reflects simultaneous ownership intensity across both brands rather than a statistical moderation effect. Higher values indicate stronger concurrent ownership across both brands, whereas lower values reflect weaker ownership towards one or both brands (Ai and Norton, 2003). To further interpret conjoined ownership patterns, respondents were categorised into theoretically defined ownership configurations ([1] “x High/y High”; [2] “x High/y Low”; [3] “x Low/y Low”) based on median splits of psychological ownership towards each brand (Kinard and Capella, 2006). These sub-groupings were specified a priori to reflect distinct ownership configurations within a multi-brand engagement field and were used for descriptive and interpretive purposes alongside the primary test of the structural model performed on the full sample. For a full sample overview, see Web Appendix I (supplementary material).

To clarify the ownership configurations, the research design allows for two procedural pathways to the high/low ownership condition (i.e. selecting a high-psychological ownership brand followed by a low-psychological ownership brand, or vice versa). These pathways differ only in the order of elicitation rather than the underlying ownership structure. Conceptually, both produce the same configuration of a high psychological ownership focal brand paired with a low psychological ownership rival brand. To assess whether this procedural difference introduced bias, we compared these groups after aligning focal (high ownership) and rival (low ownership) objects. The results indicate no systematic differences in engagement patterns across the constructs of interest. Although some minimal variation is observed at the construct level, these differences are not consistent in pattern and do not alter the overall interpretation of the results. Accordingly, and consistent with our theoretical focus on ownership configuration rather than elicitation sequence, these pathways are treated as a single “High/Low” ownership condition.

To measure psychological ownership and territoriality, we adapted Avey et al.'s (2009) promotive and preventative psychological ownership scales, respectively, to suit the research context (replacing “organisation” with individuals' selected football brands). We used Dessart et al.'s (2016) consumer engagement scale to represent positively valenced engagement, whilst negatively valenced engagement items were adapted and formulated from Naumann et al. (2020) and Azer and Alexander's (2020) negatively valenced influencing behaviours, respectively. To measure apathy, we utilised the eight-item “apathy/indifference” scale from the Neuropsychiatric Inventory (Cummings et al., 1994), adapting the items so respondents answered on behalf of themselves, measuring apathy experienced towards their chosen brands. For full list of measures, see Web Appendix I (supplementary file).

Regarding instrument and statistical checks for assuring data reliability and validity (see Table 1), Harman's single-factor variance test provides an avenue to explore common method bias. The test checks all variables included and examines whether one factor accounts for more than 50% of the variance (Silber et al., 2024). The test generated a score of 26.99%, well below the 50% threshold for detecting common method bias. Performing a reliability check on all the measured constructs, the Cronbach's alpha (α) scores were all above the recommended 0.70 cut-off level (Schomburgk and Hoffmann, 2023). We additionally performed a heterotrait–monotrait (HTMT) test to assess discriminant validity (multicollinearity); all levels are below the accepted 0.900 level (Akman et al., 2019).

The primary mediation analyses were conducted on the full sample using the continuous conjoined ownership index. Analyses on the ownership sub-groups are presented alongside to illustrate how ownership configurations manifest within the engagement field.

Prior to conducting mediation analysis, we performed a series of independent samples t-tests to examine any statistical differences concerning the identified outcome variables across the sample. For reference, the means (M) and standard deviations (SD) of constructs for the full sample are contained in Table 2. Specifically, psychological ownership (t = 6.26, p < 0.001), positively valenced engagement (t = 5.25, p < 0.001), negatively valenced engagement (t = 2.33, p < 0.05) and territoriality (t = 4.13, p < 0.001) were all statistically higher with brand x, with only apathy (t = −3.33, p < 0.001) significantly greater for brand y.

We additionally performed a series of one-way analysis of variance tests (ANOVA) across the sub-groups. For reference, the means (M) and standard deviations (SD) of constructs for each ownership sub-group are contained in Table 2.

Specifically, significant differences (p < 0.001) between all three ownership sub-groups ([1] “x High/y High”; [2] “x High/y Low” and [3] “x Low/y Low”) were yielded for psychological ownership felt towards brand x (F (2, 150.70) = 199.04) and brand y (F (2, 151.70) = 260.49, p < 0.001); positively valenced engagement towards brand x (F (2, 135.18) = 106.51) and brand y (F (2, 134.08) = 134.08, p < 0.001); territoriality felt over brand x (F (2, 125.69) = 83.83) and brand y (F (2, 122.40) = 70.91, p < 0.001); and apathy felt over brand x (F (2, 160.85) = 45.52) and brand y (F (2, 158.03) = 65.38, p < 0.001). No significant differences were uncovered across all three sub-groups regarding negatively valenced engagement towards brand x (F (2, 124.63) = 0.96, p = 0.39) nor brand y (F (2, 127.62) = 204.12, p = 0.386).

To explore the strength and valence of relationships between psychological ownership and positively/negatively valenced engagement, PROCESS by Hayes (2012) was implemented to test the hypotheses. To capture the extent to which psychological ownership was simultaneously high (or low) across both focal brands, we computed a conjoined ownership index as the multiplicative product of ownership towards brand x and ownership towards brand y (calculated in SPSS as [Brand x PO * Brand y PO]; Baron and Kenny, 1986). This operationalisation reflects simultaneous ownership intensity rather than a statistical moderation effect. We considered its effect both directly influences the outcome variables, and indirectly through the hypothesised mediators, allowing estimation of the direct and indirect associations between conjoined psychological ownership and the proposed outcome variables (Hayes, 2012). Performing Hayes (2012) PROCESS modelling with 5,000 bootstrap samples, the analysis is outlined as follows: first, we discuss the results for the full sample (see Table 3 for full dissemination of results), then discuss the results within the ownership sub-groups (see Table 4 for full sub-group results), concluding each hypothesis with implications.

H1a predicted the conjoined psychological ownership felt between brand x and brand y (labelled “xy PO”) would positively influence positively valenced engagement with brand x. The results indicated that xy PO significantly predicted x positive engagement (β = 0.111; CI: 0.094, 0.128). H1b predicted that xy PO would positively influence positively valenced engagement with brand y. Examining the analysis, xy PO significantly predicted y positive engagement (β = 0.124; CI: 0.109, 0.140).

Next, examining the ownership sub-groups, the results for H1a indicated that xy PO significantly predicted x positive engagement ([1]”x High/y High”: β = 0.096; CI: 0.059, 0.134; [2] “x High/y Low”: β = 0.062; CI: 0.015, 0.110; [3] “x Low/x Low”: β = 0.191; CI: 0.137, 0.246). Examining the analysis for H1b, xy PO significantly predicted y positive engagement across the “x High/y High” group (β = 0.092; CI: 0.056, 0.128) and “x Low/y Low” group (β = 0.196; CI: 0.153, 0.239). There was no relationship uncovered for the “x High/y Low” group (β = 0.008; CI: −0.027, 0.043). Therefore, H1 is supported.

These results support and build on prior work that demonstrates psychological ownership influences positively valenced engagement (Kumar and Nayak, 2019b) by introducing the novel finding that psychological ownership between multiple brands can also enhance positively valenced engagement.

H2a predicted that xy PO would negatively influence negatively valenced engagement towards brand x. The results did not yield a significant relationship between xy PO and x negative engagement (β = −0.007; CI: −0.022, 0.008). H2b predicted that xy PO would negatively influence negative engagement with brand y. The results likewise did not yield a significant relationship between xy PO and y negative engagement (β = 0.006; CI: −0.009, 0.020).

Examining the ownership sub-groups, for H2a the results did not yield a significant relationship between xy PO and x negative engagement across all three sub-groups ([1] “x High/y High”: β = −0.010; CI: −0.057, 0.037; [2] “x High/y Low”: β = 0.001; CI: −0.027, 0.043; [3] “x Low/y Low”: β = −0.052; CI: −0.121, 0.017). For H2b, the results likewise did not yield a significant relationship between xy PO and y negative engagement ([1] “x High/y High”: β = −0.022; CI: −0.067, 0.023; [2] “x High/y Low”: β = −0.020; CI: −0.056, 0.016; [3] “x Low/y Low”: β = −0.033; CI: −0.097, 0.032). Therefore, H2 is not supported.

No evidence was found that conjoined ownership was associated with negatively valenced engagement. The lack of support for H2 indicates that, despite the research context being typically viewed as “us against them” (Yun et al., 2021), conjoined psychological ownership across multiple brands does not appear to predict negatively valenced engagement towards either brand directly.

H3a predicted that territoriality felt over brand x would function as an indirect pathway linking xy PO with negatively valenced engagement towards brand y. Examining the mediation path, territoriality was found to partially mediate xy PO and brand y negative engagement (indirect effect: β = 0.026; CI: 0.015, 0.037). H3b hypothesised that territoriality felt over brand x would positively mediate the relationship of xy PO influencing positively valenced engagement with brand x. Examining the mediation path, territoriality was found to partially mediate this relationship (indirect effect: β = 0.084; CI: 0.073, 0.098).

Examining the sub-groups, regarding H3a, only for the “x High/y Low” group was territoriality found to partially mediate xy PO and brand y negative engagement (indirect effect: β = 0.020; CI: 0.006, 0.036). This did not generate a statistically significant indirect effect for the “x High/y High” (indirect effect: β = −0.013; CI: −0.009, 0.043) and “x Low/y Low groups” (indirect effect: β = −0.003; CI: −0.052, 0.040). Concerning H3b, territoriality was found to fully mediate within the “x High/y Low” group (indirect effect: β = 0.055; CI: 0.019, 0.093) and partially mediate for the “x Low/y Low” group (indirect effect: β = 0.111; CI: 0.063, 0.167). No significant mediation was unearthed for the “x High/x High” group (indirect effect: β = 0.034; CI: −0.005, 0.073). Therefore, H3 is partially supported.

These findings support prior work around how consumers can engage in adverse behaviours when protective feelings over psychologically owned objects arise (Micu and Ashley, 2022). Within the ownership configurations, only the “x High/y Low” sub-group had a significant indirect effect upon negatively valenced engagement (H3a). These results further uncover a corresponding simultaneous “response” to territorial infringement via positively valenced engagement (H3b). Interestingly, the “x Low/y Low” sub-group experienced a significant sense of territoriality that influenced positively valenced engagement (H3b), nuancing with prior work that shows that low forms of psychological ownership create passive engagement states (Harmeling et al., 2017).

H4a predicted that apathy felt over brand y would negatively mediate the relationship between xy PO and team x positively valenced engagement. Examining the mediation path, apathy partially mediated xy PO and brand x positive engagement (indirect effect: β = −0.043; CI: −0.056, −0.033). H4b predicted that apathy felt over brand x would negatively mediate the relationship between xy PO and team y positively valenced engagement. Apathy was found to partially mediate xy PO and brand x positive engagement (indirect Effect: β = −0.038, CI: −0.048, −0.029).

Examining the ownership sub-groups, for H4a, apathy partially mediated xy PO and brand x positive engagement within the “x High/y Low” (indirect effect: β = −0.074; CI: −0.105, −0.042), and “x Low/x Low” groups (indirect effect: β = −0.033; CI: −0.064, −0.009). No significant mediation was yielded for the “x High/y High” group (indirect effect: β = 0.001; CI: −0.009, 0.016). Regarding H4b, apathy partially mediated xy PO and brand x positive engagement across all three groups ([1] “x High/y High”; Indirect Effect: β = −0.016, CI: −0.031, −0.005; [2] “x High/y Low”; Indirect Effect: β = −0.035, CI: −0.071, −0.002; [3] “x Low/y Low”; Indirect Effect: β = −0.034, CI: −0.059, −0.016). Therefore, H4 is partially supported.

These results nuance prior work suggesting a lack of psychological ownership inhibits engagement (Karahanna et al., 2015; Reppmann et al., 2025). The findings suggest that apathy had a negative relationship with positively valenced engagement; that is, the less apathy exhibited, the stronger influence on positively valenced engagement, remarkably even within the “x Low/y Low” ownership sub-group. This lends support to Kamleitner and Feuchtl's (2015) proposition that psychological ownership matters even when at low thresholds.

The purpose of this research was to examine how a focal actor's psychological ownership felt towards multiple brand influences nuanced engagement outcomes towards associated firms, and through what mechanisms (i.e. territoriality, apathy) aid or inhibit this process. The research demonstrates that psychological ownership can act as a fruitful mechanism for brand growth, amplifying positively valenced engagement (even for “lurkers” within a service ecosystem) whilst having no identified influence on negatively valenced engagement. Mindfulness is required regarding territorial infringement, which can generate negatively valenced engagement towards opposing brands. This advances extant understanding about psychological ownership by coupling exploration with the (multi) actor engagement literature. This helps explain the complex dynamics of concurrent multi-brand engagement in contemporary markets where consumers regularly and simultaneously interact with multiple competing brands. A summary of the research and results is provided in Table 5.

In response to the research question at hand, our theoretical contributions of this study are as follows: Firstly, the empirical analysis reveals significant differences in how consumers engage with multiple brands simultaneously. Notably, a focal actor's conjoined psychological ownership between multiple brands does not directly result in negatively valenced engagement towards either firm, highlighting this as a fruitful mechanism for brands to pursue as a means of increasing engagement among their consumer bases. Whilst psychological ownership has been increasingly gaining attention in the managerial and marketing literature (Baker et al., 2021), its application is limited to dyadic interactions between consumers and single brands (Kumar and Nayak, 2019a), with more research needed into how consumers navigate multi-foci engagement scenarios (Kim et al., 2024). This study shows conjoined psychological ownership has a positive impact upon engagement (Jussila et al., 2015) and contributes to engagement in configuration-dependent ways. High ownership towards a focal brand produces different engagement outcomes depending on whether ownership towards competing brands is simultaneously high or low (Kim et al., 2024).

Furthermore, this study builds on previous research in marketing by extending psychological ownership to understand negatively valenced and disengaged engagement outcomes (Baker et al., 2021). Prior work on psychological ownership has primarily examined positive manifestations, overlooking potentially undesirable negatively valenced engagement responses that cause serious harm for brands (Baker et al., 2021; Kim et al., 2024; Koklic et al., 2025). The introduction of territoriality as a mediator reveals a critical relationship between psychological ownership and engagement outcomes. While psychological ownership initially showed no direct relationship with negatively valenced engagement, for fans who hold higher levels of psychological ownership towards one object, yet lower levels to another, the presence of territoriality enables psychological ownership to generate significant unfavourable engagement towards competing brand(s). Consumer evaluations and subsequent behaviours can shift when competing options with distinctive criteria are presented jointly versus separately (Hsee, 1996). These empirical results demonstrate that within a multi-actor setting, negatively valenced engagement outcomes can be triggered from (perceived) territorial infringement by another brand.

Research on territorial expressions has focussed almost entirely on detrimental outcomes for the self and others (Kirk et al., 2018). We provide empirical evidence of territoriality generating positive outcomes for high psychological ownership brands whereby a contrasting low psychological ownership oppositional brand is co-presented that are of benefit for brand managers to consider (Kuchmaner et al., 2025). Importantly, the finding of a significant indirect effect only suggests a more complex relationship, where territorial responses can simultaneously generate both positively and negatively valenced engagement outcomes (Naumann et al., 2020). This demonstrates that engagement intensity does not follow a single linear pathway but instead emerges from distinct ownership constellations. This nuanced view challenges the traditional dichotomy of positively versus negatively valenced engagement, suggesting a more integrated perspective that provides insight helping brands anticipate and manage consumer reactions in competitive environments.

Finally, we conduct empirical research into passive, low psychological ownership consumers. The literature on “lurkers” is nuanced; some scholars have demonstrated that low psychological ownership imposes resource constraints on consumers (Lee and Park, 2025), creating a barrier for further interactions (Reppmann et al., 2025) and are thus less likely to engage (Karahanna et al., 2015). Other scholars have noted that psychological ownership still matters even at low thresholds (Kamleitner and Feuchtl, 2015), such as brand purchasing (Kumar, 2022). Our results highlight that despite possessing low levels of psychological ownership towards multiple engagement objects, consumers' apathy in fact has a negative influence on engagement; low psychological ownership still produces a significant influence on positively valenced engagement and remarkably can still induce territorial responses preceding favourable engagement. This supports Kumar's (2019) assertion that although lurkers' engagement may not be visible, it is still important to examine and consider for scholars and managers alike. In response to calls to examine consumers with low psychological ownership and their subsequent engagement for associated brands (Kuchmaner et al., 2025; Kumar, 2022; Kumar and Nayak, 2019a), our finding further reinforces that engagement valence and intensity are shaped by ownership configurations across brands rather than isolated psychological states.

Rather than modelling engagement as a function of isolated antecedents, we demonstrate that patterned combinations of ownership across brands structure both the direction and intensity of engagement resource allocation. This has some meaningful implications for managers.

We encourage managers to induce a sense of psychological ownership over their firm(s) or products to enhance customers' engagement dispositions. Prior work has demonstrated that psychological ownership can favourably influence engagement outcomes (Kumar and Nayak, 2019a); our results show that the conjoined ownership felt towards competitors as well can further enhance consumers' positive resource contributions towards the firm's offering(s). Managers can accomplish building a sense of psychological ownership over their brand(s) by emphasising the routes to ownership (self-identity, belonging, self-efficacy, accountability) in their promotion and fan activation strategies, such as increased opportunities to participate/contribute, and referring as “your” or “our” club/brand (Palau-Saumell et al., 2022) to empower consumers to defend their firm against competition and perceived territorial infringements (Peck et al., 2020).

Firms operating within high-involvement service contexts, such as sport, have long been considered to maintain exclusive loyalty with their customers (Funk and James, 2001), where individuals engage with one firm sequentially (i.e. a brand; Hollebeek et al., 2022). Prior work has shown that fans can share their engagement with multiple teams across leagues (McDonald et al., 2010); our research finds a novel phenomenon of consumers exhibiting favourable attitudes and behaviours towards more than a single brand in the same sporting league. The analysis generated a distinct sub-group of consumers who hold higher degrees of psychological ownership towards two brands. Only when territorial instincts activate are detrimental responses observed when infringement upon object(s) whereby psychological ownership exists is perceived. In categories with high brand competitor salience, consumers acquire a comparative point to evaluate their purchased brand(s) in relation to other brands in their repertoire (Sharp and Romaniuk, 2010). We advise managers, particularly those operating within high involvement categories such as sport or luxury brands, to practice mindfulness regarding viewing their fans as exclusively their customers; their fans hold complex, multi-dimensional relationships with other firms or exogenous touchpoints outside of the brand's control (Jaakkola and Alexander, 2024).

Managers should further consider updating their customer relationship management systems to capture richer marketing analytics on consumers' psychological ownership, not only felt over their brand but also towards competitors. Our research identifies multi-brand psychological ownership as a fruitful mechanism for brands to pursue commercial growth, with no direct links predicting negatively valenced engagement uncovered. We acknowledge difficulties for practitioners to adopt large multi-item scales, such as Avey et al.'s (2009) 16-item psychological ownership scale, in existing fan analysis like end-of-season surveys or segmentation analysis (Kunkel et al., 2022). This research nonetheless highlights the opportunity for managers to capture data (such as on a single 7-point Likert scale) assessing an individual's team preference across all brands in the category. Such is an uncommon practice in sport, and can be utilised to identify high-potential consumers that are more likely to respond favourably to targeted messaging and promotions (i.e. exclusive deals, such as priority or discounted ticketing) that present opportunities to simultaneously engage with multiple high psychologically ownership brands (e.g. a sporting match between the teams). This provides a valuable tool to heighten engagement (e.g. word-of-mouth, purchase likelihood, attendance) that is otherwise challenging through traditional, dyadic relationships.

The research showed that territorial instincts may activate detrimental responses when infringement upon object(s), whereby psychological ownership exists, is perceived by individuals. Whilst the sub-groups identified provide a profile for managers to consult and tailor specific messaging alleviating point of infringement, we also unearthed a link between territoriality and positively valenced engagement, whereby (perceived) infringement reinforces positive responses towards the receiving brand. As such, managers may utilise territorial responses in an ethical manner (e.g. light banter, humour to opposing brands via social media) to stimulate engagement and attention with their firm (e.g. “fanning the flames”; Sewak et al., 2025).

Finally, limited work has looked at non-fans in sport (McDonald et al., 2023). The results unearth a considerable sub-group of consumers based on psychological ownership that, whilst appearing to be passive or even disengaged, still show signs of engagement and may still be lurking in the background within a customer base. Prior research has demonstrated that factors such as restrictions (geographic, temporal and financial; Collins et al., 2016) impede people's engagement despite their attention. This builds on previous work guiding managers, by providing further descriptions of this profile of consumers who typically are overlooked or excluded. We suggest managers may appeal to this segment of casual fans with content and access to connect with other fans via non- and limited-access packages that offer flexibility via traditional mediums (e.g. offering physical ticketing over digital forms), thereby providing touchpoints with the brand/other fans, and content to build a sense of psychological ownership and induce more favourable engagement outcomes.

While this study contributes to the understanding of psychological ownership between multiple engagement objects and its impact on engagement valence and intensity outcomes, it is not without limitations which provide fertile ground for future research.

Our sample comprised fewer respondents who held high levels of psychological ownership towards both teams compared to the “x High/y Low” and “x Low/y Low” sub-groups. This may be due to the pertinent research context of sport, where the likelihood of people following two sports firms to an equally strong extent is lower. As such, the theorised “amplification of positively valenced engagement” may be prevalent in greater amounts across differing product/service categories or even engagement aggregations (e.g. local versus national contexts; Clarke et al., 2022). As such, we encourage future research to examine other service contexts in which individuals' roles may not be so constricting on having multiple positively valenced engagement objects.

In several cases, significant indirect effects via territoriality and apathy were observed, with the direct association between conjoined psychological ownership and engagement outcomes remaining significant. This pattern suggests that the proposed mechanisms account for part, but not all, of the association. Therein exists some other mechanism whose unidentified presence suggests an alternative, positive and significant mediator not explained in the analysis. Prior research has conceptualised consumers engage with parallel contexts; holding membership duality can generate ambivalence between the two; accepted ways for one to conduct themselves (i.e. roles) may conflict, which have been conceptualised to trigger “Actor Disengagement Behaviours” for actors to temporarily or permanently halt resource contributions from heightened role conflict or strain (Alexander et al., 2018) instance, Micu and Ashley (2022) found abandonment as a territorial response in face of increased (perceived) threat. This may indicate that consumers may seek to disengage, rather than continue in the relationship (negatively) engaging. Future work should empirically consider the strain on an actor's resources, and subsequent engagement outcomes.

We only examined psychological ownership at the individual level. Collective engagement, referring to when a group senses an object as “OURS” (Baker et al., 2021), is an emerging field that would be of interest to examine how psychological ownership over brands manifests when it is shared, particularly in contexts where clear in-groups and out-groups are formed (i.e. “this is OUR team; NOT yours”; Pierce et al., 2025).

Finally, we did not account for nested brands; brands that operate as sub-units or alongside parent brands (for example, many sports teams are introducing women's teams alongside their men's teams for fans to co-consume). Potential “cross-pollination” transfer effects between these brands remain an exciting and emerging area of research (Clarke et al., 2022; Doyle et al., 2024); understanding how best to support women's sports teams alongside male counterparts (Bramley et al., 2025).

The authors would like to thank Dr Alex Belli (The University of Melbourne) for his time, guidance and analytics materials provided during the data analysis period.

This research has been approved by a Huma Research Ethics committee (H-2023-094).

The supplementary material for this article can be found online.