University students doing good online: typologies, motivators, and psychological outcomes of online prosocial behavior Authors: Sisi Tao, Gustavo Carlo Published: 2026 ABSTRACT BackgroundAs digital technologies increasingly shape the social worlds of university students, there remains a limited understanding of how they engage in prosocial behavior online, why they do so, and how these acts affect their well-being. Existing research has largely adapted offline frameworks or focused on performative forms (“like” on social media) of engagement, leaving more nuanced typologies and motivational processes underexplored.MethodWe conducted in-depth group interviews with 20 university students, recruited through purposive sampling to ensure a diverse range of demographics, to explore the types, motivations and psychological impacts of online prosocial behavior. This was done using a theory-informed thematic analysis.ResultsFive typologies of online prosocial behavior emerged: social support, instrumental support, online donations, cyberbullying intervention, and digital allyship. The first four typologies shared labels with well‑known offline forms of helping, but their mechanisms were transformed by features of the online environment, such as scalability, anonymity, and weak‑tie interactions. In contrast, digital allyship constituted a distinct meso-level form of prosociality, in which individuals leveraged their online presence to visibly advocate for and signal solidarity with vulnerable groups. Across all typologies, students highlighted value‑based motives, including empathy, personal moral commitments, and a sense of social responsibility, and they did not attribute their actions to complying with others’ requests or seeking public recognition. Psychologically, online helping was associated with both positive and negative psychological outcomes. An emergent process, which we term ‘digital prosocial resilience’, captured how students described their capacity to continue engaging in online prosocial behavior despite no feedback or backlash, sustained by internal satisfaction derived from acting in line with their values.ConclusionsThis study extends the conceptualization of offline prosocial behavior into digital ecosystems by identifying five distinct behavioral typologies. Additionally, by contextualizing the underlying value-driven motivations and psychological impacts, the findings offer targeted, practical insights into cultivating digital prosocial resilience among university students.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1186/s40359-026-05069-1. Introduction Prosocial behavior, defined as voluntary actions intended to benefit others such as helping, sharing and comforting, has long been recognized as a cornerstone of healthy individual development and social cohesion [10, 16]. Engaging in such acts is not only vital for recipients but also psychologically rewarding for the actor, fostering a sense of purpose and life satisfaction [4]. As such, understanding what motivates prosocial acts and how they are expressed is central to positive psychology and essential for promoting moral education and civic engagement. The rise of the internet has profoundly transformed how university students engage in prosocial behavior, creating new avenues for helping (e.g., from providing emotional support in online communities to organizing aid initiatives) with unprecedented speed and scale [6]. Key features of digital platforms, such as anonymity, broad reach, and instant connectivity, have fundamentally altered the dynamics of giving and receiving help [27]. While these affordances can amplify kindness and mobilize support across boundaries, they also bring new challenges, including reduced opportunities for social recognition and increased risks of exposure to online conflict or emotional fatigue [14]. Despite this profound shift, scientific understanding has lagged. Much of the existing research has focused on visible and often performative prosocial behaviors such as “slacktivism” and “like” on social media, leaving deeper questions unanswered [32]. For example, what nuanced forms of prosocial behavior appear online? What motivates youth to help others in digital spaces? What psychological effects, both positive (such as greater well-being) and negative (such as increased stress), do these actions have? The present study aims to addresses these gaps. Informed by the multidimensional model of prosocial behaviors [11] and utilizing a theory-informed thematic analysis, we aim to delineate the distinct typologies of online helping, while providing a rich, descriptive contextualization of the underlying motivations and psychological outcomes associated with these digital acts. The forms of online prosocial behavior The rapid evolution and widespread adoption of digital technologies has significantly expanded university students’ prosocial reach by connecting individuals across geographical barriers; lowered social barriers such as those related to appearance or social status; and fosters novel forms of prosocial behavior, including online activism, widespread information sharing, and the formation of virtual support communities [27]. The inherent visibility and persistence of many online actions further encourage engagement by offering avenues for social recognition [6]. Collectively, these features of the online context not only transfer traditional offline prosocial behaviors such as offering support or sharing resources into digital spaces, but also give rise to new forms of helping that are rarely achievable offline. Empirical research revealed how individuals leverage digital platforms to assist others in diverse ways. One primary way is online information and resource sharing, where individuals disseminate knowledge, advice, and critical resources to enhance community resilience and support. For instance, parents of children with acute lymphoblastic leukemia, individuals managing various medical conditions, and adolescents facing chronic illnesses frequently utilize online forums to exchange health information [3, 22, 37]. Another prominent manifestation is online emotional and peer support, involving the provision of empathy, comfort, and practical advice within digital communities. This is evident in mental health initiatives like the Buddy Project, where youth offer support to peers facing emotional challenges, and in online networks where university students and adolescents connect for reassurance and encouragement [5, 9, 31]. Additionally, digital platforms have fundamentally transformed charitable giving. Youth now actively donate to causes or individuals through online crowdfunding, with engagement often significantly influenced by prevailing social media trends and peer networks [21]. Despite growing interest in online prosocial behavior, most existing studies focus on concrete actions such as “sending a smiley” or “comforting someone online” [15]. These approaches often miss the nuanced and diverse forms of prosocial behavior that digital environments now enable and continuously reshape. As the online context evolves, it gives rise to novel types of helping behaviors that extend beyond traditional, easily observed acts (see [32]; for a review). Therefore, it is crucial to explore the broader typologies of online prosocial behavior to provide a more holistic understanding of how these different forms may interconnect and influence one another. Motivations for online prosocial behavior University students engage in online prosocial behavior for a multitude of reasons, which contemporary research typically categorizes into intrinsic and extrinsic motivators [8]. Intrinsic motivators stem from internal values and emotional satisfaction, while extrinsic motivators are driven by the pursuit of external rewards like social approval. Much of our foundational knowledge comes from frameworks developed for offline interactions. For instance, the multidimensional model proposed by Carlo and Padilla-Walker [11] has been instrumental in distinguishing various forms of prosocial behavior, and there is now a substantive body of work based on the model that has advanced our understanding of the unique developmental trends and correlates of distinct forms of offline prosocial behaviors (e.g., Wong et al., [48], Xiao [51], . Based on this approach, for example, researchers have examined six distinct forms of prosocial behaviors [12]. Altruistic behavior is primarily driven by selfless concern for others, motivated by empathy and moral principles without expectation of personal reward. Compliant behavior occurs in response to direct requests or social pressure, often to meet expectations or avoid disapproval. Emotional behavior is prompted by emotionally charged situations, typically involving empathy or personal distress. Dire behavior arises in emergencies or crises where immediate, significant help is needed. Public behavior is carried out in visible situations, motivated by the potential for social approval or recognition. Anonymous behavior involves helping without seeking acknowledgment or being observed by others. While this framework provides a robust vocabulary for describing prosocial acts, the critical question remains: how do these motivations manifest and evolve within the unique architecture of the internet? Translating offline models to the digital world is not straightforward, as the features of online environments fundamentally alter the dynamics of helping. On one hand, the visibility and permanence of online actions can amplify extrinsic motivations. As Lu et al., [24] found in online health communities, the desire to build a positive reputation can drive users to share accurate information. On the other hand, the option for anonymity can foster intrinsically motivated behaviors by reducing social pressure and the need for public acknowledgment. For example, Xia et al., [50] noted that during the COVID-19 pandemic, many shared verified information primarily for the personal satisfaction of helping others, not for external rewards. This evolving landscape reveals critical gaps in our current understanding. As a recent systematic review by Nuttall et al., [32] highlights, much of the existing research simply adapts offline frameworks without adequately accounting for the distinctive features of digital life—such as varied anonymity, global cultural contexts, and shifting social norms. Many studies also adopt a narrow focus on one or two motivators, neglecting the interplay between intrinsic and extrinsic drivers. Furthermore, although youth increasingly participate in civic engagement online, such as [52], the motivations behind social justice and digital civic engagement remain under-explored. These limitations contribute to a fragmented understanding that fails to capture the full, nuanced picture of why youth choose to help in online spaces. Psychological outcomes of online prosocial behavior Engaging in online prosocial behavior yields a complex and often contradictory set of psychological outcomes. The literature reveals a distinct duality: while the potential for positive emotional benefits is significant, these rewards are often accompanied by substantial psychological risks. On one hand, the positive psychological impacts are profound, primarily centered on fostering social connection and emotional validation. Research shows that connecting with peers who share similar life experiences online can powerfully reduce feelings of isolation and build a strong sense of community [5]. For marginalized groups in particular, these spaces offer a vital lifeline to “feel less alone” [13]. The perceived safety and anonymity of online environments can also facilitate self-disclosure, leading to emotional validation and empowerment that fosters personal growth and autonomy [23]. For some, these interactions provide tangible hope, normalize difficult experiences, and offer accessible coping mechanisms [46]. On the other hand, these supportive environments are simultaneously fraught with psychological risks. Individuals seeking or providing help may encounter hostility, mockery, or judgment instead of support, leading to significant distress [31, 43]. Beyond direct conflict, online engagement can induce stress from social comparison, burdensome friendship expectations, and the pressure of social media metrics [47]. Furthermore, the very act of engaging can lead to exposure to triggering or harmful content, creating an environment that can be emotionally overwhelming [13, 46]. Yet, the literature reveals several critical gaps in our understanding of online prosocial behavior. First, while researchers document the presence of adversity, they rarely examine the psychological processes individuals use to navigate such challenges. More importantly, the connection between the motivations for helping and the resulting psychological outcomes remains unclear. For example, it is not yet known whether intrinsic motivations (such as empathy or personal values) actively buffer individuals against negative psychological effects when they help others online. This potential protective mechanism could fundamentally shape how people experience online prosociality. Clarifying these causal links is essential for moving beyond a simple cost-benefit analysis and for developing a deeper, process-oriented understanding of online helping behavior. The current study Informed by a multidimensionality conceptual model of prosocial behaviors [11] and using a theory-informed thematic analysis approach, the present study seeks to explore the types, motivations, and psychological outcomes of online helping through the lived experiences of university students. Specifically, we aimed to address the following questions: What types of online prosocial behavior do youth exhibit?What motivates youth to engage in online prosocial behavior?What are the positive and negative psychological impacts of youth online prosocial behavior? What types of online prosocial behavior do youth exhibit? What types of online prosocial behavior do youth exhibit? What motivates youth to engage in online prosocial behavior? What motivates youth to engage in online prosocial behavior? What are the positive and negative psychological impacts of youth online prosocial behavior? What are the positive and negative psychological impacts of youth online prosocial behavior? Method Design We used a qualitative design based on hybrid deductive–inductive thematic analysis [17], informed by Carlo and Randall’s (2002) multidimensional model of prosocial behavior. This approach allowed us to apply established prosocial categories as sensitizing concepts while remaining open to novel, data‑driven themes specific to online contexts. For methodological transparency and rigor, the COREQ checklist is provided in Supplementary Material 1. We conducted online focus group interviews rather than individual interviews because our analytic focus was on how young adults in Hong Kong higher education jointly construct and negotiate understandings and norms surrounding online prosocial behavior. Methodologists note that focus groups are particularly well suited to examining social norms and moral or sensitive topics, as the interaction between participants allows shared assumptions, points of tension, and contested meanings to surface in ways that one-to-one interviews may not capture [29]. To mitigate potential social desirability pressures, we permitted pseudonyms, emphasized that both positive and negative experiences (including choosing not to help) were welcome, and actively invited divergent perspectives during facilitation. Participants Participants were 20 university students (12 women, 8 men) aged 19–31 years (M = 25.5, SD = 3.54) enrolled in higher education programmes in Hong Kong. We used purposive sampling to recruit students who were currently enrolled and actively engaged in online interaction (e.g., regular social media or online community use). Recruitment was conducted via university mailing lists and social media posts distributed through student networks; interested students contacted the research team by email and were then screened against inclusion criteria. From the pool of volunteers, we selected participants to maximize variation in gender, study programme, and family socioeconomic background. No invited participants dropped out after scheduling. Table 1 summarizes key demographic characteristics. Ethical approval was obtained from the Human Research Ethics Committee of the Education University of Hong Kong [Ref No. 2024-2025-0140], in accordance with the Declaration of Helsinki. All participants provided written informed consent before data collection. Table 1Participants’ demographic informationVariablesMean/SD or FrequencyAgeMean = 25.5, SD = 3.54GenderFemale = 60%, male = 40%Family income1Less than HK$ 6,000 = 5%HK$10,000–19,999 = 20%HK$20,000–29,999 = 20%HK$30,000–39,999 = 20%HK$40,000–49,999 = 15%HK$50,000 or above = 20%Educational levelHigher diploma2 = 30%Undergraduate = 40%Master = 30%1 Median Monthly Household Income (Hong Kong, 2024): HKD 29,3002 A “Higher Diploma” in Hong Kong is a post-secondary qualification that roughly equivalent to an associate degree Table 1Participants’ demographic informationVariablesMean/SD or FrequencyAgeMean = 25.5, SD = 3.54GenderFemale = 60%, male = 40%Family income1Less than HK$ 6,000 = 5%HK$10,000–19,999 = 20%HK$20,000–29,999 = 20%HK$30,000–39,999 = 20%HK$40,000–49,999 = 15%HK$50,000 or above = 20%Educational levelHigher diploma2 = 30%Undergraduate = 40%Master = 30%1 Median Monthly Household Income (Hong Kong, 2024): HKD 29,3002 A “Higher Diploma” in Hong Kong is a post-secondary qualification that roughly equivalent to an associate degree Participants’ demographic information Less than HK$ 6,000 = 5% HK$10,000–19,999 = 20% HK$20,000–29,999 = 20% HK$30,000–39,999 = 20% HK$40,000–49,999 = 15% HK$50,000 or above = 20% Higher diploma2 = 30% Undergraduate = 40% Master = 30% 1 Median Monthly Household Income (Hong Kong, 2024): HKD 29,300 2 A “Higher Diploma” in Hong Kong is a post-secondary qualification that roughly equivalent to an associate degree Procedure Data were collected via four online focus group interviews conducted on Zoom over two weekends in August 2024. Each group comprised five students. After individual identity verification in the virtual waiting room, participants could choose a pseudonym during the discussion, a procedure intended to reduce social pressure and support candid disclosure. No non‑participants (e.g., caregivers, peers) were present during the sessions. Each focus group lasted approximately 90 min and was facilitated by the first author, with a research assistant present to manage technical issues and note‑taking. All interviews followed a semi‑structured interview guide (see Supplementary Material 2) developed from Carlo and Randall’s [12] qualitative work on prosocial behavior, adapted to the online context. The guide comprised: Ice‑breaker questions about everyday online activities;Open and probe questions about specific examples of online helping, motivations, and emotional reactions;Closing, reflective questions inviting additional experiences or contrasting views. Ice‑breaker questions about everyday online activities; Ice‑breaker questions about everyday online activities; Open and probe questions about specific examples of online helping, motivations, and emotional reactions; Open and probe questions about specific examples of online helping, motivations, and emotional reactions; Closing, reflective questions inviting additional experiences or contrasting views. Closing, reflective questions inviting additional experiences or contrasting views. The guide was piloted with two university students who were not part of the final sample; minor wording changes were made to improve clarity and cultural relevance (e.g., replacing platform names and adding examples of local online practices). At the beginning of each focus group, the interviewer clarified that “online” and “digital” prosocial behavior referred broadly to actions occurring on any platforms participants used, including but not limited to social media (e.g., Instagram, Facebook), messaging applications (e.g., WhatsApp), local discussion forums (e.g., LIHKG), and online marketplaces or crowdfunding sites. Participants were explicitly encouraged to share experiences from any other platforms they considered relevant to their online helping. Interviews were audio‑recorded (Zoom audio tracks) with participants’ consent and transcribed verbatim in traditional Chinese by the trained research assistant. We did not return transcripts to participants for member checking, consistent with a reflexive thematic analysis approach that emphasizes researcher interpretation, but we preserved participants’ wording in quotations to maintain authenticity. Field notes were made during and immediately after each group, documenting notable interaction patterns, emotional tone, and preliminary analytic insights (e.g., moments of disagreement, laughter after discussing backlash). To ensure thematic sufficiency, we systematically tracked code generation across the four focus group transcripts. Following the coding of the third focus group, we observed a decline in the generation of novel inductive codes. By the conclusion of the fourth focus group, the data yielded no substantively new codes or sub-themes related to the typologies or motivations. Consequently, we determined that the dataset possessed sufficient information power and thematic depth to address the research aims without requiring additional data collection. All focus groups were conducted in Cantonese, and audio recordings were transcribed verbatim in written Chinese by the trained research assistant. For reporting purposes, illustrative quotations were translated into English. The first author, a bilingual developmental psychologist, prepared the initial translations, which were then checked by her colleague, who is a bilingual professor in linguistics, against the original Chinese transcripts to ensure semantic equivalence and preservation of contextual nuance. Any discrepancies were resolved through discussion, prioritizing the conveyance of participants’ intended meaning over literal word-for-word translation, while retaining key culturally specific expressions where necessary. Researcher reflexivity and group dynamics The first author is a Chinese female developmental psychologist who grew up in a middle‑SES family and has resided in Hong Kong for 8 years. Her ethnicity, language, and educational trajectory are broadly similar to those of many participants, which likely facilitated rapport and ease of communication in the focus groups. At the same time, she does not represent the full breadth of experiences among Chinese young adults in Hong Kong (e.g., those from more disadvantaged, rural, or non‑university pathways), and her professional role as a faculty member may have shaped how students perceived the research context and framed their narratives. The second author is a male developmental psychologist and expert in prosocial development of Latino ethnicity who grew up in a low‑SES household and has resided in the United States throughout his life. His positionality brings a different cultural lens on youth development, prosociality, and digital contexts, and provided an external perspective during data interpretation. Together, the authors are PhD‑level researchers with expertise in developmental psychology and prosocial behavior who come from distinct and diverse backgrounds. We acknowledge that our positionalities and prior commitments to viewing young people as active moral agents may have influenced the questions we asked, the aspects of the data we foregrounded (e.g., value‑driven motives, resilience), and the interpretations we advanced. Readers should therefore interpret the findings with appropriate caution, recognizing the potential for researcher bias despite our efforts to engage in ongoing reflexive discussion and to preserve discrepant cases in the analysis. No prior relationship existed between the interviewer and participants. When inviting students, the email and study information sheet introduced the interviewer as a university researcher interested in understanding “how students use digital spaces to support others” and emphasized that there were no right or wrong answers. At the beginning of each group, the interviewer reiterated that both positive and negative experiences of helping (or choosing not to help) were welcome. We explicitly invited dissenting or alternative views during the focus groups. Throughout analysis, we engaged in regular debriefing discussions to scrutinize how our disciplinary backgrounds and expectations might have shaped coding and theme development, and we retained discrepant cases to avoid over-harmonizing participants’ perspectives. Data analysis We employed a hybrid deductive-inductive thematic analysis [17] to analyze the transcripts in Excel. This hybrid approach allowed us to leverage established prosocial frameworks as a priori sensitizing concepts while remaining open to novel, data-driven typologies that the existing theory could not account for. The analysis proceeded in three stages. First, deductive coding was applied using the Multidimensional Prosocial Behaviors Model [11] as a codebook to identify known forms of helping (e.g., Instrumental Support, Donations). Second, inductive coding was utilized for data segments that did not fit these predefined categories, allowing for the emergence of novel themes and unique online motivations. To ensure analytic rigor and consistency in the application of the deductive framework, the first author and the trained research assistant independently co-coded 25% of the transcripts. Discrepancies were resolved through collaborative discussion to refine code definitions. The interrater reliability (Cohen’s kappa = 0.86) indicated strong consistency in the application of the theoretical coding scheme. Following this validation step, the remaining transcripts were coded, and the initial set of 45 codes was iteratively reviewed and synthesized into a formal hierarchical coding tree. This finalized structure comprises three first-order categories (Typologies, Motivations, and Psychological Impacts) that encompass a total of 15 s-order sub-themes addressing the core research questions. The complete coding tree, including definitions and representative quotations, is presented in Supplementary Material 3. As our findings represent an aggregate, theoretical synthesis of the entire dataset viewed through the existing theoretical framework, individual participants are not positioned to validate them [30]. We therefore did not invite participants to validate the findings. Results The analysis shows that university students in Hong Kong higher education (hereafter referred to as ‘students or ‘university students’ for brevity) engage in five main types of online prosocial behavior: social support, instrumental support, online donations, cyberbullying intervention, and digital allyship. These are driven largely by empathy, personal values, and a sense of social responsibility rather than a desire for public recognition. Across these behaviors, students reported both beneficial outcomes (e.g., increased happiness, purpose, and connection) and significant risks (e.g., stress, backlash, and feeling underappreciated), and described a capacity to continue helping despite hostility or lack of appreciation, a pattern we conceptualize as digital prosocial resilience. Table 2 summarizes these core findings by outlining the typologies, motivations, psychological outcomes, and digital prosocial resilience that are elaborated in the sections that follow. Table 2Types, motivations, and psychological outcomes of youth online prosocial behaviourThemesExamplesTypologiesSocial SupportLeaving supportive comments for a peer on Instagram facing exam stress.Instrumental SupportAnswering technical questions on forums.Online DonationsPosting stationery online for donation to those in need.Cyberbullying InterventionCommenting on Facebook to support a person criticized for their outfit.Digital AllyshipCreating an Instagram post to promote products made by visually impaired individuals.MotivationsEmotionalContributing financially to a family in crisis after feeling moved by their social media story.AltruisticProactively making posts to raise awareness for charity without external prompts.DireDonating to a family in crisis after learning of a sudden death leaving dependents.AnonymityGiving financial help to a struggling online friend without public acknowledgmentNon-Compliant and Non-PublicHelping online without telling friends or family about the benevolent acts.Personal ValuesSpeaking up against unfair comments online, driven by the belief in doing what’s right.Sense of Social ResponsibilityReporting unlicensed street vendors online to prevent food poisoning.Psychological ImpactsPositive Emotional BenefitsFeeling happy and satisfied after contributing to a cause onlineNegative ExperiencesRisking becoming a target when defending someone from cyberbullying.Digital Prosocial ResilienceWriting constructive comments online without visible gratitude but knowing it improves the atmosphere. Table 2Types, motivations, and psychological outcomes of youth online prosocial behaviourThemesExamplesTypologiesSocial SupportLeaving supportive comments for a peer on Instagram facing exam stress.Instrumental SupportAnswering technical questions on forums.Online DonationsPosting stationery online for donation to those in need.Cyberbullying InterventionCommenting on Facebook to support a person criticized for their outfit.Digital AllyshipCreating an Instagram post to promote products made by visually impaired individuals.MotivationsEmotionalContributing financially to a family in crisis after feeling moved by their social media story.AltruisticProactively making posts to raise awareness for charity without external prompts.DireDonating to a family in crisis after learning of a sudden death leaving dependents.AnonymityGiving financial help to a struggling online friend without public acknowledgmentNon-Compliant and Non-PublicHelping online without telling friends or family about the benevolent acts.Personal ValuesSpeaking up against unfair comments online, driven by the belief in doing what’s right.Sense of Social ResponsibilityReporting unlicensed street vendors online to prevent food poisoning.Psychological ImpactsPositive Emotional BenefitsFeeling happy and satisfied after contributing to a cause onlineNegative ExperiencesRisking becoming a target when defending someone from cyberbullying.Digital Prosocial ResilienceWriting constructive comments online without visible gratitude but knowing it improves the atmosphere. Types, motivations, and psychological outcomes of youth online prosocial behaviour RQ1: common types of online prosocial behavior Across the focus groups, five common types of online prosocial behavior emerged: (1) social support, (2) instrumental support, (3) online donations, (4) cyberbullying intervention, and (5) digital allyship. The first four mostly occur at the individual person level and parallel offline prosocial behaviors, while the fifth occurs at the community level and leverage unique aspects of the digital context. Each form is detailed below with representative examples. Social support Participants frequently offered emotional support or companionship online, often through comments, private messages, or posts. For example, Fion (Group 1) shared, “Recently, I saw someone on Instagram wanting to jump off a building due to exam stress, and I left encouraging comments, suggesting ways to relax.” Others described consoling peers diagnosed with serious illnesses, using simple gestures like supportive comments or listening via chat to comfort those in need. Nurturing online relationships was especially important during the COVID-19 pandemic, when in-person contact was limited. Reby (Group 2) explained, “During the pandemic, it was really boring, so I started going online to meet friends and play together… chat online and like their posts.” By reaching out, engaging in shared activities, and showing appreciation for friends’ content, participants supported each other’s well-being and reduced feelings of isolation. Instrumental support Participants also frequently provide instrumental support or expertise to others online. This involved helping people solve problems or learn, especially in areas of the participants’ own knowledge. For example, Tony (Group 1) stated, “I have helped answer technical engineering questions in online forums that only specific people might know.” By lending his expertise in an academic domain, Tony was able to assist strangers in an online community. Other students similarly described answering questions on Q&A sites, helping classmates with coursework via messaging apps, giving advice in interest-based groups (e.g., coding help, language learning tips), or sharing helpful information or links (such as study materials or opportunities) with others who might benefit. Online donations Students share resources and donate money through online platforms. For instance, Eva (Group 2) noted, “I post items I don’t need, like stationery, online and give them to those in need.” Similarly, another participant explained using an online platform to donate to a charity after learning about its cause. In this way, students leveraged e-marketplaces, social media groups, or crowdfunding sites to contribute resources to individuals or organizations. These examples illustrate online charitable behavior, extending offline practices of donation and gift-giving into digital spaces. Cyberbullying intervention Intervening to protect or defend others in online conflicts was another form of prosocial behavior identified, particularly in the context of cyberbullying. Several participants described situations where they stood up for someone being harassed or unfairly criticized online. Kim (Group 2) shared, “I defended a girl on Facebook who received derogatory comments about her outfit while traveling. I commented that everyone has the freedom to dress as they choose.” In this example, Kim took action to counter cyberbullying behavior and support the victim in a public online forum. Participants also shared effective strategies for intervening in cyberbullying, such as posting counter-comments and reporting hurtful content to the webmaster of the online platform. Digital allyship Some participants engaged in advocacy-oriented prosocial behaviors, using social networks to champion causes or support vulnerable groups. Zoe (Group 1), for example, recounted, “I helped an NGO by creating an Instagram post to promote products made by visually impaired individuals.” In doing so, she leveraged her online presence to raise awareness and generate support for a social cause. The digital allyship also encompassed actions like promoting charitable campaigns, amplifying important social messages, and encouraging others to support community initiatives. Students took advantage of the broad reach of social media to act as allies or ambassadors for causes they cared about. RQ2: motivations of online prosocial behavior Participants articulated a rich array of underlying motivations for engaging in prosocial behavior online, revealing both classic prosocial motives and factors unique to the online context. Below we describe each motivation theme with supporting examples. Emotional A number of students indicated that their online helping behaviors are driven by emotional reactions to others’ situations. In other words, empathic concern (e.g., feeling moved or saddened by someone’s plight) motivated them to act. Participants described instances where seeing someone in distress online evoked strong emotions that spurred them to help. For example, Emily (Group 1) recounted reading about a family’s misfortune (an accident) on social media and feeling compelled to contribute financially to their cause because of the emotional impact of the story: “I thought it was so tragic, so I decided to help them a bit.” Altruistic Participants also emphasized pure altruistic intentions as a driving force behind their online prosocial activities. They described helping others simply because they believed it was the right thing to do, without expecting anything in return. Fion (Group 1) recounted how she observed peers engaging in online charity entirely out of goodwill: “Actually, no one told them to do it, but they proactively made posts to draw attention to the issue or find ways to help… It’s purely out of kindness.” In this example, Fion notes that these individuals were not prompted or rewarded – their actions were self-initiated and motivated by a genuine desire to help others. Dire Some participants reported being especially motivated to help when they encountered dire situations or urgent needs online. In these cases, the severity of someone’s hardship or the emergency nature of an appeal spurred action. For instance, one of Emily’s (Group 1) net friends involved a tragic story of a family in crisis (a sudden death leaving dependents), which led her to donate money through a posted link. Participants indicated that when an appeal or story online conveyed that someone was in serious trouble (such as a medical emergency, accident, or desperate circumstance), they felt a heightened obligation to step up. The online context makes these dire situations visible (through viral posts or crowdfunding campaigns), and students often reacted with quick generosity or support upon learning of them. Anonymity Students tend to help others online anonymously. Several students expressed that they feel more comfortable or drawn to prosocial acts that do not put them in the spotlight. Reby (Group 2), for example, described quietly giving financial help to an online friend in need: “During the pandemic, I transferred some money to a friend I met online who was struggling financially. I didn’t tell anyone else about it.” Participants frequently noted that the online medium allows for (and perhaps encourages) a level of anonymity in helping. They could donate under a pseudonym, send support via private messages, or use features like anonymous forums to give advice. Many found this appealing because it shifted focus to the act of helping itself rather than any personal credit. Non-compliant and non-pubic Participants did not report helping because someone asked them to (compliant motivation) or helping to gain public recognition or approval (public motivation). No student described engaging in online helping due to a direct request or obligation from others – all the prosocial acts were self-initiated rather than compliance with a personal ask or instruction. Likewise, students consistently denied seeking any kind of spotlight or praise for their online good deeds. In fact, several emphasized the opposite, that recognition was not a factor. For example, Bonnie (Group 2) reflected, “When I help someone online, I don’t really think much about it afterward. It’s just about solving their problem in the moment, not about recognition.” Similarly, Fion (Group 1) mentioned she doesn’t usually tell friends or family about her online benevolent acts because “it doesn’t seem necessary.” Personal values Participants often cited personal values and internal rewards as key motivations for helping online. Many described prosocial behavior as consistent with their moral principles, identity, or desire for emotional gratification. For example, Fion (Group 1) shared a belief in humanity’s goodness and a commitment to reducing harm: “I feel that humanity is inherently good, and we should strive to lessen harmful behaviors like cyberbullying.” Some also mentioned religious fulfillment, as Tony (Group 1) explained: helping others online gives him a “short burst of happiness” and makes him feel like he’s “done something good, like accumulating karma.” When confronting cyberbullying or unfairness, participants were driven by a sense of moral duty. Kim (Group 2) recounted: “I saw people making judgmental comments about a girl’s travel attire… I spoke up, saying that everyone has the freedom to wear what they like and that their comments were unfair. It felt important to stand up for what’s right.” Reby (Group 2) added, “When I see something wrong, I feel like I need to step in because it’s unfair for someone to face it alone.” Tony (Group 1) echoed this, acknowledging the challenge and risk of intervening: “standing up for someone feels like the right thing to do… it’s about values and doing what’s just.” Sense of social responsibility Beyond personal value, many students voiced a broader sense of responsibility to improve their online community or society as a motivation. They saw their prosocial actions as contributions toward a kinder, safer online environment. Fion (Group 1) emphasized this perspective, saying, “I hope our actions online can improve the environment and make it a kinder place. People are already struggling so much in life.” Another participant, Tony (Group 1), mentioned reporting unsafe practices (unlicensed street vendors) via an online hotline, out of concern “that other customers… might end up with food poisoning.” His action was motivated by protecting strangers in the community from harm. These examples reflect a prosocial drive grounded in social responsibility, where individuals feel motivated by civic-minded values and the desire to make the online (and offline) world better for others. RQ3: psychological and social impacts of engaging in online prosocial behavior Participants often described these motivations and their emotional experiences as intertwined: acting from empathy, personal values, or a sense of responsibility was frequently followed by feelings of satisfaction and meaning, whereas situations where their values were questioned or rejected were more likely to be associated with stress or self-doubt. Positive emotional benefits Nearly all participants described experiencing positive emotions as a result of helping others online. Engaging in prosocial behavior gave them feelings of happiness, satisfaction, and meaningfulness. For instance, Zoe (Group 1) said, “The happiness from helping someone online is short-lived but still meaningful when you see it makes a difference.” Students commonly mentioned feeling “happy,” “good about myself,” or “satisfied” after they had assisted someone or contributed to a cause. Some noted that knowing they had eased someone’s burden or brightened someone’s day translated into their own improved mood or self-esteem. Negative experiences While the overall emotional impact was positive, participants also discussed some negative or challenging aspects of engaging in prosocial acts online. In certain situations, particularly those involving conflict or public visibility, helping others could expose the helper to stress, criticism, or other adverse outcomes. Tony (Group 1) described this risk vividly, noting, “When you stand up for someone being cyberbullied, you risk getting dragged into a heated argument or even becoming a target yourself.” Several participants echoed this sentiment, recounting that they had been “insulted”, “flamed,” or drawn into protracted online disputes after trying to help or defend someone. Resilience The final impact theme centers on participants’ perceptions that their online prosocial efforts often go unnoticed or unappreciated and can even lead to attacks. Yet, they still believe these actions matter. Students acknowledged that unlike some offline helping, digital prosocial acts do not always yield visible gratitude or recognition. For example, small acts like writing supportive comments, giving advice, or moderating a discussion can pass with little fanfare. Fion (Group 1) observed this reality, stating, “Even small actions like giving constructive comments can help improve the online atmosphere, though these efforts often go unnoticed.” Some mentioned that rarely did anyone praise them for intervening or donating online; in some cases, they were attacked after engaging in an online argument. As with anonymous aid, the beneficiaries might not even know who helped them. Nevertheless, students derived a quiet sense of pride in knowing they had acted in line with their values and had potentially improved someone’s life or the community climate, illustrating a form of digital prosocial resilience—the capacity to continue helping online despite indifference or hostility. Discussion The present study sought to provide a nuanced understanding of online prosocial behavior among university students by exploring the types of their actions, the motivations driving them, and the resultant psychological impacts. The findings show that students engage in a diverse repertoire of prosocial behaviors, driven primarily by strong internalized values rather than a desire for external rewards. While these acts foster well-being, they can also carry distinct emotional risks. Students exhibited distinct digital prosocial resilience, as evidenced by their ability to persist in online prosocial behavior despite facing hostility or a lack of explicit appreciation. This resilience was fuelled by the internal satisfaction they derived from acting in accordance with their values. A diverse repertoire of online prosocial behaviors In response to our first research question, university students engage in a diverse range of online prosocial behaviors. Five distinct types emerged from the data: (1) social support, (2) instrumental support, (3) online donations, (4) cyberbullying intervention, and (5) digital allyship. While the nomenclature of the first four forms aligns with established offline frameworks (e.Carlo and Randall [12], , our analysis reveals that the digital context does not merely host these behaviors but potentially transforms their psychological mechanisms. Specifically, digital affordances may reshape cost-benefit appraisals: anonymity lowers personal costs by reducing risks of face-to-face retaliation or reputational damage, while scalability enhances perceived benefits through broader reach, such as one forum post aiding thousands asynchronously. Such shifts enable online prosociality to operate beyond offline limits of physical proximity and strong-tie relationships (e.g., helping a neighbor or classmate). For instance, answering a question on Quora (a U.S.-based social question-and-answer website) shifts from a dyadic exchange to a ‘one-to-many’ contribution where a single act of advice-giving can asynchronously assist thousands of strangers. Similarly, online social support leverages anonymity to bypass social barriers, allowing students to offer deep emotional validation to peers they have never met, which rarely happens in the traditional support networks. Thus, while the forms appear traditional, their execution relies on a unique digital logic that decouples prosociality from established relationships. The emergence of digital allyship as a distinct theme is particularly noteworthy. We conceptualize this form of prosociality as a meso-level phenomenon encompassing public yet interpersonal online acts. In these acts, individuals leverage their personal platforms to visibly amplify and support vulnerable groups or individuals (e.g., sharing marginalized voices such as #MeToo, or urging community aid such as #CALFIRE). This behavior harnesses the unique affordances of digital platforms (e.g., broad reach and visibility) to facilitate community-level, civic-minded action that transcends micro-level, individual-to-individual support, yet remains distinct from macro-political mobilization. Crucially, digital allyship diverges from several related constructs. It differs from online activism, as it does not target formal political institutions or specific policy outcomes [28]. It is distinct from civic engagement, operating at a meso-level of interpersonal support focused on “standing with” marginalized groups rather than “participating in” structured civic systems [2]. Furthermore, it contrasts with moral advocacy, which centers on persuading others of a moral stance, whereas digital allyship prioritizes signalling support and creating visibility for others [1]. Finally, it separates itself from identity solidarity behaviors, as its primary motivations stem from personal values and social responsibility rather than shared group identity [36]. Consequently, we argue that digital allyship constitutes a distinct behavioral category within the typology of online prosocial behaviors. This categorization is justified by its unique operational criteria, scale, and intent, specifically: (1) public visibility (acts are identifiable, not anonymous); (2) beneficiary focus (targets are specific people or groups, rather than systemic or institutional structures); and (3) relational intent (aimed at demonstrating symbolic solidarity and validation, rather than persuasion, status-seeking, or political mobilization). As digital spaces become increasingly central to youth development, these findings urge researchers to move beyond merely quantifying the frequency or visibility of online actions. Future research could examine the underlying motivations and internal values driving these behaviors. Such a nuanced understanding is essential for guiding moral and civic education programs designed to cultivate not just outward engagement, but the internal ethical reasoning necessary for sustained, meaningful prosocial action both online and offline. Value-driven motivations The second research question explored the motivations behind these actions. In line with research based on the PTM and a multidimensional approach [11], which shows that students are motivated by emotional, altruistic, anonymous, and dire situations, the present findings indicate the absence of compliant (i.e., acting on request) and public (i.e., seeking recognition) motivations. Moreover, two new motivations emerged: personal values and a sense of social responsibility. These findings underscore the predominance of internal motives and the relative rejection of external pressures in much of the online prosocial behaviors expressed by university students. The rejection of compliant and public motivations by our participants can likely be attributed to the characteristics of the online environment. Many platforms allow for anonymous or semi-anonymous interactions, reducing opportunities for reputation-building and peer recognition, which diminishes the influence of motivations seeking social approval [27]. Unlike face-to-face interactions where direct requests can create strong social pressure, online interactions can be easily ignored without consequence, making prosocial actions more likely to be self-initiated and driven by internal factors such as empathy and personal values [34]. Furthermore, many online acts of help, such as anonymous forum posts and reporting mechanisms, are designed for privacy rather than visibility, making public recognition less relevant. This aligns with research on online health support communities, where anonymity reduces social evaluation apprehension and encourages genuine helping behaviors (e.g., Wright and Li [49], . Participants were significantly motivated by personal values to do good things online, which encompass deeply held moral identities and religious beliefs. These motivators have long been recognized as strong drivers of offline prosocial behavior [10]. For example, in their review, Hardy and Carlo [19], demonstrated that moral identity (which encompasses strong moral values and principles) predicts a wide range of prosocial actions among adolescents and young adults. Similarly, Saroglou and Cohen [39], found that religiosity correlates positively with volunteering, philanthropy, and charitable giving, especially when benefiting in-group members. Many world religions explicitly promote compassion, generosity, and service to others as moral imperatives, providing both external norms and internalized values that encourage helping behaviors [40]. Adding to this literature, the present findings suggest that personal values can also play a crucial role in motivating university students to engage in online prosocial activities. Additionally, a sense of social responsibility emerged as a key motivator, especially for digital allyship. This may be because the internet largely facilitates connections with like-minded communities, enabling individuals to engage according to their values and expertise without the pressure of social judgment. Social media’s reach and algorithms increase the visibility of community-level issues, such as #MeToo, fostering awareness and a sense of responsibility. Simple actions, like sharing content, reporting threats, or commenting, can significantly impact public narratives and contribute to the community. These low-effort, high-impact tools empower individuals to make meaningful contributions, emphasizing the role of social responsibility in motivating online prosocial behavior. Notably, although digital allyship is public by nature, it is motivated by altruism rather than the desire for public recognition. Psychological impacts and digital prosocial resilience In addressing the third research question, our study revealed a complex duality in the psychological outcomes of online prosocial behavior. While university students reported overwhelming positive emotional benefits from their engagement, they simultaneously acknowledged significant social risks and emotional costs. Crucially, this tension between satisfaction and stress highlighted a third, emergent dimension: the demonstration of digital prosocial resilience as a necessary capacity for sustained engagement. Consistent with the “warm glow” effect observed in offline contexts [20, 35], participants described profound feelings of happiness, satisfaction, and meaningfulness. In the online setting, this sense of efficacy is amplified by scalability; the ability to reach a wide audience or solve niche problems fosters a strong sense of purpose that reinforces well-being [18]. However, this satisfaction coexists with distinct digital risks. Participants reported stress, emotional exhaustion, and exposure to personal attacks, particularly during cyberbullying interventions [31, 44, 45]. The architectural features of the internet exacerbate these challenges: unlike fleeting offline confrontations, online disputes are persistent and public, while anonymity often emboldens aggressors to launch more severe attacks than would occur face-to-face [42]. Online environments also alter moral agency, making helpers’ sense of control vaguer since the impacts may feel indirect and hard to gauge (e.g., anonymous comments or donations yield no quick proof of success). The tension between positive emotions and digital risks requires a mechanism that enables youth to persist. Our findings indicate such a mechanism: digital prosocial resilience, which is an interpretive thematic pattern capturing how participants described and made sense of their continued helping despite backlash or lack of recognition. We conceptualize this not as a separate personality trait, but as a domain-specific manifestation of general resilience adapted to the digital ecology [26]. While general resilience involves adapting to broad life adversity (poverty or disease), digital prosocial resilience specifically involves navigating cyber-social threats, such as not being recognized or being attacked. Furthermore, it is distinct from moral steadfastness (rigid adherence to duty). While steadfastness implies enduring pressure unchanged, our findings suggest this resilience operates through a dynamic motivational process. Drawing on Self-Determination Theory [38], intrinsic motivation serves as the primary buffering mechanism. When extrinsic feedback is negative (e.g., being “flamed”) or absent (e.g., zero “likes”), the intrinsic value of “doing the right thing” replenishes the student’s psychological resources, preventing withdrawal. This finding offers a useful conceptual proposition for future research: whereas current research on digital resilience largely focuses on coping with victimization (e.g [33]), , our analysis suggests that, in participants’ meaning-making, ongoing online prosocial engagement under adverse conditions is sustained by an internal moral feedback loop: acting in line with personal values generates a sense of satisfaction that, in their accounts, supports continued helping. However, we offer this as a conceptual interpretation of participants’ narratives, rather than as an empirically tested causal mechanism, and highlight it as a possible avenue for future research. Developmental and sociocultural context Our findings must be interpreted in light of the specific developmental and sociocultural context of the sample. Participants were young adults in higher education in Hong Kong, spanning late adolescence to early adulthood (19–31 years). This age range encompasses heterogeneous developmental stages (e.g., traditional undergraduates versus older postgraduate students), who may differ in identity consolidation, civic engagement, and digital repertoires [41]. Thus, our group-based design did not allow us to systematically compare these subgroups, and future work could examine such within-sample differences. Moreover, participants’ online helping unfolded within Hong Kong’s digital ecosystem, where platform preferences (e.g., Instagram, WhatsApp, LIHKG), language practices (Cantonese, written Chinese, English), and local norms for public expression shape how support, advocacy, and confrontation are performed. Finally, as young adults socialized in a predominantly Chinese, collectivistic society, their prosocial narratives are likely informed by sociocultural scripts emphasizing relational harmony, concern for in-group others, and cautious engagement with contentious public issues [25]. These contextual features might both facilitate and constrain particular forms of online prosocial behavior, and could limit the generalizability of our findings to youth in other cultural or institutional settings. Contributions to theory and practice This study advances our empirical understanding of digital prosociality in several ways. Theoretically, it extends established offline prosocial frameworks [12] by showing how digital affordances transform four familiar forms of helping (e.g., social support, donation) and introduce a fifth, digital allyship, as a meso-level adaptation. These forms retain offline labels but operate through digital mechanisms like scalability and anonymity, which reshape processes such as cost-benefit appraisals and moral agency. Digital allyship leverages visibility to signal solidarity with vulnerable groups. It bridges individual helping and civic engagement without implying status-seeking or conflation with activism. The findings also contextualize established motivational mechanisms within the digital sphere. Among these university students, value-based drivers (empathy, moral commitments, social responsibility) dominated, while compliant or recognition-seeking motives were not endorsed. Additionally, the study conceptually outlines ‘digital prosocial resilience’ as a potential domain-specific manifestation of resilience in online contexts. This construct captures students’ capacity to sustain prosocial engagement despite exposure to hostility or lack of explicit appreciation, supported by intrinsic satisfaction derived from acting in line with their values. This extends current work on digital resilience, which has focused primarily on coping with victimization, by explaining how active prosocial agency can be maintained under adverse or indifferent online conditions. Practically, the findings hold implications for educators and platform designers. For moral and civic education, the results suggest that supporting authentic online prosociality involves cultivating empathy, stable moral values, and a sense of social responsibility, as well as explicitly addressing the emotional demands of helping in public, contested digital spaces. Programs that build digital prosocial resilience (e.g., teaching strategies for managing backlash and coping with invisibility of impact) may help university students continue to engage constructively online. For platform designers, the study underscores the value of affordances that enable users to help in flexible, low‑friction ways, allow for varying degrees of anonymity, and highlight impact without overemphasizing metrics of public approval, thereby protecting users while fostering intrinsically motivated, community‑oriented engagement. Limitations and future directions While this study offers insights into online prosociality, its limitations must be acknowledged to contextualize the scope of the findings. First, our sample consisted exclusively of university students (emerging adults) in Hong Kong. While this developmental stage is critical for identity exploration and digital engagement [7], the motivations and behaviors observed here may not fully extrapolate to younger adolescents or ‘youth’ as a broader category. For instance, the competency required for digital allyship may be less developed in younger cohorts. Furthermore, the specific cultural context of Hong Kong (characterized by high digital density and a unique civic climate) may shape how these behaviors manifest. Future research should therefore examine whether these typologies persist across different cultural settings and broader developmental stages, including early and middle adolescence. Second, our hybrid deductive-inductive thematic analysis provided rich narrative depth but inherently lacks quantitative generalizability. We did not measure the frequency of these behaviors or statistically test the causal pathways between motivations and psychological outcomes. Additionally, reliance on retrospective self-reports in focus groups may introduce recall or social desirability biases, although our focus on “shared meaning” rather than individual frequency mitigates this to some extent. Also, the retrospective self-report cannot help us determine the causal mechanism between motivation and psychological outcomes. Future studies would benefit from mixed methods designs or real-time data capture (e.g., ecological momentary assessment) to validate the prevalence and psychological impact of the typologies identified here. Conclusion This study examined how university students in Hong Kong engage in online prosocial behavior, revealing five interrelated forms of helping whose mechanisms are shaped by the affordances of digital environments. Across these typologies, students framed their actions as grounded in internal motivations rather than in compliance or the pursuit of recognition. At the same time, they navigated a complex emotional landscape in which meaningful connection and purpose coexisted with stress, backlash, and feeling overlooked. The emergence of digital prosocial resilience highlights that sustaining constructive engagement online requires not only opportunities to help, but also internal resources to cope with hostility and indifference. Although based on a qualitative sample of emerging adults within a single cultural context, the findings offer a useful conceptual lens and practical starting point for educators and designers seeking to foster authentic, value‑driven, and resilient forms of prosocial participation as social life becomes increasingly digital. Supplementary Information Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2. Supplementary Material 3. Supplementary Material 3.