Introduction

Prosocial behavior—actions aiming to help, benefit, and contribute to other people, collective groups, or society as a whole—are critical to facilitating and maintaining healthy and resilient societal dynamics (Grant & Dutton, 2012; Penner et al., 2005). Existing research has examined the innate individual characteristics underpinning prosocial behaviors (e.g., Meglino & Korsgaard, 2004; Penner et al., 1995) and contextual factors that can motivate a temporary increase in prosociality (e.g., De Dreu & Nauta, 2009; Grant, 2008a). However, to ensure the sustained development of a healthy and resilient social environment, we need better knowledge of the factors that can foster long-lasting increases in individuals’ prosocial behavior. In this research, we build on attachment theory to propose that an interpersonal factor, receiving social support, is associated with higher felt security, which can motivate long-term engagement in prosocial behavior both in the context in which the support was received (e.g., organizational citizenship behavior) and in further reaching ones (e.g., societal contributions). We first review research on the antecedents of prosocial behavior and attachment theory to demonstrate receiving social support as a promising predictor, and felt security as a key mechanism, driving higher prosociality in the long term. Then we build theory on how individuals who receive social support across diverse forms and sources would experience a sustained sense of security which, in turn, associates positively with their motivation to engage and their actual engagement in prosocial behavior in the long term.

This paper offers two main contributions. First, we enrich the prosocial behavior literature through focusing on an interpersonal factor—receiving social support—as a potential sustainable resource from which individuals draw to motivate their prosociality in the long term. Specifically, we build on attachment theory to identify felt security as a novel mechanism that explains the relationship between receiving social support and long-term engagement in prosocial behavior. Whereas prior research has focused primarily on static individual differences as predictors of prosocial behavior (Meglino & Korsgaard, 2004; Penner et al., 1995) and short-term interventions to enhance it (e.g., De Dreu & Nauta, 2009; Grant, 2008a), we suggest that this relational construct of receiving social support along with the higher level of felt security it cultivates could be key in driving a virtuous cycle of prosociality.

Second, we underscore the diverse forms of prosocial behavior that receiving social support could spur. Prosociality is bigger than only an ambition or any single behavioral manifestation of that ambition (Bolino & Grant, 2016). Our multi-study and multi-method empirical evidence provides robust findings across diverse outcomes ranging from contributing to one’s local community, to supporting organizational diversity initiatives, to engaging in social impact practices, to serving on nonprofit boards. The collective findings reaffirm how the many behavioral indicators of prosociality can be predicted by the interpersonal factor of receiving social support through a higher sense of felt security and prosocial motivation.

Theoretical Background

Antecedents to Short- vs. Long-Term Prosocial Behavior

Our research considers prosocial behavior broadly as “actions that promote or protect the welfare of individuals, groups, or organizations” (Bolino & Grant, 2016: 5). Organizational psychologists and management scholars have focused on prosocial behavior occurring inside organizations, including but not limited to organizational citizenship behaviors (OCBs), or voluntary behaviors that contribute to organizational effectiveness but are not recognized by the formal reward system (Organ, 1988; Podsakoff et al., 2000). We adopt the broader definition of prosocial behavior that encompasses any act people undertake that benefits the welfare of others with whom one has direct interaction or not, a group or community, society, or even future generations (Bolino & Grant, 2016). Here we review research on the antecedents of broadly defined prosocial behavior, integrating often-isolated literatures.

Past research has largely focused on individual traits as predictors of prosocial behavior, suggesting that a prosocial personality (e.g., Eisenberg et al., 2002), concern for others (Korsgaard et al., 1997; McNeely & Meglino, 1994), and prosocial values (e.g., Grant, 2008b) contribute to prosociality—both the desire to contribute and actual engagement in activities that benefit others. Substantial research has also examined the impact of social value orientation—the weight people place on collective vs. individual interests on decision processes (Messick & McClintock, 1968). Having a stronger collective orientation predicts higher engagement in collectively beneficial behaviors including helping others (McClintock & Allison, 1989), using public transportation (Van Lange et al., 1998), and protecting the environment (Garling et al., 2003).

A smaller but growing research body has investigated contextual factors influencing prosociality. Some research shows that job and leader characteristics can influence employees’ engagement in prosocial behavior within their organizations. For example, employees are more likely to help others within their organizations when their work is clearly defined, when they have autonomy and task variety, and when they receive timely feedback (e.g., Grant, 2008a). Consistent with our thinking on the potential impact of interpersonal relationships, employees who report to more transformational, supportive, trustworthy, or grateful leaders are more helpful to colleagues at work (e.g., Podsakoff et al., 2000). Moreover, some organizational characteristics predict prosocial behavior at work (e.g., Moorman et al., 1998). Specifically, employees perceiving that their organization values them and cares about their wellbeing were more likely to help coworkers and engage in tasks beyond requirements (Moorman et al., 1998), suggesting that receiving organizational support can increase employees’ prosocial behavior inside their organization.

Prior research has also examined the effects of interventions aimed to promote prosocial behavior. For example, receiving help from a fellow student led to helping another student on the same online platform—a form of generalized reciprocity (Baker & Bulkley, 2014). Receiving gratitude from their managers led fundraisers to make more voluntary calls (Grant & Gino, 2010). Individuals who reflected on their recent experiences of giving, contributed more money to their own university and donated more to an earthquake relief fund (Grant & Dutton, 2012). Further, recalling times when they were concerned about confirming negative gender stereotypes increased women’s feelings of solidarity with other women, which motivated them to provide more support for other women in their profession (Cortland & Kinias, 2023). Mindfulness meditation interventions also led individuals to feel and behave more prosocially during days in which they meditated in the morning, and to be more financially generous with others immediately following the interventions (Hafenbrack et al., 2020).

Our review suggests that research on prosocial behavior has largely overlooked factors that can foster long-term increases in prosociality. Still, it offers helpful insights into potential antecedents to higher long-term prosociality. Specifically, extant studies seem to imply three primary mechanisms through which prosociality, including generalized reciprocity, is heightened. These mechanisms include affect, learning, and value internalization (Penner et al., 2005). First, affective experience can motivate prosocial actions. When people are activated by the distress of others (e.g., Eisenberg & Fabes, 1991) or by the experience of moral emotions (Tangney et al., 2007), they are likely more motivated to respond prosocially. Research on generalized reciprocity has suggested that a person who receives help from someone pays it forward to a third party mainly due to gratitude (Bartlett & DeSteno, 2006; DeSteno et al., 2010) or positive mood (Emmons & McCullough, 2003) they experience from receiving help. Second, individuals engage in prosocial behavior through social learning, as they observe others’ prosociality and imitate it (e.g., Eisenber & Fabes, 1991). Finally, individuals are motivated to engage in prosociality when they internalize prosocial values (Dovidio, 1984) and maintain a desire to present themselves as good, moral individuals (e.g., Schwartz & Howard, 1984; Steele, 1988). Among these three mechanisms, affective experiences can be relatively fleeting occurrences that likely only motivate prosocial behaviors temporarily, whereas learning and the internalization of prosocial values may cultivate sustained and long-term prosociality. Therefore, factors that facilitate learning and prosocial value internalization could potentially promote long-term prosocial behavior.

We suggest that receiving social support could be one such factor. It can enable learning, as when social support comes in the form of identifiable role models, this improves academic motivation and performance (e.g., Dasgupta, 2011; Stephens et al., 2014). It also helps that, albeit temporary, receiving social support can repeatedly evoke emotions such as gratitude. However, drawing on attachment theory, we propose that beyond affect, learning, and value internalization as conceptualized in the extant prosociality and generalized reciprocity literatures, receiving social support can foster long-term prosociality through strengthening one’s felt security.

Attachment Theory

Attachment theory is regarded as a key framework to understand individuals’ interpersonal processes, and it describes how long-term interpersonal relationships shape people’s development and interactions with others (e.g., Hazan & Shaver, 1987). Grounded in developmental psychology, attachment theory (Ainsworth et al., 1978; Bowlby, 1969) posits that individuals tend to seek proximity to others in times of distress and as a protection from threats. When supportive others (e.g., parents, caregivers, spouses, supervisors) are available and responsive, they develop secure attachment. Knowing that an attachment figure will be present and available in times of need provides a “secure base” to explore the environment confidently, and it shapes the beliefs and expectations individuals hold about themselves and others, which is reflected in felt security (Mikulincer & Shaver, 2005; Taylor, 2011).

Although the attachment system is first formed during childhood with primary caregivers, research has shown that it both tends to be stable (Fraley et al., 2011; Harms, 2011) over time and can be updated through adulthood relationships (Arriaga et al., 2014). Specifically, research has shown that attachment styles developed during early life extend to many years later, into the adult and work years (Eisenberg et al., 2014; Hart et al., 2007; Hazan & Shaver, 1990). For example, Collins and Read (1990) found correlations between the attachment relevant characteristics of adults’ parents and their dating partners, which suggests that adults are attracted to potential partners who recreate their childhood patterns of support responsiveness. These findings are particularly relevant because of the far-reaching impact that attachment has in interpersonal relationships, including work relationships (Little et al., 2011). Childhood attachment figures are parents or teachers, and in adult life they shift to romantic partners and other meaningful figures such as colleagues and supervisors.

Studies have demonstrated that secure attachment leads to a greater sense of felt security, allowing people to shift mental resources from a defensive to a more externally oriented mode, reflected as stronger empathy, openness to others, trust, and helping (for reviews, see Mikulincer & Shaver, 2007; Steele & Steele, 2008). For example, Little and colleagues (2011) found that secure attachment at work enables a positive affective state that leads to prosocial behaviors in the form of OCBs, including helping coworkers. Similarly, secure attachment predicts constructive functioning, altruism, and civic virtue within a group or an organization (Huffmeier et al., 2014; Luke et al., 2020; Richards & Schat, 2011; Rom & Mikulincer, 2003). The sense of security obtained in one context can shape attitudes and behaviors in another context. In particular, there is a connection between adults’ experiences with their social relationships in the personal domain and how they approach relationships in the work domain. For example, Petriglieri and Obodaru (2019) examined dual-career couples and found that experiencing a partner as a “secure base” had a sustained positive effect on individuals’ professional development.

Building on attachment theory, we theorize that receiving social support from others can strengthen people’s felt security, which then increases their prosocial motivation and behavior in the long-term beyond proximal relationships or contexts in which the support has been received. Research on prosociality or generalized reciprocity has largely omitted individuals’ felt security as a potential mechanism. By focusing on felt security, we also answer recent calls for research on the integration of attachment theory to explain individual differences in and beyond the workplace (Harms, 2011; Yip et al., 2018).

Hypothesis Development

Receiving Social Support and Prosociality

Receiving social support, defined as “any process through which social relationships might promote health and wellbeing” (Cohen et al., 2000: 4), has been shown to covary with helping others (Johnson et al., 1989; see Taylor, 2011). Social support can take different forms, including instrumental (practical actions or provisions), emotional (warmth and nurturance), informational (knowledge, advice, or feedback), and companionate (activities with others) support (Cohen et al., 2000). There is ample evidence that receiving social support of varied forms can buffer stress and improve healthy behaviors and physical, emotional, and psychological wellbeing (e.g., Cohen et al., 2000; Taylor, 2011).

Sources of social support include family, friends, supervisors, mentors, coworkers, and organizations. In the management literature, receiving social support from direct supervisors has been shown to positively predict health and workplace wellbeing and buffer against turnover intentions (Hammig, 2017; Rugulies et al., 2006). Informal mentorship, sponsorship, and support from peers and supervisors positively predict global women leaders’ workplace wellbeing (Cortland & Kinias, 2019), and receiving social support can enable employees to successfully navigate work-family conflict (Carlson & Perrewé, 1999; Kossek, et al., 2011).

Although most research on the relationship between receiving social support and prosociality has assumed prosocial actions being reciprocated only within the group that provided the support, work on generalized reciprocity (Baker & Bulkley, 2014; Molm et al., 2007) suggests that prosocial tendencies resulting from social support could span more broadly. For example, Cojuharenco et al., (2016) experimentally manipulated individuals’ sense of social connectedness. Despite the instruction to write about generosity toward a specific person, this manipulation boosted participants’ sense of connectedness with others in general and their community at large, which led them to immediately afterward exert greater efforts on socially responsible behaviors including recycling, making environmentally conscious purchases, and financially contributing to an NGO that promotes ethical business trade. Similarly, Baker and Bulkley (2014) showed that participants who received help were subsequently more likely to help third parties. Thus, receiving social support seems to “fill up” individuals’ reservoir of psychological resources, enabling them to offer support to others more broadly. Following previous research on generalized reciprocity, we first hypothesize that receiving social support is associated with higher engagement in prosocial behavior, within and beyond the immediate context. Here we seek to replicate prior findings and to establish receiving social support as a valid contributor to generalized prosociality.

Hypothesis 1

Receiving social support is associated with higher individual engagement in prosocial behavior both within and beyond the immediate context.

Moving beyond validating receiving social support as a predictor of generalized prosociality, our research seeks to demonstrate the longer-term benefits of receiving social support on prosociality and the role of felt security, as suggested by attachment theory, in sustaining this relationship. Indeed, much of the value of our present research lies in showing that the relationship between receiving social support and prosociality is not at all a tenuous or fleeting one, but rather strong, sustainable, and persistent, and this is largely because one’s steady feeling of security represents the pillar that supports this relationship. We explain our theorizing below.

Receiving Social Support, Felt Security, and Long-Term Prosociality

Receiving social support leads individuals to feel accepted and cared for by others, and perceiving others as available and supportive in times of need strengthens people’s felt security (Isen et al., 1976; Mikulincer et al., 2005; for a review, see Carlson et al., 1988). Further, attachment theory suggests that felt security generated by experiencing a secure base can persist within individuals and have lasting effects (Fraley et al., 2011; Harms, 2011; Mikulincer & Shaver, 2007). Specifically, individuals feeling more secure tend to experience less defensiveness toward and are less likely to feel threatened by others (Mikulincer & Shaver, 2005). Felt security can also encourage an explorative tendency. Petriglieri and Obodaru (2019) suggest that individuals who experience a sense of security from partners are more likely to engage in exploratory career behaviors and ongoing professional identity development. With reduced defensiveness, lower sense of threat, and an explorative tendency, individuals with felt security are likely to form more and stronger connections with others within and beyond work. In turn, this can enable subsequently developing more genuine concern for others’ well-being, thereby engendering a stronger motivation to behave in ways that benefit others (Mikulincer et al., 2005; Shaver & Mikulincer, 2009). Such motivation has been referred to as prosocial motivation, which concerns people’s desire to engage in activities that protect, promote, and benefit the wellbeing of others (Batson, 1987; Grant, 2007; Grant & Berg, 2011). Indeed, research has suggested that felt security positively relates to relevant outcomes. These include willingness to help others (Mikulincer & Shaver, 2007; Mikulincer et al., 2005), motivation to engage in behavior that benefits a partner’s wellbeing (Schrage et al., 2002), volunteering (Erez et al., 2008; Gillath et al., 2005), and commitment to social and environmental issues (Mikulincer & Shaver, 2013). Given that felt security developed from receiving social support is likely stable, we expect its positive relationship with prosocial motivation to be long-lasting.

A strong prosocial motivation naturally leads individuals to find and engage in opportunities to benefit others (Batson, 1987; Grant, 2007). While the direct relationship between prosocial motivation and prosocial behavior has been well documented in the literature (e.g., Caillier, 2016; Grant & Berg, 2011; Grant, 2008b), we include prosocial motivation in our model for three reasons. First, we aim to demonstrate how felt security relates to prosocial motivation as well as to prosocial behavior, given both are important constructs of prosociality. Second, we wish to verify the validity of prior studies showing the link between prosocial motivation and prosocial behavior. Third, showing that felt security is related to higher prosocial motivation would suggest that it could also lead to a variety of different prosocial behaviors beyond the specific ones measured in our set of studies. Taken together, we hypothesize that receiving more social support predicts long-term higher levels of prosocial behavior, and that this relationship is sequentially mediated by higher felt security and prosocial motivation.

Hypothesis 2

Receiving social support is associated with higher individual engagement in prosocial behavior for the long term.

Hypothesis 3

The relationship between receiving social support and individuals’ long-term engagement in prosocial behavior is serially mediated by higher felt security, which in turn is associated with higher prosocial motivation.

Overview of Studies

We tested our hypotheses using four studies that complement each other in important ways. Study 1 utilized a large cross-sectional dataset from a global business school alumni survey with significant representation of participants in Asia, Europe, and North America to validate past research showing the positive relationship between receiving social support and prosociality. Study 2 examined the relationship between receiving social support and prosocial behavior over a multi-year time period using a large longitudinal dataset of American adults. In a retrospective survey of working Americans, Study 3 investigated the relationship between receiving social support and long-term prosocial behavior, sequentially mediated by felt security and prosocial motivation. Finally, Study 4 tested our proposed mediators—felt security and prosocial motivation—in an experiment. Table 1 summarizes the measure or manipulation of social support and measure of prosocial behavior for each of the studies. The authors have obtained institutional review board approval for this research. Informed consent was obtained from the participants at the start of each study.

Table 1 Summary of Variables and Measures

Study 1

Study 1 leveraged survey data to test Hypothesis 1, concerning the positive relationship between receiving social support and engaging in prosocial behavior within and beyond the immediate environment. Aiming to integrate across prior studies that focus separately on more proximal and further reaching prosocial behavior, in this study we measured social support as workplace social support, and prosocial behavior both in the immediate context (e.g., workplace) and in further reaching ones (e.g., community or society).

Sample and Procedure

Alumni of masters-level or executive education programs from an international graduate business school with campuses in Europe, Asia, and the Middle East participated in an online survey about their work experiences, roles and identities, and contributions. Alumni received email invitations from the dean to participate with unique identifier survey links through a survey management company that administered the survey, and a total of 5715 alumni completed the survey (response rate = 10.5%). Although the response rate is not high, it is standard in this type of survey questionnaires (Sax et al., 2003), and this is a large number of participants (n = 5715). These items were part of a much longer survey where international business school alumni responded to other questions about their careers. Cortland and Kinias (2019) used a subset of this dataset to test hypotheses about social support leading to work satisfaction through reducing the experience of stereotype threat using the female alumnae responses only. Here, we use the entire sample with social support predicting prosocial behavior and control for work satisfaction to isolate our hypothesized relationship between social support and prosocial behavior.

Although all participants had advanced business education, they were diverse in terms of degree (70% MBA), age (59% Generation X: age 37–52 at time of data collection), seniority (35% held C-Suite or CEO/President positions), gender (77% men), and the country and continent in which they lived and worked. Participants represented 119 countries (no more than 13% of the sample from any one country), with 60% of participants located in Europe, 15% in Asia, 9% in North America, 15% in all other regions combined.

Measures

Receiving Social Support

This study used the same measure of workplace social support as Cortland and Kinias (2019), which examined the role of social support in women’s experiences of stereotype threat and work satisfaction. Participants indicated (1 = yes, 0 = no) whether they have received the following sources of social support during their careers: “formally assigned mentors/sponsors” (formal mentors/sponsors); “mentors/sponsors not formally assigned” (informal mentors/sponsors); “supportive supervisors” (supervisors); “strong peer support” (peers), and “seeing people like you succeed in senior management positions” (role models). The five responses were summed to create a composite measure of social support, and standardized.Footnote 1

Prosocial Behavior in the Proximal Context: Workplace

To assess prosocial behavior within their organizations, participants indicated their contribution to diversity and wellbeing enhancing practices. Specifically, participants indicated the extent to which they personally created or contributed to five specific initiatives focused on enhancing diversity and inclusion (a formal employee diversity and inclusion program; developing and identifying female talent for leadership positions; developing and identifying talent from group(s) under-represented in leadership positions; supportive work arrangements that enable employees to meet work and family responsibilities; providing parental leave opportunities beyond legal mandates). Responses were recorded on a scale from 0 = not at all to 3 = I created or co-created. The items formed a reliable (α = 0.84) scale, so were averaged to form a composite, and standardized.

Prosocial Behavior in Further Reaching Contexts: Community and Society

Participants indicated their individual prosocial behavior beyond their proximal context (organization) in two ways: (1) economic development and societal impact, and (2) non-profit board membership. Economic development and societal impact practices.Footnote 2 Participants indicated the extent to which they personally created or contributed to five initiatives focused on economic development and societal impact (economic development in your region, identifying and increasing sourcing from local suppliers; protecting the environment/sustainability; protecting or promoting human rights; charitable giving). Responses were on a scale ranging from 0 = not at all to 3 = I created or co-created, and because they formed a reliable measure (α = 0.73), they were averaged to form a composite, and standardized.

Non-profit board membership. For a final measure of prosocial behavior in a further reaching context, participants indicated their service in response to the question, “On how many non-profit boards of directors do you serve?”. Given a positively skewed distribution (78% zero, 15% one, 7% two or more), we dichotomized the responses as follows: 0 = zero non-profit boards of directors, 1 = one or more non-profit board of directors.

Control Variables

To reduce concerns about potential confounding demographic and job-relevant factors, age, job status, participant gender, degree program, company size, and job scope served as control variables. We included these specific control variables as they both correlate with receiving social support and likely impact the extent to which individuals contribute to prosocial practices (Piliavin & Siegl, 2007). Specifically, age, job status, degree program, company size, and job scope are important control variables that account for qualifications to engage in certain types of prosocial behavior, such as being a member of a non-profit board. We control for gender, as prior work indicated gender differences in prosocial behavior, such that women tend to engage in more of some types of prosocial behavior than men (Eagly et al., 2003; Piliavin & Siegl, 2007), as well as gendered expectations of others to engage in prosocial behavior (Lee & Huang, 2018) aligned with social role theory (Eagly, 2009). Moreover, we control for age as prior research has documented an age effect, such that older people tend to engage in prosocial behavior to a greater extent than younger people (Mayr & Freund, 2020).

Additionally, we control for life satisfaction and work satisfaction given work has shown that happier people are more likely to engage in behaviors benefitting others (Kushlev et al., 2022). Thus, we isolate the relationship between receiving social support and engaging in prosocial behavior from a generally positive attitude and positivity toward work (Cohen et al., 2000; Moorman et al., 1998). Specifically, we also controlled for the single item of the Diener et al. (1985) scale included in the survey (“All things considered, how satisfied are you with your life as a whole these days?”; 1 = completely dissatisfied, 5 = completely satisfied) and Cortland and Kinias’ (2019) work satisfaction scale. The three-item work satisfaction scale (“At this stage in your life, how satisfied are you with the following: (1) Work that is meaningful and satisfying, (2) Opportunities for career growth and development, and (3) Professional accomplishments; 1 = not at all satisfied, 5 = extremely satisfied) was reliable in this sample (α = 0.83), so was averaged to form a composite control variable.

Results

Table 2 contains the means, standard deviations, and bivariate correlations for all Study 1 variables.

Table 2 Study 1 Descriptive Statistics and Bivariate Correlations

Analytic Approach

To test Hypothesis 1, we analyzed social support predicting prosocial behavior in proximal and further reaching contexts using a series of Ordinary Least Square (OLS) regression analyses with social support as the predictor variable and prosocial behavior (diversity and wellbeing enhancing practices, economic and societal development practices) as outcomes. Binary Logistic regression tested the relationship between social support and non-profit board membership. Covariates were included in each model.

Hypothesis Test

Consistent with Hypothesis 1, social support significantly predicted prosocial behavior in proximal and further reaching contexts on all three measures. Specifically, social support predicted diversity and wellbeing enhancing practices (b = 0.120, p < 0.001), economic development and societal impact initiatives (b = 0.079, p < 0.001), and non-profit board membership (b = 0.078, p = 0.039).

As a robustness check, we also ran these regressions without the covariates and found that social support predicted diversity and wellbeing enhancing practices (b = 0.104, p < 0.001), and economic development and societal impact practices (b = 0.068, p < 0.001), but it did not predict non-profit board membership (b = − 0.010, p = 0.771). We interpret the nonsignificant effect on non-profit board membership when covariates are excluded as logical, given the covariates account for qualifications to serve on non-profit boards (e.g., age and job status). Social support therefore facilitates diversity and economic development practices robustly and non-profit board membership only above and beyond qualifications (i.e., among those with required experience). We also examined gender and national cultural context (Asia versus Europe versus North America versus all other continents) as potential moderators. Neither gender nor national context qualified reported results. See Table 3 for results of Study 1.

Table 3 Study 1 Regression Analyses Results

Discussion

Study 1 supports Hypothesis 1 by showing the positive relationship between receiving workplace social support and prosocial behavior both in the proximal context (within the organization) and in further reaching ones (community and broader society). One outcome reflecting participants’ individual behaviors captured prosocial behavior within the organization (diversity and wellbeing enhancing practices) and two outcomes captured prosocial behavior outside the organization (economic development and societal impact practices; non-profit board membership). Supporting internal validity, the effects are robust controlling for age, gender, and seniority. Particularly important given that participants self-reported on the variables of primary interest, the results were robust with the inclusion of life satisfaction and work satisfaction measures in models. This helps to alleviate concerns about potential alternative explanations connected to response bias.

One limitation of this study is potential selection bias. It is possible that individuals who are conscientious and agreeable were more likely to participate in the study, truncating the range of responses. Also, although the Study 1 sample included respondents across geographic locations, it is less educationally and economically diverse than ideal (see Henrich et al., 2010). Specifically, given that all participants had studied advanced business, questions of generalizability to populations with different educational backgrounds and earning potential remain. This matters as education and wealth have been shown to relate to prosocial behavior (e.g., Piff et al., 2010).

In Studies 2 through 4 where we tested our additional hypotheses, we paid special attention to selecting samples with substantially more educational and employment diversity, despite them being geographically less diverse (all drawn from US populations), with complementary methodologies. Studies 2 through 4 focus on the duration of the relationship between receiving social support and prosocial behavior through longitudinal (Study 2) and retrospective (Study 3) surveys, directly measure the hypothesized processes of felt security and prosocial motivation (Studies 3 and 4), and enable testing causality through experimental methodology (Study 4).

Study 2

In Study 2, we tested whether receiving social support predicts engagement in prosocial behavior over time using an archival dataset with survey responses relevant to our hypotheses at three time points over a nineteen-year period. Here we examine the predictive power of people receiving social support on their prosocial behavior (Hypothesis 1), over time (Hypothesis 2). By testing prosocial behavior as predicted by receiving social support at earlier time periods, this disentangles our findings from potential reverse causation and enables the understanding of longevity of the relationship between receiving social support and prosocial behavior.

We used data from the Wisconsin Longitudinal Study (WLS, accessed on https://www.ssc.wisc.edu/wlsresearch/), a multi-wave study of a random sample of people who graduated from high schools in Wisconsin, US in 1957 (Herd et al., 2014). Our cross-lagged analysis (Finkel, 1995; Little et al., 2007; Peng et al., 2023) focuses on the years in which participants answered questions relevant to our hypotheses, specifically 1992, 2004, and 2011. The total number of observations in our analysis was equal to 8057. Relevant to participant diversity, the sample was 52% women, and education ranges from less than 1 year of college to postdoctoral studies, with a mean of less than 2 years of college (note that the WLS dataset does not include participant ethnicity or race).

Measures

Receiving Social Support

To assess social support, we adapted the emotional support measure that Piliavin and Siegl (2007) used as a covariate in their study of the health benefits of volunteering using the WLS dataset. These questions were: “Is there a person in your family, with whom you can really share your very private feelings and concerns?”, “Is there a friend outside your family with whom you can really share your very private feelings and concerns?”, “During the past month, have you received advice, encouragement, moral or emotional support from parents?”, “During the past month, have you received advice, encouragement, moral or emotional support from friends, neighbors, co-workers?”. Participants indicated 1 = yes or 0 = no to each, and we computed a composite measure of social support for every year of interest (i.e., 1992, 2004, 2011) by summing the four responses. We then standardized the variable.

Prosocial Behavior in the Community

We adapted Piliavin and Siegl (2007)’s volunteering scale to measure prosocial behavior. Piliavin and Siegl (2007: 454) defined volunteering as “taking actions, within an institutional framework, that potentially provide some service to one or more other people or the community at large”, which measures prosocial behavior well because the focus is on activities that benefit others. Specifically, this comprises questions about community centers (“What is your level of involvement with community centers?”), neighborhood improvement organizations (“What is your level of involvement with neighborhood improvement organizations?”), and charity organizations (“What is your level of involvement with charity or welfare organizations?). Responses ranged from 0 (not involved) to 4 (a great deal). We averaged the responses to form a composite score for each participant and standardized it.

Control Variables

Participants reported their gender, education, household income (log transformed) and marital status. A supplementary cross-lagged analysis controlled for these variables and found no change to results. One of the benefits of cross-lagged analysis is that it rules out potential spurious relationships due to unmeasured individual difference variables by controlling for the auto-regressive effects (i.e., the effect of the previous measure on the next of the same variable) (Burkholder & Harlow, 2003; Peng et al., 2023). We therefore report the analysis without including the control variables.

Results

Table 4 contains the means, standard deviations, and correlations for all Study 2 variables.

Table 4 Study 2 Descriptive Statistics and Correlations

Analytic Approach

We examined whether social support predicts prosocial behavior in subsequent years using a cross-lagged analysis (Finkel, 1995; Little et al., 2007; Peng et al., 2023). We estimated the structural equation model in Stata 18 and used maximum likelihood with missing values as estimation method.

Hypothesis Tests

Figure 1 reports the results of the cross-lagged analysis. We tested measurement invariance by estimating a configural invariance model (CFI = 0.991 and RMSEA = 0.023) and a metric invariance model (CFI = 0.988 and RMSEA = 0.023) (Widaman et al., 2010). Results supported metric invariance as the change in fit statistics were below the cutoff points of 0.01 for both CFI and RMSEA (Chen, 2007).

Fig. 1
figure 1

Study 2 Cross-Lagged Model Testing the Effect of Social Support on Prosocial Behavior. Structural equation model estimation: Maximum likelihood with missing values. The reported cross-lagged model does not include the covariates. Including the covariates does not meaningfully change the results. Number of observations = 8071. Model fit statistics: RMSEA = .032, TLI = .976, CFI = .995. *p < .05; **p < .01; ***p \(\le\) .001

Consistent with Hypothesis 1 and Hypothesis 2, social support was a significant predictor of prosocial behavior years later. The cross-lagged path from social support to prosocial behavior was significant across both intervals (social support in 1992 predicted prosocial behavior in 2004: b = 0.045, p \(\le\) 0.001; social support in 2004 predicted prosocial behavior in 2011: b = 0.047, p \(<\) 0.001). Model fits indexes overall fall in the good range: RMSEA = 0.032, TLI = 0.976, CFI = 0.995 (Bentler & Bonett, 1980; Hu & Bentler, 1999; Steiger, 2007).

Discussion

Study 2 provides evidence for the long-lasting relationship between receiving social support and engagement in prosocial behavior over a decade later. Consistent with Hypothesis 1, individuals who received social support are more likely to contribute to organizations aimed at benefiting others and society. More importantly, we found support for Hypothesis 2, as receiving social support positively predicts engagement in prosocial behavior many years later. While Study 2 demonstrated the relationship in time sequence—receiving social support occurring before engaging in prosocial behavior—and helps to rule out potential influence of unobserved factors through cross-lagged analysis, it did not directly test the process through which the relationship develops. Thus, the next step is to investigate the mechanisms explaining the relationship.

Study 3

In Study 3, we tested the long-term relationship between receiving social support in early life and the engagement in prosocial behavior many years later (Hypothesis 1 and 2), serially mediated by current felt security and prosocial motivation (Hypothesis 3) through a preregisteredFootnote 3 retrospective survey with working Americans.

Sample and Procedure

Two hundred fifty working adults (44% women; 83% White) participated in an online study in exchange for $0.70 payment. Participants were 30 years old or older (M = 46.11, SD = 11.14), recruited from Prolific Academic to complete a study about social and work-related perceptions and behaviors. After excluding 16 participants who did not complete the study or failed the attention check, we analyzed a final sample of 234 complete responses. Data are available at: https://rb.gy/yz6o9r.

Measures

Early Life Social Support

For our assessment of early life social support, participants indicated the informational and emotional support they received during their childhood and adolescence through responding to four items adapted from Cohen et al., (2000: “How often did you have someone who:” “Listened to you talk about your private feelings?” “Expressed interest and concern in your well-being?” “Suggested some action you should take in order to deal with a problem you were having?” “Gave you information that made a difficult situation easier to understand?”). As this was a reliable scale (α = 0.92), we averaged responses, and to enable comparisons of effects across studies, we also standardized the aggregate scale.

Felt Security

Participants indicated the security they feel at the current stage of their lives, by responding to the Luke et al., (2020) scale (sample items: “secure”, “safe”, “looked after”). Response options were anchored on a 1 (strongly disagree) to 7 (strongly agree) scale and were averaged to form a composite (α = 0.97).

Prosocial Motivation

Participants reported their prosocial motivation at the current stage of their life, by responding to Grant’s (2008b) scale: “I care about benefiting others”, “I want to help others”, “I want to have a positive impact on others”, and “It is important to me to do good for others”. Response options were anchored on a 1 (strongly disagree) to 7 (strongly agree) scale and were averaged to form a composite (α = 0.97).

Prosocial Behavior: Workplace, Community, and Society

We measured the extent to which participants engage in prosocial behavior at the current stage of their life, by using measures adapted from Study 1: diversity and wellbeing enhancing practices and economic development and societal impact practices.Footnote 4 For diversity and wellbeing enhancing practices, participants indicated the extent to which they currently engage in seven specific behaviors adapted from Study 1 (supporting colleagues’ ability to meet both their work and family responsibilities; emotionally supporting colleagues after they become parents; practically supporting colleagues after they become parents; making sure you give voice to women in meetings; making sure you give voice to underrepresented minorities in meetings; organizing practices to ensure women have the opportunity to thrive; organizing practices to ensure underrepresented minorities have the opportunity to thrive). Here, responses were recorded on a scale from 1 = not at all to 7 = extremely. The items formed a reliable (α = 0.95) scale, so were averaged to form a composite, and standardized. For economic development and societal impact practices, participants indicated the extent to which they are currently engaging in the same five specific behaviors as in Study 1. Here, responses were recorded on a scale from 1 = not at all to 7 = extremely. The economic development and societal impact practices scale was again reliable in this sample (α = 0.89), so was averaged to form a composite, and standardized.

Control Variables

Reported analyses control for age, gender, education, race/ethnicity, and number of subordinates participants supervise as a proxy for job status, as these variables can both correlate with receiving social support and impact the extent to which individuals contribute to prosocial practices. To reduce participant fatigue, we did not control for work and life satisfaction as we did in Study 1 given that results from Study 1 held with or without these controls.

Results

Table 5 contains the means, standard deviations, and correlations for all Study 3 variables.

Table 5 Study 3 Descriptive Statistics and Bivariate Correlations

Analytic Approach

To test the indirect effect of received social support on prosocial behavior through felt security and prosocial motivation, we employed the PROCESS macro by Andrew Hayes (Model 6 in Hayes, 2017). Social support was the predictor, felt security was the first mediating variable, prosocial motivation the second mediating variable, and prosocial behavior the outcomes. This command was run separately for each continuous outcome measure of prosocial behavior with 10,000 bootstraps. We report analyses controlling for gender, age, education, race/ethnicity, and job status.

Hypothesis Tests

Table 6 contains all direct effects of the mediation model predicting prosocial behavior from social support through felt security and prosocial motivation. Demonstrating robustness, hypotheses were also fully supported in the absence of covariates.

Table 6 Study 3 Direct Effects on Felt Security, Prosocial Motivation and Prosocial Behavior

Model 1: Social Support → Felt Security → Prosocial Motivation → Diversity and Wellbeing Enhancing Practices

The indirect effect of social support on diversity and wellbeing enhancing practices through felt security and prosocial motivation was 0.100. The 95% bootstrap confidence interval for this indirect effect did not include zero (0.058 to 0.156), indicating that this effect was statistically significant.

Model 2: Social Support → Felt Security → Prosocial Motivation → Economic Development and Societal Impact Practices

The indirect effect of social support on economic development and societal impact practices through felt security and prosocial motivation was 0.097, and the 95% bootstrap confidence interval for this indirect effect did not include zero (0.054 to 0.155), indicating that this effect was statistically significant.

Discussion

Study 3 findings support Hypothesis 1 and 2 with early life social support predicting current prosocial behavior. Moreover, findings support Hypothesis 3 with felt security and prosocial motivation sequentially mediating the indirect relationship between receiving social support on prosocial behavior within and beyond the immediate context, and over time. Further suggesting internal validity, the effects are robust controlling for gender, age, education, race/ethnicity, and job status.

While Study 3 demonstrated the relationship in retrospective time sequence—receiving early life social support predicts current felt security, prosocial motivation and engagement in prosocial behavior, it did not test alternative mechanisms or leverage random assignment to conditions to make causal inferences appropriate. Thus, the next step is to test causation and investigate our hypothesized sequential processes against alternative mechanisms potentially explaining the effect.

Study 4

In Study 4, we randomly assigned participants to experimental conditions of social support or control. We tested the relationship between receiving social support and engagement in prosocial behavior (Hypothesis 1), serially mediated by felt security and prosocial motivation (Hypothesis 3) through a preregisteredFootnote 5 experiment with working Americans.

Sample and Procedure

Four hundred working adults (49% women; 86% White) participated in an online study in exchange for $0.50 payment. Participants were 30 years old or older (M = 45.34, SD = 10.13), recruited from Prolific Academic to complete a study about social and work-related perceptions and behaviors. After excluding six participants who did not complete the study or failed the attention check, we analyzed a final sample of 394 complete responses. Data are available at: https://rb.gy/yz6o9r.

Receiving Social Support vs. Control Condition

Participants were randomly assigned to one of two experimental conditions dummy coded for analyses: social support = 1 versus control = 0. To be consistent with the other studies, we standardized the variable. A separate pilot study demonstrated that participants in the social support condition reported feeling more supported, comforted, and protected (α = 0.86, adapted from Jakubiak & Feeney, 2016) than participants in the control condition, t (208) = 6.27, p < 0.001. In both conditions, participants recalled a recent event and then wrote about it in a text box (participants were required to type at least 300 characters in the text box). Specifically, in the social support condition, the instructions to the recall and writing task read:

Please think about a situation that happened during the past few months in which you received support from a friend, family member, co-worker, or other social contact. This support might include practically doing something to help you, providing emotional support or useful advise, or keeping you company. Please try to focus on one specific situation in which you felt supported by this other person and describe: (1) the situation, (2) who helped/supported you, (3) how this person helped/supported you, and (4) how you felt when they helped/supported you.

In the control condition, the instructions read:

Please think about what you did last Tuesday. Please describe: (1) the things you did during the day, (2) any people with whom you interacted, (3) what happened in any such interaction(s), and (4) how were you feeling that day.

Measures

Felt Security

Following the experimental manipulation, participants indicated their feelings of security by responding to the Luke et al. (2012) scale used in Study 3 (α = 0.97).

Prosocial Motivation

Participants indicated their prosocial motivation by responding to Grant’s (2008b) scale used in Study 3 (α = 0.95).

Prosocial Behavioral Intentions: Workplace, Community, and Society

We measured prosocial behavior intentions with the same measures used in Study 3 adapted to reflect intentions rather than prior behaviors. For diversity and wellbeing enhancing practices, participants indicated the extent to which they would be interested in engaging in seven specific behaviors in their careers. The items formed a reliable (α = 0.92) scale, so were averaged to form a composite, and standardized. For economic development and societal impact practices, participants indicated the extent to which they would be interested in engaging in five specific behaviors in their careers. The items formed a reliable (α = 0.86) scale, so were averaged to form a composite, and standardized.

Control Variables

As in Study 3, reported analyses control for age, gender, education, race/ethnicity, and number of subordinates participants supervise as a proxy for job status. We also ran an additional study, reported in the Supplementary Material, with life satisfaction (Diener et al., 1985) and work satisfaction (Cortland & Kinias, 2019) also included as control variables.

Alternative Potential Mediators

To rule out alternative potential mechanisms explaining the effect of social support on prosocial motivation, we measured gratitude (α = 0.94) (adapted from Bartlett & DeSteno, 2006), and positive (α = 0.92) and negative affect (α = 0.89: PANAS; Watson et al., 1988). In an additional study reported in the Supplementary Material, we also included moral identity internalization (Aquino & Reed, 2002), prosocial identity (Grant et al., 2009), and efficacy in helping (McAllister et al., 2007) as potential mediators.

Results

Table 7 contains the means, standard deviations, and correlations for all Study 4 variables.

Table 7 Study 4 Descriptive Statistics and Bivariate Correlations

Analytic Approach

To test the indirect effect of received social support on prosocial behavior through felt security and prosocial motivation, we employed the PROCESS macro by Andrew Hayes (Model 6 in Hayes, 2017). Social support was the predictor, felt security was the first mediating variable, prosocial motivation the second mediating variable, and measures of prosocial behavior were the outcomes. This command was run separately for each continuous outcome measure of prosocial behavior with 10,000 bootstraps. We report analyses controlling for age, gender, education, race/ethnicity, and job status.

Hypothesis Tests

Table 8 contains all direct effects of the mediation model predicting prosocial behavior from social support through felt security and prosocial motivation. The bootstrapped path models are reported in Table 9.

Table 8 Study 4 Direct Effects on First-Stage Proposed Mediators, Prosocial Motivation and Prosocial Behavior
Table 9 Study 4 Indirect Effects of Social Support on Prosocial Behaviors

Model 1: Social Support → Felt Security → Prosocial Motivation → Diversity and Wellbeing Enhancing Practices. The indirect effect of social support on diversity and wellbeing enhancing practices through felt security and prosocial motivation was 0.051. The 95% bootstrap confidence interval for this indirect effect did not include zero (0.029 to 0.079), indicating that this effect was statistically significant.

Model 2: Social Support → Felt Security → Prosocial Motivation → Economic Development and Societal Impact Practices. The indirect effect of social support on economic development and societal impact practices through felt security and prosocial motivation was 0.050, and the 95% bootstrap confidence interval for this indirect effect did not include zero (0.029 to 0.078).

To examine potential alternative mechanisms explaining the indirect effect of social support on prosocial behavior through prosocial motivation, we ran multiple mediation models to compare and contrast mediators (Model 6 in Hayes, 2017). Social support was the predictor, felt security, gratitude, positive affect, and negative affect, were the first mediating variables, prosocial motivation was the second mediating variable, and measures of prosocial behavior were the outcomes. This command was run separately for each continuous outcome measure of prosocial behavior with 10,000 bootstraps. Results show that the social support manipulation influenced felt security (b = 0.298, p < 0.001), gratitude (b = 0.479, p < 0.001), and negative affect (b = 0.154, p = 0.044), but did not predict positive affect (b = 0.045, p = 0.550).

When examining the indirect effects of the social support manipulation on prosocial behavior, felt security was a significant first-stage mediator for both outcomes (as reported above). Gratitude was also a significant first-stage mediator for diversity and wellbeing enhancing practices (95% CI [0.035, 0.097]); and for economic development and societal impact practices (95% CI [0.034, 0.094]). Positive affect and negative affect were non-significant first-stage mediators (all 95% CIs included zero). However, when all the alternative mediators were in the equation, the only significant indirect effect was the one we hypothesized. Specifically, social support indirectly predicted both diversity and wellbeing enhancing practices (95% CI [0.006, 0.056]), and economic development and societal impact practices (95% CI [0.005, 0.057]). Importantly, the indirect effects through gratitude were not significant for diversity and wellbeing enhancing practices (95% CI [− 0.010, 0.019]) or for economic development and societal impact practices (95% CI [− 0.010, 0.018]). These results demonstrate that felt security is a key part of the process through which receiving social support leads to prosocial motivation and behavior that goes above and beyond the other potential alternative mechanisms. Notably, demonstrating robustness, hypotheses were fully supported in the absence of covariates, and we found no evidence of gender or race/ethnicity as boundary conditions.

See the Supplementary Material (Table S1) for the comparisons between alternative mediation models including gratitude, moral identity internalization, prosocial identity, and efficacy in helping in the supplemental experiment with additional processes measured. Felt security was the only statistically significant first-stage mediator in these analyses.

Discussion

Study 4 findings are consistent with causal inference supporting Hypothesis 2 with the social support manipulation increasing prosocial behavior relative to the control condition. Moreover, findings support Hypothesis 3 with felt security and prosocial motivation serially mediating the indirect effect of receiving social support on prosocial behavioral intentions within and beyond the immediate context (from organization to society).

General Discussion

Four studies tested and found support for hypotheses that receiving social support boosts prosocial behavior through higher felt security, which in turn leads to higher prosocial motivation. Evidence emerged across different datasets including cross-sectional, longitudinal, retrospective, and experimental data, with diverse samples (i.e., global business school alumni, adults around retirement age from the state of Wisconsin, and US working adults), both with and without important demographic, contextual, and psychological control variables. We found that receiving different forms of social support (emotional, instrumental, informational, and/or companionate) from various sources (e.g., peers, mentors and managers at work, friends and family outside work) leads people to engage in prosocial behavior in the short- and long-term. And the range of prosocial behavior included those directed toward targets both inside one’s organizations (i.e., diversity and wellbeing enhancing practices) and well beyond it (i.e., taking part in economic development and societal impact initiatives, serving on non-profit boards, volunteering in the local community).

Theoretical Implications

The present research advances prior research in at least three ways. First, it identifies and tests an interpersonal antecedent—receiving social support—to higher levels of prosocial behavior for the long term. In doing so, we not only extend existing research on prosociality to examining long-term, rather than short-term, relationships with higher prosocial behavior, but also broaden the perspective on predictors of prosocial behavior to include individuals’ experience with others in their social environment. Specifically, the relationship between receiving social support and prosocial behavior does not only exist in short-term, one-time exchange, as its replenishing effect could last over years, sustaining itself once developed. While past studies have suggested that prosocial behavior can be encouraged more when individuals experience a stronger need to reciprocate or a stronger self-efficacy to make changes, they nevertheless have found this positive effect to last only within a short timeframe (Grant et al., 2008; Rhoades & Eisenberger, 2002). Our research suggests that receiving social support, given its various sources and forms, as well as its effects on individual resources and perceived connection with others in general, can have a long-term boosting effect on engagement in prosocial behavior.

Second, drawing on attachment theory, we argue and find evidence that receiving social support relates to long-term prosocial behavior through higher levels of felt security. In doing so, we demonstrated a novel mechanism—felt security—that to our knowledge has not been included in prior studies on generalized exchange or prosociality. In particular, we showed that the mediating effect of felt security went beyond potential alternative mechanisms of gratitude, positive affect, negative affect, as well as moral identity internalization, prosocial identity, and efficacy in helping in our findings. By showing that felt security underpins how receiving social support relates to long-term prosociality, we add to the growing body of research suggesting that insights from attachment theory have the potential to explain individual behaviors in both work and non-work domains for the long term (e.g., Petriglieri & Obodaru, 2019; Maitlis, 2022).

Third, we answer the calls for greater attention to the varied types of prosociality (Bolino & Grant, 2016). Specifically, our studies speak to the argument that there is potential “value in distinguishing prosocial behavior based on their target” (Bolino & Grant, 2016: 635). Prior research tends to study prosocial behavior as a composite measure, an approach that risks overlooking the nuances of different types of prosocial behavior. We addressed this issue by distinguishing separately prosocial behavior directed toward different targets, and relatedly, those occurring in the proximal context (e.g., workplace diversity practices) and beyond it (e.g., economic development and societal impact practices). In particular, we examined novel operationalizations of prosocial behavior, such as diversity and well-being enhancing practices, non-profit board membership, and economic development and societal impact practices that to our knowledge have not been examined by the prosocial behavior literature but are clearly pro-social and relevant to improving organizations and society. Increasing felt security might be key to facilitating the internalization of prosocial values that can lead to a broad range of prosocial behaviors for an extended period of time.

Limitations and Future Research

Our studies have several limitations that open avenues for future research. We sought to examine a variety of different types of prosocial behavior individuals could engage both in the same or proximal context in which the support was received (e.g., workplace) or beyond it (e.g., community or society), yet these measures remain a limited set. Other types of prosocial behavior beyond the ones examined in this research will be worth researchers’ attention. Some prosocial behaviors such as whistleblowing not only require a desire to do good but also a tremendous amount of courage as they may involve risks that can hurt one’s career or, in some extreme cases, even life. Suggestive of such effects, women who see themselves as central in their networks are more likely to confront sexism when they encounter it (Brands & Rattan, 2020). But can receiving social support also promote other prosocial behavior, and if so, how? Future research may investigate which kinds of prosocial behavior could be more effectively facilitated by receiving social support, and more broadly, which interpersonal or contextual factor is most effective in fostering any specific kind of prosocial behavior.

In addition, we measured prosocial behavior through self-reported engagement in activities, and specifically, used a self-reported cross-sectional survey in Study 1, which could involve selection bias. For instance, it could be the case that individuals who are more agreeable and conscientious were more inclined to participate in our study, thus potentially limiting the generalizability of Study 1 results to individuals relatively high on these dimensions. However, the fact that the primary relationships are maintained with the inclusion of control variables and emerge across the population study (Study 2), a retrospective survey (Study 3) and experimentally (Study 4) in other samples reassure us that the relationships observed were not only the result of any possible sampling bias. Still, future studies should target diverse samples and leverage more objective measures of prosocial behaviors.

Moreover, Study 3 did not control for work satisfaction and life satisfaction, which could potentially have a positive impact on felt security. While we acknowledge this as a limitation, in the Supplementary Study we control for these two variables and the results hold. Our four studies complement each other in important ways and, as a package, they overcome the limitations of each in isolation. Specifically, Study 1 has benefits of a large globally diverse sample with the limitations of cross-sectional design with potential sampling bias, which is balanced by Study 2’s random sample of respondents from the population and time-lagged design, demonstrating the long-term effect of receiving social support on prosocial behavior. Study 3’s retrospective survey design cannot as strongly show time sequencing, but does identify processes of current felt security, current prosocial motivation, and current prosocial behavior. Finally, Study 4 has the limitation of outcomes being only behavioral intentions, but its experimental design offers causal evidence that receiving social support leads to increased prosocial behavior intentions through felt security and prosocial motivation.

Further, we did not differentiate among the different forms and sources of social support, as we are interested in social support as broadly perceived by individuals. Future work could tease apart potential variance in relationships between types of social support and types of prosocial behavior and contextual amplifiers. Lastly, given our focus was on felt security (driven by receiving social support), we did not compare different attachment styles. We invite future research to examine how different attachment styles shape the relationship between receiving social support and engaging in long-term prosocial behavior.

Practical Implications

Our research provides practical guidance for organizations interested in promoting more prosocial behavior among their employees (Brockner & Sherman, 2020). Specifically, our findings suggest that creating opportunities for employees to receive ample support from their colleagues and/or supervisors is effective, and it is also relatively easy to implement. Organizations striving to increase OCBs (see Parke et al., 2021) or corporate social responsibility output through employee volunteering might consider strengthening the social support system for their employees. In addition, beyond organizational contexts, our research suggests that providing social support to each other among family, friends, and community members can encourage sustained, long-term prosociality.

Conclusions

This work identifies receiving social support as a key factor that positively predicts individuals’ felt security, which in turn boosts prosocial motivation and engagement in prosocial behavior directed toward targets both in the proximal and in further reaching environments. Most strikingly, receiving social support can have a long-lasting impact on increasing prosocial behavior that could span over a decade. In short, providing social support seems a simple, yet effective, way to stimulate long-term prosocial behavior in organizations and society.