Mitigating the risk that peer-initiated task conflict escalates into diminished helping: roles of passion for work and collectivistic orientation

Abstract This study unravels the link between employees’ exposure to peer-initiated task conflict—defined as the extent to which they perceive that coworkers systematically contest and attack their opinions—and their engagement in helping behavior. Beliefs about interpersonal conflict might mediate this link, and two personal resources, passion for work and collectivistic orientation, arguably have moderating roles. To test these predictions, this study relies on survey data from employees who work in the banking sector, which confirm that peer-initiated task conflict diminishes helping behavior, because the focal employees come to believe coworkers are responsible for their emotion-based quarrels. Employees’ passion for work and collectivistic orientation buffer this harmful dynamic. Organizations thus should recognize that exposure to overcritical colleagues can undermine voluntary work behaviors, as well as consider how they might help reduce the force of this negative dynamic by enabling employees to find ways to draw from their supportive personal resources.


Introduction
A helpful colleague is invaluable; when employees devote extra effort voluntarily to helping their colleagues complete their tasks, it leads to beneficial outcomes all around. Dedicated staff members, or helpers, who undertake peer-oriented citizenship behavior enhance team and organizational effectiveness (Bachrach et al. 2006;Borman and Motowidlo 1993;Podsakoff, Morrison, and Martinez 2018;Reisel et al. 2010), and these employees likely prompt favorable views among their peers and their supervisors (Hui, Lam, and Law 2000;Korsgaard et al. 2010), as well as a sense of personal fulfillment for themselves (Hoption 2016;Lemoine, Parsons, and Kansara 2015). Yet extra-role work activities also might be disadvantageous, if generous offers of assistance prevent the helpers from completing their own formal job duties (Bergeron 2007;Koopman, Lanaj, and Scott 2016). Some recipients of aid also might read offers of assistance as a form of condescension or signal of the helpers' belief they cannot do the work on their own (Organ 1988;Podsakoff, Morrison, and Martinez 2018).
Considering these contradictory outcome possibilities, determining the factors that lead employees to either embrace or avoid this form of work-related voluntarism remains a pertinent research goal (De Clercq, Rahman, and Haq 2019;Zhao and Guo 2019). For example, employees are less likely to go out of their way to contribute to the well-being of their peers or their employer in general if they suffer from coworker incivility (Liu, Zhou, and Che 2019), ostracism (Peng and Zeng 2017), or bullying (Devonish 2013). Such negative relationship factors all create a sense that employees are not respected or taken seriously by peers (Podsakoff, Morrison, and Martinez 2018). The central premise of this study is that when coworkers systematically initiate content-based disputes or contest offered opinions-which we label peer-initiated task conflictemployees experience an additional source of resource depletion that also may steer them away from discretionary helping activities (Jungst and Janssens 2020). We know of only a single study that identifies a negative relationship between task conflict and helping behavior (Ng and Van Dyne 2005); otherwise, our understanding of how this harmful dynamic unfolds is minimal, which means that organizational decision makers lack comprehensive insights into why or when hardships due to peer-initiated idea clashes may cause employees to halt extra-role work behaviors that have the potential to benefit them and their organization (Hui, Lam, and Law 2000;Podsakoff et al. 2009).
With this research, we aim to detail relevant factors that underpin or impact this translation of peer-initiated task conflict into diminished helping behavior. First, we postulate that a critical conduit through which this translation may materialize is that employees develop beliefs that their coworkers introduce emotion-based tensions into their relationships (Pooja, De Clercq, and Belausteguigoitia 2016). These beliefs about interpersonal conflict can manifest in different ways, such as a conviction that coworkers are responsible for their discomfort, annoyance, or even anger in daily interactions (Jehn and Mannix 2001;Kisamore et al. 2014). Consistent with the logic underlying conservation of resources (COR) theory (Hobfoll 2001;Hobfoll et al. 2018), the perception that peers persistently express viewpoints in opposition to their own may lead employees to stop helping those peers, because they hold them accountable for emotional hardships in their interpersonal relationships, which gives the employees a way to shield their own self-esteem resources (De Dreu and Weingart 2003;Pierce and Gardner 2004). Second, and also in line with COR theory, we predict that this detrimental process is less likely if the employees possess two relevant personal resources, passion for work and collectivistic orientation, that mitigate self-depreciating ruminations about peer-initiated idea clashes (Meier et al. 2013)-and thus diminish frustrations with interpersonal conflict and a subsequent reluctance to perform discretionary helping behaviors. Employees' passion for work reflects the excitement that they experience from working hard (Baum and Locke 2004), and their collectivistic orientation speaks to excitement derived from maintaining group harmony (H. Wang, Lu, et al. 2017).
With these considerations, we seek to provide several contributions to management research. First, we theorize how peer-initiated task conflict, a threat to employees' sense of self-worth (De Dreu and Weingart 2003), may reduce the likelihood that they work to advance and ensure their peers' success, as informed by their beliefs that the coworkers are culprits in their interpersonal quarrels (Jehn and Mannix 2001). Systematic attacks on their opinions make it seem as if they can never get things right, according to their colleagues, so employees may suffer emotion-based hardships that ultimately lead them to stop helping their colleagues. Previous research acknowledges a positive link between task conflict and interpersonal conflict (e.g., Pluut and Curşeu 2013;Simons and Peterson 2000), but we go further to consider how this link may hamper workrelated voluntarism. In so doing, we pinpoint a possible downward spiral for employees: Their peer interactions are marked by continuous idea clashes, leading to negative beliefs that escalate into complacency toward coworkers' success, which might prompt the coworkers to become even more contentious toward them.
Second, contingency perspectives on the detrimental consequences of content-based conflict are both necessary and informative (Bai, Han, and Harms 2016;Boros , 2020). With our application of such a perspective, we clarify how the likelihood of helping behavior, as a response to peer-initiated task conflict and associated beliefs about interpersonal conflict, is lower among employees equipped with valuable personal resources (Hobfoll et al. 2018). As noted by previous studies, the harmful outcomes of task conflict are not automatic-and some of this conflict type even can generate positive outcomes in the form of creativity (De Clercq, Rahman, and Belausteguigoitia 2017;Jehn 1995)-because certain factors can help employees cope with the corresponding challenges, such as their emotion regulation skills (Boros , 2020), emotional intelligence (Lee and Wong 2019), or trust in others (Choi and Cho 2011). We add to this research stream by focusing on two distinct sources of positive work energy that also may buffer the negative threats of task conflict: (1) how employees experience their hard work in and of itself (passion for work) and (2) whether they thrive when collaborating with others (collectivistic orientation). The focus on these two personal resources is purposeful, in that the resources reflect consistent and complementary features that may protect employees against a situation in which peer-initiated idea clashes cause them to stop offering voluntary help to others and focus only on their formally prescribed job tasks. Finally, the proposed moderated mediation dynamic combines a mediating role of interpersonal conflict beliefs with moderating roles of two personal resources. It thus adds conceptual and methodological rigor to efforts to establish a comprehensive view of key contingencies that can explain employee behaviors (Haq, De Clercq, and Azeem 2022;Yang and Yang 2020).

Theoretical background and research hypotheses
The conceptual arguments for the proposed mediating effect of interpersonal conflict and moderating effects of personal resources reflect insights from conservation of resources (COR) theory. This theory predicts that employees' work-related thoughts and actions are shaped, in significant ways, by their desire to preserve their current resource reservoirs and avoid further losses when they confront resource-depleting work circumstances (Hobfoll and Shirom 2000). It offers two critical propositions: Resource drainage due to upsetting work situations directs employees to embrace beliefs and behaviors that enable them to undo such drainage, and access to valuable personal resources can subdue the process, especially if those resources render it less likely that their suffering leads to an actual depletion of their resource bases (Abbas et al. 2014;De Clercq, Rahman, and Haq 2019).
The notion of resources in COR theory is broad, such that they entail "those objects, personal characteristics, conditions, or energies that are valued in their own right, or that are valued because they act as conduits to the achievement or protection of valued resources" (Hobfoll 2001, p. 339). In particular, and relevant to our research objectives, employees' self-esteem resources strongly determine the quality of their professional functioning. In his foundational work, Hobfoll (2001) specifies these resources as essential features, and he maintains that employees are especially driven to protect them in the workplace. Yet as conflict research establishes, "being in conflict threatens one's self-esteem and requires cognitive resources to cope with the situation" (De Dreu and Weingart 2003, p. 156). If employees sense that their coworkers systematically contest their opinions and do not take them seriously, employees' sense of self-worth comes under great threat (de Wit, Greer, and Jehn 2012;Jungst and Blumberg 2016). To escape such a situation, they might lay the blame for their poor interpersonal interactions on the coworkers who start the conflict (Simons and Peterson 2000). That is, in line with COR theory (Hobfoll et al. 2018), employees' perceptions of interpersonal conflict caused by coworkers and their associated hesitance to help those colleagues on a voluntary basis represent likely responses to peer-initiated task conflict, as means to reduce their own self-depreciating ruminations. As coping strategies, such responses allow them to express their sense of offense, sparked by colleagues who never seem to agree with their viewpoints (Jungst and Janssens 2020).
As a further application of COR theory, the likelihood that employees engage in such coping responses should be lower when they can rely on personal resources that make the responses less necessary (Hobfoll et al. 2018). In particular, we propose that even if their opinions are constantly challenged by colleagues, employees should be less likely to unleash their frustrations by calling out the interpersonal conflict if certain personal resources keep them excited about their work functioning. As noted by prior research, employees' positive work-related energy, derived from their passion (De Clercq and Belausteguigoitia 2019) or collaborative attitudes (Du et al. 2015), can help them deal with resource-draining work situations. We similarly argue that employees' passion for work and collectivistic orientation may protect them against the self-damaging thoughts that come with peer-initiated task conflict (De Dreu and Weingart 2003), such that they become less pessimistic about the quality of their interpersonal interactions (Pluut and Curşeu 2013) and ultimately more willing to continue to engage in some work-related voluntarism to help coworkers (Kisamore et al. 2014). These two personal resources diminish the risk that perceptions of peer-initiated task conflict escalate into complacency toward work activities that can address the sources of the experienced hardships.
The study's conceptual framework is in Figure 1, depicting the proposed mediating role of beliefs about interpersonal conflict and moderating roles of passion for work and collectivistic orientation. We explicitly note that the arguments underpinning the mediation link predict causality among the constructs that constitute this link (i.e., from peer-initiated task conflict, to beliefs about interpersonal conflict, to helping behavior), but the cross-sectional research design, as detailed in the Method section, does not allow us to test for the presence of such causality. The hypotheses stated hereafter thus predict relationships among constructs, rather than the effects of the constructs on one another. We detail the rationales for each of the hypotheses next.

Mediating effect of interpersonal conflict
We postulate that employees' exposure to peer-initiated task conflict relates positively to their beliefs about interpersonal conflict. In line with COR theory (Hobfoll and Shirom 2000), the frustrations initiated by coworkers who contest seemingly every viewpoint might undermine employees' sense of self-worth so much that they search for a scapegoat and assert that the coworkers cause emotion-laden quarrels (Boros , 2020;de Wit, Greer, and Jehn 2012). For example, systematic disagreements about task-related topics may prompt employees to develop self-damaging thoughts about their ability to make positive work contributions (De Dreu and Weingart 2003),  to which they respond by accusing the disagreeable colleagues of creating a dysfunctional work environment, filled with negative energy and tension (Boros , 2020). These accusations serve as coping tactics through which employees express their irritations and safeguard their self-esteem resources (Hobfoll et al. 2018). Their exposure to peer-initiated task conflict thus becomes less troubling, because they can shift responsibility for their negative feelings from themselves to their colleagues (Simons and Peterson 2000). In a related way, employees who feel attacked by opposing viewpoints expressed by their colleagues may interpret the unfavorable treatment as a sign of limited recognition for their contributions or daily work efforts (Avgar, Lee, and Chung 2014;Liao and Sun 2020). In the presence of such signals, accusations that colleagues are the cause of dysfunctional relationships may seem justified (Bowling et al. 2010). If they cannot express their own opinions without worrying about having their ideas systematically contested, employees also might realize that their coworkers apparently are trying to make them look incompetent, especially to organizational authorities (de Wit, Greer, and Jehn 2012;Jungst and Blumberg 2016). Faced with such evidence, employees might form beliefs about the interpersonal adversity that their coworkers create (Bai, Han, and Harms 2016). In contrast, if coworkers respect their opinions, employees should hold more positive views of the quality of their work relationships and their coworkers' contributions to them (De Dreu and Weingart 2003), such that they have no reason to accuse them of causing any negative interpersonal dynamics (Hobfoll et al. 2018;Pluut and Curşeu 2013). It is less likely that they suffer from a poor self-image in this scenario (Pierce and Gardner 2004), with beneficial consequences for their beliefs about interpersonal conflict. We accordingly predict: Hypothesis 1: There is a positive relationship between employees' exposure to peer-initiated task conflict and their beliefs about interpersonal conflict.
Employees who are convinced that their coworkers are responsible for emotion-based quarrels also may be reluctant to go out of their way to help the coworkers, particularly in ways that are not explicitly required in their job descriptions. Consistent with COR theory, by halting extrarole work activities, employees can safeguard their self-esteem resources and constrain their suffering from interpersonal conflict (De Dreu and Weingart 2003;Hobfoll et al. 2018). That is, the realization of poor peer relationships may escalate into a diminished motivation to assist coworkers, because it offers a way to vent dismay with the experience of emotion-based tensions (Kisamore et al. 2014;Pooja, De Clercq, and Belausteguigoitia 2016). Further in line with the COR logic, halting extra-role helping activities, in response to beliefs about interpersonal conflict, may create resource gains for employees, related to their personal fulfillment (Hobfoll and Shirom 2000). In particular, to the extent that their coworkers seemingly do not care about maintaining high-quality interpersonal relationships, employees may experience a sort of guilty pleasure in rejecting their requests for help (Lu, Zhou, and Leung 2011).
Similarly, employees who are convinced that their colleagues are responsible for their emotionbased hardships likely have limited motivation to contribute to the work success of those colleagues by offering discretionary helping activities (Lau and Cobb 2010;Podsakoff, Morrison, and Martinez 2018). Previous studies reveal that when employees suffer from interpersonal conflict, it generates negative work-related sentiments in them, such as diminished job satisfaction (Briggs, Jaramillo, and Noboa 2015) or commitment (Pooja, De Clercq, and Belausteguigoitia 2016). These sentiments likely make employees indifferent about whether their colleagues' performance would be compromised in the absence of their voluntary help (Pooja, De Clercq, and Belausteguigoitia 2016;Saks 2019). Conversely, employees who hold positive beliefs about the quality of their peer relationships should be keen to leverage these beliefs in supporting their voluntary work activities, so that they can obtain even more resource gains (Hobfoll et al. 2018;Hui, Lam, and Law 2000). One particular path though which this positive outcome can be realized is by being a "good organizational soldier," which tends to generate favorable leader evaluations (Podsakoff, Morrison, and Martinez 2018). Ceteris paribus, we predict: Hypothesis 2: There is a negative relationship between employees' beliefs about interpersonal conflict and their helping behavior.
In combination, these arguments imply a critical mediating role of employees' beliefs about interpersonal conflict. Their perception that coworkers initiate idea clashes increases the chances that employees halt their helping activities, informed by their allegations that the coworkers are responsible for the negative energy permeating their interpersonal interactions (Bai, Han, and Harms 2016;Simons and Peterson 2000). To the extent that they are the victims of persistent idea clashes, the chances that employees allocate personal time to extra-role work activities are slim, due to their convictions about impoverished relationship dynamics and the associated desire to protect their sense of self-worth (Bowling et al. 2010;De Dreu and Weingart 2003). Similarly, interpersonal conflict mediates the escalation of other resource-draining work conditions, such as limited interactional justice (Bouckenooghe, De Clercq, and Deprez 2014) or workplace bullying (Parach and Shahzad 2017), into negative work outcomes. We complement such research by postulating: Hypothesis 3: Employees' beliefs about interpersonal conflict mediate the relationship between their exposure to peer-initiated task conflict and their helping behavior.

Buffering effect of passion for work
According to COR theory, the resource-draining effect of unfavorable work conditions is mitigated to the extent that employees can counter the resource drainage with valuable personal resources (Abbas et al. 2014;Hobfoll and Shirom 2000). We similarly propose that the probability that peer-initiated task conflict translates into beliefs about interpersonal conflict is lower among employees who also are passionate about their work (Baum and Locke 2004). The positive work energy stemming from such passion enhances employees' inclinations to see task-related disputes as challenges, instead of threats to their self-image (De Clercq and Belausteguigoitia 2017), so it becomes less likely that they adopt tactics to protect their self-esteem resources (De Dreu and Weingart 2003). Similarly, passionate employees often search proactively for solutions to upsetting work situations, which reduces the need to look for culprits for these situations (Gulyani and Bhatnagar 2017). They even could be attracted to workplace challenges, in that finding ways to thrive in the face of challenges generates a sense of personal accomplishment that resonates with their passion for work (Houlfort et al. 2015;Tr epanier et al. 2014). From this angle, employees' passion for work may generate resource gains, in the form of personal satisfaction, if they can maintain the quality of their relationships even with overcritical coworkers (Ryan and Deci 2000). The desire to preserve their sense of self-worth by forming beliefs that coworkers cause emotionbased conflicts thus is lower in this scenario.
This predicted buffering role of passion for work, in combination with the mediating role of interpersonal conflict, points to a moderated mediation dynamic (Hayes, Montoya, and Rockwood 2017). Passion for work constitutes a valuable personal contingency that influences the indirect negative link between suffering from peer-initiated task conflict and helping behavior, through beliefs about relation-based quarrels. Among employees who can rely on unbridled work excitement (Baum and Locke 2004), criticisms that coworkers are responsible for negative interpersonal interactions should be less powerful in terms of channeling their exposure to conflict-seeking colleagues into a refusal to engage in voluntary activities to aid these colleagues. This personal resource mitigates self-depreciating thoughts that disagreeable colleagues threaten to spark (De Clercq and Belausteguigoitia 2017), so it is less likely that employees halt their work-related voluntarism (Kisamore et al. 2014). In contrast, employees with limited passion for work may rely on their beliefs about impoverished interpersonal relationships as more salient explanations for why peer-initiated task conflict justifies their tarnished helping activities.
Hypothesis 4: The indirect negative relationship between employees' exposure to peer-initiated task conflict and helping behavior through their beliefs about interpersonal conflict is moderated by their passion for work, such that this indirect relationship is weaker at higher levels of passion for work.

Buffering effect of collectivistic orientation
The logic of COR theory similarly suggests that the likelihood that employees criticize their coworkers for invoking emotion-laded quarrels, to protect their self-depreciating ruminations about their frequently contested opinions (Bowling et al. 2010), should be mitigated when they give priority to collective over individual interests. Collectivistic people seek to maintain harmony, even at the expense of their personal situation (Moorman and Blakely 1995;Varma et al. 2009), so they likely have less desire to unleash their frustrations about peer-initiated task conflict in the form of accusations about the presence of interpersonal conflict, which would undermine such harmony (De Dreu and Weingart 2003;Pooja, De Clercq, and Belausteguigoitia 2016). Similarly, a collectivistic orientation tends to make employees more benevolent, even toward those who argue with them (Du et al. 2015;Van Dyne et al. 2000), so they likely are more forgiving of colleagues who continuously express opposing opinions (Semerci 2019). Fundamentally, this personal resource transfers employees' focus from individual concerns (e.g., frustrating colleagues who always disagree with them) toward their duty to ensure group harmony, so they are less likely to issue allegations that coworkers are responsible for the negative relationship dynamics (H. Wang, Lu, et al. 2017).
Similar to the passion for work, we predict a moderated mediation effect of collectivism. Employees' collectivistic orientation serves as a boundary condition of the negative, indirect relationship between peer-initiated task conflict and helping behavior, through beliefs that coworkers are responsible for emotion-based disputes. When employees have a natural tendency to maintain group harmony (Du et al. 2015), they are less likely to develop a conviction that their coworkers are the source of dysfunctional relationship dynamics in the presence of systematic contestations of their opinions, so they remain ready to extend help. Their collectivistic tendencies counter self-depreciating musings about persistent attacks on their viewpoints (Semerci 2019), which diminishes the chances that they halt their helping efforts due to allegations about impoverished coworker relationships.
Hypothesis 5: The indirect negative relationship between employees' exposure to peer-initiated task conflict and their helping behavior through their beliefs about interpersonal conflict is moderated by their collectivistic orientation, such that this indirect relationship is weaker at higher levels of collectivistic orientation.

Data collection and sample
We test the hypotheses with survey data collected in September 2020 from a Portuguese banking firm. The firm operates in the capital city of Lisbon, employs more than 1,000 employees, and is involved in a broad set of activities, including personal and corporate banking. In Portugal, the banking sector is marked by significant uncertainty and turmoil (Borges and Tavares 2020;Figueiredo et al. 2016). In particular, the aftermath of the 2008 financial crisis deprived the Portuguese financial system of access to the European inter-banking market, and this uncertainty-inducing situation has intensified with the COVID-19 pandemic, a crisis with effects that spill over into the workplace and spur different viewpoints about task-related topics (De Clercq and Pereira 2021). Intense competitive rivalry in this industry in Portugal (Borges and Tavares 2020) also means that employers rely on employees to go beyond their formal duties. From this perspective, it is highly relevant to investigate how persistent idea clashes might prevent extrarole helping activities, as well as how this negative process can be contained. With a single-firm focus, we also avoid the risk of unobserved organization-level differences, such as a general organizational climate that encourages work-related voluntarism (Bolino et al. 2010).
Although Portugal thus offers a relevant setting, the conceptual arguments that inform the hypothesized relationships are country-neutral. That is, we expect the relationships to arise in various countries, though perhaps with different strengths, depending on culture-driven forces. For example, the uncertainty-avoidant tendencies that mark Portugal (Hofstede, Hofstede, and Minkov 2010) suggest that attacks on individual opinions may feel highly intrusive, so employees respond vigorously with accusations about negative relationship dynamics and a reluctance to undertake discretionary work behaviors. But Portugal also features high levels of collectivism, so employees may be less likely to use peer-initiated task conflict as an excuse to refrain from helping behaviors that add to the collective (Podsakoff, Morrison, and Martinez 2018). These contrasting dynamics make Portugal a compelling setting that can offer valuable insights for firms that operate in similarly risk-averse, collectivistic cultures.
The survey, consistent with well-established methods (Brislin 1986), was first translated into Portuguese by a bilingual translator, then back-translated into English by a different colleague. After corrections of a few minor discrepancies, the survey was finalized in Portuguese. The administration of the survey occurred electronically, through an institutional license of the Microsoft Forms software held by the university of one of the authors. The employees of the participating organization were very familiar with this survey tool and considered it easy to navigate. The survey tool also adheres to ethical standards with regard to the confidentiality of data collection and storage. Yet another advantage of our reliance on an online survey tool, instead of a paper-and-pencil equivalent, is that it enabled data collection without physical contact between the researchers and participants, a critical issue during the pandemic.
In addition to the implications of the data collection tool itself, several other measures ensured that we complied with ethical standards and protected participants' rights. In particular, target respondents were informed that their answers would be treated with absolute confidentiality, that no individual-level data would ever be released, that our sole and research-driven interest was in detecting general patterns in aggregated data, and that they could withdraw from the study at any time. The survey also emphasized that there were no right or wrong responses, that respondents probably would give varying answers to specific questions, and that it was very important for them to complete the survey as truthfully as possible. These specifications diminish the danger of social desirability and acquiescence biases (Jordan and Troth 2020;Spector 2006).
We received a complete list of organizational employees, from which we randomly selected 400 employees for possible participation, using a random digit calculator. From these 400 contacted employees, we received 255 responses, for a response rate of 64%. This response rate admittedly is not extremely high, but it exceeds rates typically obtained from surveys in prior organizational research (e.g., Baruch and Holtom [2008] report an average response rate of 53% for studies that collect data from individual employees). Moreover, the organization's senior management regarded this response rate as excellent, relative to their previous internal data collection projects. The final sample consisted of 43% women; 94% of the respondents had worked for their organization for more than 10 years; 81% had at least some supervisory responsibilities; and 49% had a front-office job (i.e., direct interactions with customers), while 51% had a back-office job.

Measures
The measurement items for the five central constructs use seven-point Likert anchors (ranging from 1 ¼ "strongly disagree" to 7 ¼ "strongly agree").

Peer-initiated task conflict
To assess employees' exposure to task-related disputes, we relied on a four-item scale of task conflict (De Clercq, Thongpapanl, and Dimov 2009). In light of our conceptual focus on how such conflict originates from coworkers, we adapted the original wording to reflect employees' perceptions that their coworkers initiated the conflict, such as "My coworkers systematically initiate disagreements about task-related issues" and "My coworkers systematically express opinions which are in conflict with mine" (Cronbach's alpha ¼ .78).

Beliefs about interpersonal conflict
We measured employees' convictions that coworkers cause emotion-based relationship tensions with a four-item scale of interpersonal conflict (Pooja, De Clercq, and Belausteguigoitia 2016). Considering our interest in employees' beliefs that their colleagues are responsible for this negative situation, we also adapted these items, such that employees noted how the actions by their colleagues influenced relationship quality. For example, they indicated whether "The actions of my coworkers make it so that I often get angry while working with them" and "The actions of my coworkers make it so that I often experience tensions in my relationships with them" (Cronbach's alpha ¼ .87).

Helping behavior
We assessed employees' voluntary efforts to help peers using a seven-item scale of helping behavior (Williams and Anderson 1991). Two sample items were "I go out of my way to help coworkers" and "I assist coworkers with their work, even when not asked" (Cronbach's alpha ¼ .86). Using a self-rated scale of peer-oriented helping activities is in line with prior research (e.g., Lin, Koopman, and Wang 2020; Rubenstein, Allen, and Bosco 2019), which reasons that other assessors, such a colleague or supervisor, only have incomplete insights about the entire set of discretionary helping behaviors that employees undertake toward all colleagues (Chan 2009).

Passion for work
The extent to which employees experience positive energy from their work was assessed with a five-item scale of passion for work (Baum and Locke 2004). Respondents expressed their agreement with statements such as "I love to work" and "I derive most of my life satisfaction from my work" (Cronbach's alpha ¼ .83).

Collectivistic orientation
To measure the extent to which employees give precedence to group over personal interests, we relied on a four-item scale of collectivistic orientation (Triandis and Gelfand 1998). 1 Two sample items were "The well-being of my peers is important to me" and "If a peer gets a prize, I would feel proud" (Cronbach's alpha ¼ .77).

Control variables
The statistical models controlled for four individual factors: gender (1 ¼ female), organizational tenure (1 ¼ 5 years or less, 2 ¼ 6-10 years, 3 ¼ 11-15 years, 4 ¼ 16-20 years, 5 ¼ more than 20 years), job level (1 ¼ staff, 2 ¼ supervisor, 3 ¼ management), 2 and job type (front-office or back-office, with the latter as the base category). 3 Women tend to exhibit greater propensities to help coworkers voluntarily (Belansky and Boggiano 1994), employees who have been part of their organization for a longer time or occupy higher job levels might be more confident that they will be successful in their voluntary work efforts (De Clercq, Haq, and Azeem 2020;Podsakoff, Morrison, and Martinez 2018), and the type of job that employees do might influence the extent to which opportunities for helping present themselves (Podsakoff, Morrison, and Martinez 2018).

Construct validity
The validity of the study's focal constructs was assessed with a confirmatory factor analysis of a five-factor measurement model (Anderson and Gerbing 1988). The fit of this model was good: v 2 (218) ¼ 399.80, confirmatory fit index ¼ .94, incremental fit index ¼ .94, Tucker-Lewis index ¼ .93, and root mean squared error of approximation ¼ .06. In support of convergent validity, each measurement item had strongly significant factor loadings (p < .001), and the values of the constructs' average variance extracted (AVE) exceeded, or were close to, the benchmark of .50 (cf. .47 for helping behavior, .49 for passion for work). In support of discriminant validity, the AVE values were all higher than the squared correlations of the associated construct pairs, and the fit of the 10 models that included constrained construct pairs (correlations between two constructs fixed) was significantly worse than the fit of the corresponding unconstrained pairs (correlations between constructs could vary) (Dv 2 (1) > 3.84, p < .05) (Lattin, Carroll, and Green 2003).

Statistical procedure
To empirically test the proposed moderated mediation framework, we relied on a well-established procedure (e.g., De Clercq, Fatima, and Jahanzeb 2021;Haq, De Clercq, and Azeem 2022;Mitchell, Boyle, and Nicholas 2021;Yang and Yang 2020) that combines hierarchical moderated regression analysis with the Process macro technique (Hayes 2018). Specifically, to assess the direct relationship between peer-initiated task conflict and beliefs about interpersonal conflict (Hypothesis 1), the direct relationship between beliefs about interpersonal conflict and helping behavior (Hypothesis 2), and the moderating effects of passion for work and collectivistic orientation on the direct relationship between peer-initiated task conflict and beliefs about interpersonal conflict (as a preliminary assessment of Hypotheses 4 and 5), we ran several multiple regression models. In line with the recommendation by Aiken and West (1991), the constitutive variables of these product terms were mean-centered.
To assess the presence of mediation (Hypothesis 3), we applied the Process macro's Model 4 and calculated the effect size of the indirect relationship between peer-initiated task conflict and helping behavior, through beliefs about interpersonal conflict, as well as the associated bootstrap 95% confidence interval (CI; Hayes 2018). The bootstrapping technique avoids the problems that arise when indirect effects are asymmetric and do not follow a normal distribution (MacKinnon, Lockwood, and Williams 2004). To formally evaluate the presence of moderated mediation (Hypotheses 4 and 5), we applied the Process macro's Model 7 and calculated the conditional indirect effects of peer-initiated task conflict (and corresponding CIs) at three levels of passion for work and collectivistic orientation, respectively: one standard deviation (SD) below their mean values, at their mean values, and one SD above their mean values, as specified in the Process macro (Hayes 2018). 4

Main analysis
We present the zero-order correlation coefficients and descriptive statistics in Table 1, then the  hierarchical regression results in Table 2, in which Models 1-3 predict beliefs about interpersonal conflict, and Models 4-6 predict helping behavior. The variance inflation factors of each regression coefficient in each model are below 5.0, so we do not have concerns about multicollinearity (Studenmund 1992).
Consistent with our prediction in Hypothesis 1 that employees who feel upset by coworkers for constantly contesting their opinions are more likely to criticize these coworkers for causing emotion-based quarrels that harm their relationships, Model 2 reveals a positive relationship between peer-initiated task conflict and beliefs about interpersonal conflict (b ¼ .223, p < .001). We also find support for our prediction in Hypothesis 2 that these convictions about quarrels steer employees away from voluntary work efforts, as apparent in the negative relationship between beliefs about interpersonal conflict and helping behavior in Model 6 (b ¼ À.184, p < .01). The assessment of mediation by interpersonal conflict beliefs indicates an effect size of À.041 for the indirect relationship between peer-initiated task conflict and helping behavior, through these beliefs. The CI of this indirect relationship does not span 0 ([À.094; À.005]), in support of the mediation that we predict in Hypothesis 3 (Table 3). To evaluate the theorized buffering effects of the two personal resources, we estimate the peerinitiated task conflict Â passion for work and peer-initiated task conflict Â collectivistic orientation product terms in predicting beliefs about interpersonal conflict (Model 3, Table 2). Both product terms are negative and significant (b ¼ À.085, p < .05; b ¼ .À.166, p < .01, respectively). In line with the graphs of the moderating effects in Figures 2 and 3, the corresponding slope analyses indicate that the relationship between peer-initiated task conflict and beliefs about interpersonal conflict is positive and strongly significant at low levels of passion for work (b ¼ .370, p < .001) and collectivistic orientation (b ¼ .370, p < .001) but weakly or not significant at high levels of both personal resources (b ¼ .119, p < .10; b ¼ .038, ns, respectively). That is, we find evidence of the predicted buffering effects.
The assessment of moderated mediation by passion for work (Table 3) reveals that the effect sizes of the indirect relationship between peer-initiated task conflict and helping behavior grow weaker at higher levels of this personal resource (i.e., À.058 at one SD below the mean, À.039 at its mean, and À.017 at one SD above its mean). At the lower two values, the CI does not include 0 ([À.134; À.006] and [À.090; À.004], respectively); at the highest level, the CI does include 0 ([À.057; .009]). These findings support Hypothesis 4. Similarly, the indirect effect of peer-initiated  task conflict is subdued at higher levels of a collectivistic orientation (i.e., À.069 at one SD below the mean, À.045 at its mean, and À.022 at one SD above its mean). The CIs at the lower two values of collectivistic orientation do not span 0 ([À.135; À.018] and [À.094; À.011], respectively), whereas the CI includes 0 at the highest level ([À.067; .002]). Thus, we obtain evidence in support of Hypothesis 5 too.

Post hoc analysis
An alternative to the well-recognized approach of combining moderated regression analysis with the Process macro procedure, as adopted herein, would be structural equation modeling (SEM). However, a critical problem of SEM when testing conceptual frameworks that include multiple moderating effects is the issue of nonlinearity when calculating all possible product terms for the measurement items that load on interacting constructs (Lattin, Carroll, and Green 2003;Ping 1996). As a robustness check of the reported results, we accordingly follow Ping's (1996) recommendation and apply a path analysis, which lumps the items of each construct into a single indicator before calculating the product terms. The use of aggregate indicators to estimate moderating effects removes the aforementioned estimation difficulties due to nonlinearity (Bollen 1989;Hair et al. 2019). The results generated from this path analysis are consistent with those reported in Table 2: a positive relationship between peer-initiated task conflict and beliefs about interpersonal conflict (b ¼ .204, p < .001), negative relationships of the peer-initiated task conflict Â passion for work (b ¼ À.085, p < .05) and peer-initiated task conflict Â collectivistic orientation (b ¼ À.166, p < .01) interaction terms with beliefs about interpersonal conflict, and a negative relationship between beliefs about interpersonal conflict and helping behavior (b ¼ À.243, p < .001). 5

Discussion
This study contributes to prior research by detailing how employees' experience of peer-initiated task conflict may escalate into a reluctance to engage in discretionary helping behaviors, due to various factors. To extend research that establishes that peer exchanges marked by bullying or ostracism steer employees away from helping efforts (Devonish 2013;Peng and Zeng 2017), we specifically investigate how (1) the probability of help, in response to systematic expressions of disagreement by coworkers, may decrease because employees criticize coworkers for starting emotion-based quarrels and (2) the personal resources of passion for work and collectivistic orientation subdue this process. The statistical results confirm our conceptual predictions. In particular, employees' irritations with colleagues who persistently attack their viewpoints lead them to refuse to offer them work-related assistance, because they hold the disagreeable coworkers accountable for compromising their relationship harmony (Pluut and Curşeu 2013;Simons and Peterson 2000). In COR terminology, employees respond to resource-draining coworker treatment with negative convictions and behaviors, in their attempt to protect their self-esteem resources (Hobfoll and Shirom 2000;Jungst and Janssens 2020). These responses seem justified and are meant to avoid self-damaging ruminations about intrusive task-related disputes (De Dreu and Weingart 2003;Jungst and Blumberg 2016). Our empirical findings are theoretically interesting, in that they pinpoint the danger of a counterproductive spiral in which one unfavorable situation (systematic contestations of viewpoints by coworkers) generates another (complacent behavioral responses toward the coworkers). If such refusals to help upset the coworkers, employees could have an even harder time getting the coworkers to listen to their ideas and opinions without criticizing them (Korsgaard et al. 2010).
Another theoretical contribution of this study results from the evidence we provide about how to contain the negative spiral, namely, by ensuring that employees leverage personal resources to deal with the experienced hardships. As we hypothesized, employees' accusations that coworkers are responsible for emotion-based quarrels serve as less forceful conduits through which peer-initiated task conflict translates into a reluctance to extend voluntary help when those employees (1) are equipped with a strong passion for work (Houlfort et al. 2015) and (2) are collectively oriented (Du et al. 2015). In line with COR theory, self-depreciating ruminations resulting from coworker attacks are subdued among employees who can draw on positive work energy that helps them deal with these ruminations (Hobfoll et al. 2018). The probability that employees criticize their colleagues for impoverished relationship quality, then halt their voluntary helping efforts, decreases if employees feel vitalized by work (passion for work) or by their ability to put the interests of coworkers above their own (collectivistic orientation). These two personal resources stimulate employees, threatened by peer attacks on their opinions, to maintain a certain level of work-related voluntarism, because they refrain from accusations about the source of dysfunctional interpersonal dynamics (Kisamore et al. 2014;Lau and Cobb 2010).
In summary, we pinpoint beneficial roles of two pertinent personal features, which mitigate employees' complacency-inducing responses to peer-initiated task disputes. These findings complement prior investigations of the direct benefits of passion for work and collectivistic orientations in encouraging voluntary work behaviors (Gulyani and Bhatnagar 2017;L. Wang et al. 2013). We provide the additional insight that the detrimental influence of convictions about interpersonal quarrels, in response to peer-initiated task conflict, is attenuated among passionate and collectivistic employees. We thus reveal two distinct boundary conditions that diminish the risk of a negative loop, in which upsetting peer rejections of different viewpoints make employees "stingy" in their work efforts, which could further compromise any respect that coworkers still exhibit toward employees' opinions.

Limitations and future research
This study has some shortcomings, which indicate avenues for continued investigations. First, our theorizing assumed causality: Frustrations about peer-initiated task conflict prompt employees to form beliefs about interpersonal conflict, which then translate into diminished helping behavior. This causal logic is anchored in the robust COR framework and its postulate that resource-depleting work conditions generate negative beliefs and behaviors to avoid further resource losses (Hobfoll et al. 2018). In light of the study's cross-sectional design, the presence of reverse causality cannot be entirely eliminated though. For example, the personal satisfaction that employees derive from helping coworkers arguably could generate positive emotions that spill over onto how they experience their relationships and idea contests with coworkers. Longitudinal research, including measures of each focal construct at various points in time, would support estimations of cross-lagged effects and explicit tests of causality (Antonakis et al. 2010). Similarly, we relied on the well-established COR argument that employees tend to protect their self-esteem resources in response to upsetting, intrusive workplace experiences (Bentein et al. 2017;Hobfoll 2001), but we did not formally measure these resources, nor how they change over time. Research that uses longitudinal designs could address this issue.
Second, we focused on the buffering roles of passion for work and collectivistic orientation, noting previous studies that identify these two contingency factors as beneficial for diminishing work-related hardships (Du et al. 2015;Lavigne, Forest, and Crevier-Braud 2012). It would be interesting to consider other personal resources with potentially similar mitigating effects, such as employees' creative self-efficacy (Tierney and Farmer 2011), proactive personality (Z. Wang, Zhang, et al. 2017), or resilience (Edwards and Ashkanasy 2018). Resourceenhancing relational resources also might protect employees against the risk that their experience of peer-initiated task conflict escalates into beliefs about interpersonal conflict and reduced helping behavior, such as trust in the employer (Vanhala and Dietz 2019), group cohesion (Jung et al. 2016), or goal congruence (Chan and Lam 2011). It would be valuable to assess the relative power of each contingency for alleviating employees' suffering from idea clashes.
Third, we intentionally investigated one organization in one industry, to control for unobserved organization-or industry-level differences that may influence employees' propensities to support their peers on a voluntary basis. Yet this research design also decreases the external validity of the findings (Malhotra 2010). A useful extension would be to check for variation in employees' average helping behavior across different banking organizations, according to the organizations' internal corporate culture or the composition of their top management teams, for example (Podsakoff, Morrison, and Martinez 2018). In terms of the role of the industry, the hypothesized relationships are notably industry-neutral, so we do not expect that the signs of these relationships would vary across different sectors-but their strength might. For example, the level of competitive rivalry may have an impact on the perceived need for employees to add to organizational effectiveness with extra-role work activities (Lahiri, P erez-Nordtvedt, and Renn 2008). Moreover, the likelihood of these activities may be high, even in the presence of peer-initiated task conflict, in industries that are marked by limited external employability (Philippaers, De Cuyper, and Forrier 2017), because employees might strive particularly hard to make positive impressions on organizational leaders. Multi-industry studies could formally assess the role of these and other industry features.
Fourth, the rationale we used similarly was not country-specific, so we anticipate that the statistical findings apply to a wide cross-section of countries, even if the tested relationships might, again, differ in strength. As we explained in the Method section, in country contexts marked by high risk avoidance, employees likely experience others' systematic disagreements as very upsetting, so they respond with negative beliefs and actions (Hofstede, Hofstede, and Minkov 2010). But collectivism-in addition to aligning with the personal resource of collectivistic orientation-may subdue these behavioral responses, due to corresponding concerns about group harmony (Hofstede, Hofstede, and Minkov 2010). The statistical support for the research hypotheses implies that the former force outweighs the latter, though without empirical validation, we can only speculate. We suggest that further research explicitly consider how pertinent macro-level factors (including cultural values) might influence the strength of the hypothesized relationships.

Managerial implications
This study provides insightful implications for organizational practice. Organizational decision makers should acknowledge a significant source of coworker-induced hardships: employees' perceptions that their personal opinions are not being taken seriously or are systematically contested. These perceptions can harm the employees themselves and their coworkers, who are no longer benefitting from good quality relationships in which they can count on the help of the focal employees (Kisamore et al. 2014). Yet idea clashes also can have benefits for organizations, such as by spurring creative thinking (De Clercq, Rahman, and Belausteguigoitia 2017). Furthermore, some employees might be hesitant to admit that they are annoyed with colleagues who contest all their ideas, which could make them seem too sensitive (Jungst and Blumberg 2016). These considerations lead us to recommend that organizations should discourage disrespectful task-related disputes while also gauging employees' feedback on the emotion-based challenges that they might experience in these situations (Rai and Singh 2013). Soliciting such feedback could take place during organization-wide discussion forums or in face-to-face exchanges with supervisors or human resources representatives (Khvatova and Block 2017).
But in some organizations, confrontations of conflicting viewpoints might be part of the culture, especially if they are considered highly beneficial, such as when the core purpose of the firm is to seek radical innovations (Hoisl, Gruber, and Conti 2017). Even if such disagreements about task-related topics are desirable, organizations must still take care that their workforce does not enter into a downward spiral, in which criticisms spark self-deprecating contemplations that culminate in a refusal to undertake voluntary helping efforts, to the detriment of the helpers, their colleagues, and the overall organization. As this study indicates, this spiral can be avoided when recruitment and retention policies embrace individual passion and group concern (Du et al. 2015;Gulyani and Bhatnagar 2017). By honing and leveraging these personal resources, organizations can increase the probability that employees maintain a certain level of workplace voluntarism toward coworkers, even if they feel disheartened by task-related disputes. To the extent that employees are fueled with positive energy-whether through their own dedicated work efforts or their attention to collective interests-they likely think in less pessimistic ways about peer-initiated idea clashes and thus remain motivated to contribute to the well-being of coworkers with their own voluntary helping efforts.

Conclusion
This research elucidates the roles of interpersonal conflict and two personal resources in the conversion of peer-initiated task conflict into diminished helping behavior. Beliefs about low-quality interpersonal dynamics are key mechanisms by which systematic idea contestations leave employees reluctant to contribute to coworker success with their voluntary efforts. We also identify mitigating factors in this harmful process, namely, when employees can draw from their work-related passion and orientation toward group harmony. We hope this research serves as a catalyst for continued investigations of the harmful consequences of persistent idea clashes, including their effects on work-related voluntarism, as well as ways to address them by leveraging relevant resources among the organization's employee ranks.
Notes 1. One item ("To me, pleasure is spending time with my peers") was removed because of its low reliability. 2. The survey explained that the staff category pertained to employees who had no supervisory responsibilities, the supervisor category referred to employees who held some supervisory responsibilities, and the management category pertained to employees with significant supervisory responsibilities.
3. The survey indicated that the front-office category pertained to employees who interacted with customers directly; the back-office category instead described employees who did not have such interactions. 4. In line with the proposed theoretical framework, we estimated two models that included moderating effects of passion for work and collectivistic orientation on the relationship between peer-initiated task conflict and beliefs about interpersonal conflict, respectively, but not between beliefs about interpersonal conflict and helping behavior. A robustness check affirmed that the two personal resources did not affect the second path. 5. The detailed results of this analysis are available on request.

Notes on contributors
Dirk De Clercq is Professor of Management in the Goodman School of Business at Brock University, Canada. His research interests include entrepreneurship, organizational behavior, and cross-country studies.
Renato Pereira is Professor of General Management in ISCTE Business School at Instituto Universit ario de Lisboa, Portugal. He is also a Researcher in the Emerging Markets Research Center at ISCIM, Mozambique. His research interests include entrepreneurship, innovation, and international business, with a particular focus on Africa.