Examining selective exposure and disaster information literacy among Generation Z Instagram users

Indonesia is globally recognized as one of the most disaster-prone nations due to its location within the Pacific Ring of Fire and its acute exposure to extreme climatic variability. These geographical conditions give rise to recurring hazards—such as earthquakes, volcanic eruptions, floods, and landslides—that affect millions of people annually (Ayuningtyas et al., 2021; Khumairoh et al., 2021). Data from the National Disaster Management Authority (BNPB) recorded 5,400 disaster events in 2023, representing a substantial increase over the preceding year (Rosyida et al., 2023). The Sumedang earthquake on May 21, 2025, a 3.7-magnitude event, resulted in casualties and infrastructure damage, thereby underscoring the structural vulnerabilities of Indonesian society and the exigency for systematic disaster mitigation (Syafarudin, 2025). Such conditions necessitate robust disaster information literacy, defined as the capacity to critically access, evaluate, and utilize disaster-related data for informed decision-making, enhanced preparedness, and community resilience (Annapisa, 2019). Disaster information literacy transcends technical competencies to encompass the critical evaluation of source credibility, contextual relevance assessment, and the practical application of information toward protective measures. The Sumedang earthquake exemplifies this dynamic; local media were required to deliver both rapid updates and explanatory content to bolster preparedness, demonstrating how disaster communication integrates immediate response with preventive education. The proliferation of social media has fundamentally transformed disaster communication landscapes, shifting from linear broadcast models toward participatory, networked ecosystems (Joseph et al., 2018; Reuter et al., 2018; Seneviratne et al., 2024; Vollmer et al., 2025). Platforms such as Instagram enable local media to facilitate real-time dialogue, audience feedback, and content redistribution. This evolution mirrors global trends where digital platforms facilitate bidirectional communication, community-generated content, and collective sense-making during crises (Zhang et al., 2019). Within Indonesia’s disaster-prone context, local media occupy a critical position due to their cultural proximity, linguistic accessibility, and community embeddedness (Annapisa, 2019; Simamora et al., 2023). Unlike national outlets that prioritize broad coverage, local media emphasize proximity, cultural resonance, and accessibility, ensuring their messages are directly meaningful to affected communities. Recent studies demonstrate that Instagram has emerged as a primary platform for disaster communication among younger demographics, enabling rapid information dissemination, visual storytelling, and interactive engagement (Guidry et al., 2017; Salonen & Laaksonen, 2023; Vázquez-Herrero et al., 2019). Features such as Instagram Stories extend communication beyond linear dissemination, positioning local media as mediators that influence immediate crisis responses and broader collective interpretations of disaster events. Generation Z, born between 1997-2012, represents the most digitally native cohort, characterized by mobile-first media consumption, visual information preferences, and participatory communication practices (Evita et al., 2023; Reinikainen et al., 2020; Seemiller & Grace, 2017). According to the Association of Indonesian Internet Service Providers (APJII) in 2024, 51.9% of this group access news primarily through Instagram, illustrating a shift toward visual-oriented and mobile-first consumption. Their engagement with disaster information is shaped by digital literacy capacities that influence how they assess accuracy, relevance, and practical utility (Hatlevik et al., 2015; Palumian et al., 2023; Reddy et al., 2020). Their selective engagement can be understood through Selective Exposure Theory (Winter & Krämer, 2016), which posits that individuals prefer content aligning with psychological needs, message relevance, and social identity. This theory distinguishes between attitudinal selective exposure, where individuals prioritize content that offers comfort and reassurance, and partisan selective exposure, where social or group identity determines content preferences. Mechanisms such as cognitive dissonance reduction, confirmation bias, and affective motivation help explain why audiences often avoid conflicting or distressing disaster information (Frimer et al., 2017; Matthes et al., 2019; Metzger et al., 2020). Evidence suggests that while Generation Z exhibits highly selective behavior, they do not entirely reject contradictory information if it provides practical value, such as mitigation advice or safety instructions. This complements Uses and Gratifications Theory (UGT), which conceptualizes audiences as active agents who turn to media to satisfy cognitive, affective, and social needs. For Generation Z, Instagram serves both as a platform for rapid news access and a space for reassurance, solidarity, and a sense of collective belonging. The distinction between sought gratifications and obtained gratifications is central to explaining why certain content resonates while others fail to meet expectations (Alhabash & Ma, 2017; Sundar & Limperos, 2013). In a disaster context, the framework suggests that Generation Z engages with information on Instagram not only to stay informed but also to seek reassurance, solidarity, and the affirmation of belonging. Empirical studies corroborate this depiction. Wang (2021) found that Instagram’s visual and interactive features significantly shape news practices among university students, a demographic representative of Generation Z. Zimmermann et al. (2023) further showed that Instagram influencers are often judged as more credible than traditional outlets, highlighting the importance of authenticity and interpersonal trust. Liu et al. (2021) demonstrated that Generation Z exhibits information overload and selective discontinuance behaviors during prolonged crisis exposure, suggesting that sustained engagement requires careful balance between information provision and emotional well-being. Additionally, recent research indicates that Generation Z demonstrates health information avoidance behaviors when content triggers anxiety or cognitive dissonance (Howell & Shepperd, 2016), underscoring the importance of affective considerations in information design. From an institutional perspective, Situational Crisis Communication Theory (SCCT) (Coombs, 2022; Coombs & Holladay, 2015) provides a framework for evaluating communication strategies during disasters. SCCT emphasizes the importance of transparency, empathy, and timeliness to maintain public trust, particularly when media organizations fall within the “victim cluster” where credibility is paramount. For Generation Z, responsiveness and authenticity are central to sustained engagement, while mismanaged communication risks amplifying misinformation and increasing anxiety (Shahbazi & Bunker, 2024). By integrating these three theoretical frameworks—Selective Exposure Theory, Uses and Gratifications Theory, and Situational Crisis Communication Theory—this study advances a comprehensive understanding of disaster communication through the lens of disaster information literacy. However, research on selective exposure remains disproportionately focused on political communication (Baltezarević & Baltezarević, 2022; Winter & Krämer, 2016; Zhu et al., 2024), leaving the disaster context less explored from an information literacy perspective (Houston et al., 2015; Liu et al., 2021; 柳静, 2025). To address this gap, this study investigates three primary research questions: (1) How do psychological factors shape Generation Z’s selection of earthquake information disseminated by local media on Instagram within the context of disaster information literacy? (2) How do the message features of local media on Instagram influence Generation Z’s information choices? (3) How do the social context and media strategies influence Generation Z’s engagement with earthquake information on Instagram? The findings are expected to enrich scholarly debate on the interplay between communication practices, audience selectivity, and crisis communication, while generating practical insights to strengthen digital disaster communication in disaster-prone societies through the reinforcement of disaster information literacy.

This study adopts a qualitative case study approach to investigate how local media on Instagram disseminate earthquake information and how Generation Z selectively engages with such content in Sumedang. The case study framework is particularly appropriate as it situates the phenomenon within a concrete crisis context, allowing for a detailed examination of the relationship between message production by local media and the selective practices of young audiences. Rather than offering a descriptive account of the disaster event, this approach enables an interpretative analysis of the social, cultural, and cognitive dynamics that influence how information is filtered, trusted, or rejected in a social media environment (Yin, 2018). The qualitative orientation is based on robust epistemological considerations, acknowledging that the complexities of crisis communication cannot be adequately captured through quantitative measures that often emphasize frequency counts or general consumption patterns (Creswell, 2017; Denzin & Lincoln, 2018). Qualitative inquiry facilitates the exploration of subjective meanings, lived experiences, and negotiated practices underlying Generation Z’s selective exposure to disaster information. Anchored in the constructivist paradigm, this study employs in-depth interviews, focus group discussions, and visual discourse analysis to identify the implicit meanings embedded in communication practices and audience behavior. The case study centers on the dissemination of information regarding the Sumedang earthquake that occurred on June 31, 2025, covering the emergency and recovery phases until November 5, 2025. This case was selected due to its high communicative intensity, where local media were required to balance speed, accuracy, and credibility while responding to a crisis that directly affected their own community. The research focuses on the Instagram accounts of Sumedang-based outlets such as @radarsumedang, @inimahsumedang, @tahuekspress, and @sumedangekspresnews. Both the accounts and the timeframe were purposively selected for the richness of interactions they generated between producers and audiences, especially during the transition from emergency alerts to post-crisis recovery (Palinkas et al., 2015; Patton, 2015). To collect data comprehensively, the study implemented semi-structured interview guides and Focus Group Discussion (FGD) protocols designed across three analytical domains: psychological, message-related, and social, reflecting the dimensions of selective exposure in the context of disaster information literacy (Kallio et al., 2016; Krueger, 2015). This design ensures that the exploration of communication remains closely aligned with the overarching research questions. A pilot study was first conducted with a small group of respondents to refine the clarity, flow, and sensitivity of the questions, ensuring suitability for the communication context (Malmqvist et al., 2019). Interviews and FGDs recorded with informed consent and transcribed verbatim to maintain data accuracy and minimize interpretative distortion. Furthermore, a visual discourse analysis of digital artifacts was conducted, encompassing 45 Instagram posts (23 from @radarsumedang, 12 from @inimahsumedang, 6 from @tahuekspress, and 4 from @sumedangekspresnews), including Stories, feed posts, captions, and comments. These artifacts were systematically documented through screenshots and metadata collection, ensuring that both the content and the digital context were preserved (Altheide & Schneider, 2016; Rose, 2023). This triangulation of production practices, audience reception, and interactional dynamics is essential for analyzing the Sumedang earthquake coverage as a holistic communication phenomenon within the disaster information literacy framework. Semi-structured interviews provided detailed insights into participants’ motivations, emotions, anxieties, and content preferences, thereby clarifying why certain earthquake information was trusted, avoided, or prioritized. The interview framework was informed by Selective Exposure Theory, allowing for a deductive exploration of existing constructs, while its flexible format opened space for the inductive identification of emerging themes (Braun & Clarke, 2019; Miles et al., 2018). Complementing this, FGDs generated collective reflections on the role of peer influence, group norms, and content-sharing practices, offering a broader perspective on how Generation Z engages with disaster information in socially embedded contexts. Visual discourse and content analysis were performed on a purposive sample of Instagram content, including posts, Stories, captions, short videos, and public comments published between June 31, 2025, and November 5, 2025. Sampling criteria prioritized content containing mitigation information, resulting in high audience engagement (>100 likes or >20 comments), or triggering public debate (Krippendorff, 2022; Neuendorf, 2020). This allowed for an analysis of both the production side of communication (message framing, presentation style, utility, and credibility markers) and the reception side (comments, interactivity, and feedback). The thematic analysis process combined deductive and inductive approaches, moving systematically from data familiarization and coding to the development and refinement of cross-cutting themes using NVivo 12 software (Braun & Clarke, 2019; Nowell et al., 2017). Credibility was enhanced through the triangulation of sources and methods, while dependability and confirmability were strengthened through member checking, peer debriefing, and transparent audit trails. Data validity was achieved through the application of four interrelated criteria (Korstjens & Moser, 2018). Credibility was maintained through cross-method triangulation and participant verification of findings, ensuring the research interpretations accurately reflect the communicative reality. Transferability was enhanced by providing thick descriptions of the Sumedang context and the demographic characteristics of the Generation Z audience, allowing readers to evaluate the applicability of findings in comparable communication settings. Dependability was secured through systematic and transparent documentation of analytical procedures, resulting in an audit trail that demonstrates methodological consistency. Confirmability was supported by systematically archiving data and maintaining a reflexivity journal documenting the researcher’s assumptions, positioning, and potential biases. The study was conducted in strict adherence to ethical standards of qualitative inquiry, with a specific focus on the protection of participants within the sensitive context of disaster communication. Informed consent was obtained from all participants, with additional parental consent secured for minors. The confidentiality of communication-related data was ensured through anonymization using alphanumeric codes and encrypted storage to prevent unauthorized access. Participants retained the right to withdraw at any time without penalty. Ethical protocols were formally reviewed and approved by the institutional ethics committee, guaranteeing compliance with academic standards and securing participant protection throughout the study. Reflexivity is treated as a central methodological principle in this study. As researchers who also experienced the Sumedang earthquake, we acknowledge that our positionality may have influenced data interpretation. This insider status provided valuable contextual understanding but also required conscious efforts to distinguish between personal experiences and participants’ perspectives. Throughout the research process, we maintained reflexive journals documenting our assumptions, emotional responses, and interpretative decisions. Regular peer debriefing sessions were conducted to challenge our interpretations and ensure analytical rigor. This reflexive approach enhances the trustworthiness of findings by making visible the interpretative processes underlying our analysis (Berger, 2015; Dodgson, 2019).

Evidence from in-depth interviews and focus group discussions indicates that Generation Z’s engagement with earthquake information on local Instagram accounts is influenced by a combination of demographic factors and underlying psychological dynamics. Participants aged 15–18, primarily students from SMAN 1 Sumedang, reported more frequent use of @inimahsumedang, whereas 19-year-olds tended to prefer @radarsumedang. This variation is not purely age-related but reflects social identity construction and perceived cultural affinity with specific media outlets (Reinikainen et al., 2020; Seemiller & Grace, 2017). Gender also shaped preferences; female respondents engaged more with @inimahsumedang, perceiving it as more approachable and emotionally relevant. These findings highlight that media choice during crises is informed by perceptions of credibility, relevance, and connection rather than being random or incidental within the framework of disaster information literacy. While credibility serves as the initial filter for content selection, sustained engagement is primarily driven by trust, emotional states, and intrinsic motivations. Practical needs—such as guidance on safe locations or mitigation procedures—are closely linked to affective needs for reassurance and anxiety reduction (Ren & Yan, 2022; Stainback et al., 2020). As one respondent observed: “I trust @radarsumedang more because information is uploaded quickly and usually matches what is happening on the ground” (Interview Data, 2024). Another noted: “When I see a post clearly showing safe locations, I feel calmer, so I visit that account more often” (Interview Data, 2024). These accounts demonstrate that emotional reassurance is as vital as factual accuracy in shaping selective engagement with disaster information within the information literacy framework (Howell & Shepperd, 2016). Focus group discussions also revealed diversity in preferred content consumption formats. Many participants favored Instagram Stories for their speed and accessibility during moments of heightened uncertainty, with one commenting: “Stories usually provide immediate updates, so I check them first before the feed” (Interview Data, 2024). Others, however, preferred feed posts, which they deemed more reliable for obtaining detailed narratives and understanding the broader impact of the earthquake alongside recommended responses (Salonen & Laaksonen, 2023; Vázquez-Herrero et al., 2019). This diversity illustrates that selective exposure among Generation Z is not uniform but is shaped by temperament, social identity, and situational demands within the context of disaster information literacy. Overall, these findings underscore that earthquake communication must be understood not as a linear act of information transmission but as a dialogic process where audiences employ psychological selectivity to filter and interpret content (Reuter et al., 2018; Simamora et al., 2023; Vollmer et al., 2025). This supports the principles of Selective Exposure Theory (Winter & Krämer, 2016) while extending its application to disaster contexts by emphasizing the roles of emotional reassurance and practical motivation. For local media, the implications are clear: effective crisis communication requires more than speed and accuracy; it demands strategies that foster trust, provide accessible practical guidance, and tailor formats to the audience’s mood (Coombs, 2022). Demographic and psychographic segmentation is therefore essential to reflect the heterogeneity of Generation Z, positioning psychological dynamics as a central explanatory factor and a practical entry point for enhancing credibility, preparedness, and resilience in digital disaster communication through the reinforcement of disaster information literacy (Annapisa, 2019; Zhang et al., 2019). A visual discourse analysis of local Instagram accounts (@radarsumedang, @inimahsumedang, @tahuekspress, @sumedangekspresnews) from June 31, 2025, to November 5, 2025, identified distinct patterns in message features that shape Generation Z’s selective exposure. Triangulating post documentation with interview and FGD data revealed that participants aged 15–18 favored feed posts containing infographics and maps, which they described as “clear, concise, and saveable for later” (Interview Data, 2024). In contrast, 19-year-old participants gravitated toward @radarsumedang, which they perceived as “faster at updating and generally accurate” (Interview Data, 2024). This distinction suggests that audience selectivity is influenced not only by message format but also by psychological needs for credibility, relevance, and a perceived sense of connection with the media within the context of disaster information literacy (Wang, 2021; Zimmermann et al., 2023). Message format is central to engagement patterns. Instagram Stories were widely regarded as the most immediate source of updates because they “appear right at the top of the screen, making them easier to access during a panic” (FGD Data, 2024). Stories thus serve as rapid alerts, while feed posts offer extended narratives that provide contextualization, verification, and detailed information on impacts and mitigation strategies (Salonen & Laaksonen, 2023; Vázquez-Herrero et al., 2019). This demonstrates how digital platform affordances—specifically the speed of Stories and the depth of the feed—shape selective engagement during crisis moments. Credibility and the comprehensiveness of information are equally decisive. Older participants valued longer feed posts for their comprehensive details, particularly when they included practical guides such as safe location identification, which one respondent claimed made them “feel more at peace” (Interview Data, 2024). This convergence of cognitive needs for practical instruction and affective needs for reassurance is consistent with Uses and Gratifications Theory (Blumler, 1979; Sundar & Limperos, 2013), which views audiences as active agents choosing content to satisfy cognitive, affective, and social motivations (Wang, 2021; Zhang et al., 2019). Similarly, FGDs highlighted variations in preferences, with some participants favoring visual aids like maps and infographics to simplify complex information, while others rejected sensationalist captions for triggering unnecessary fear (Interview Data, 2024). Interactivity further reinforces credibility and trust. Participants often turned to public comments to confirm accuracy, noting that “if many comments say the same thing as what is happening on the ground, I become convinced it is true” (Interview Data, 2024). Practices such as reposting and sharing Stories enhance content circulation, particularly when the information is deemed credible and relevant (Seneviratne et al., 2024; Shahbazi & Bunker, 2024). These findings hold practical implications: effective crisis communication requires more than speed and accuracy; it must also address the psychological needs of the audience. A hybrid communication strategy is recommended, combining Stories for breaking updates (within 15 seconds of event occurrence) with feed posts and short videos for verification and mitigation (follow-up within 1-2 hours). Editorial practices should prioritize transparency, clarity, and calming language, as participants warned that “if a post is too scary, it just confuses and frightens us” (FGD Data, 2024). Given the heterogeneity of Generation Z, demographic and psychographic segmentation is equally critical. Ultimately, effective digital crisis communication relies on the integration of speed, narrative depth, practical utility, and interactivity to meet cognitive and affective needs within the context of strengthening disaster information literacy (Guidry et al., 2017; Hutabarat et al., 2023; Simamora et al., 2023; Takahashi et al., 2015). The findings of this study indicate that Generation Z’s engagement with earthquake-related information on Instagram during the Sumedang crisis was heavily shaped by social context. Access to information is not merely an individual activity but is deeply mediated through interactions with peers, family members, and school communities. One participant remarked: “I usually double-check news posts when my classmates send links or screenshots to our WhatsApp group” (Interview Data, 2024). Focus group discussions further emphasized that peer recommendations, reposts from trusted community accounts, and collective conversations serve as vital mechanisms for validating information (Liu et al., 2021; Takahashi et al., 2015; Zimmermann et al., 2023). As one participant explained: “If my friend reposts from @radarsumedang, I usually trust it more” (FGD Data, 2024). Such practices illustrate that disaster news consumption forms part of a broader collective validation process, reinforcing the argument that digital disaster communication during a crisis functions simultaneously as an information channel and a platform for shared meaning-making within the framework of disaster information literacy. Local media organizations were found to adapt their communication strategies in response to these audience dynamics. For instance, @radarsumedang emphasized speed and factual accuracy in its real-time updates, while @inimahsumedang adopted a more emotive style, prioritizing humanitarian concerns and narratives of community solidarity (Annapisa, 2019; Simamora et al., 2023). As observed by one participant: “If I want the latest updates quickly, I check @radarsumedang, but if I want stories from residents, I look at @inimahsumedang” (Interview Data, 2024). Visual discourse analysis of posts from June 31, 2025, to November 5, 2025, revealed that content combining evacuation maps, mitigation advice, and empathetic storytelling was the most widely shared. This supports the principles outlined by Situational Crisis Communication Theory (Coombs, 2022; Coombs & Holladay, 2015), which identifies transparency, empathy, and timeliness as crucial elements of crisis response. As another participant reflected: “I prefer accounts that use calming language; it feels more caring and authentic” (Interview Data, 2024). Strategies that blend practical utility with reassurance and solidarity are therefore most effective in fostering audience engagement in the context of disaster information literacy (Salonen & Laaksonen, 2023). The research also identified interactive practices as central to Generation Z’s engagement with disaster communication. Participants frequently used the comment section as a means of evaluating post accuracy, often comparing it with lived experiences shared by other users (Seneviratne et al., 2024; Shahbazi & Bunker, 2024). One explained: “If many comments say the situation is the same as what is on the ground, I feel more certain the post is correct” (Interview Data, 2024). Posts highlighting community support and survivor resilience similarly encouraged participants to share content as an expression of solidarity (Houston et al., 2015; Reuter et al., 2018). As another noted: “If a post has many comments where people support each other, I feel calmer and usually share it” (Interview Data, 2024). Conversely, hyperbolic or sensationalized posts were rejected, as participants feared they might trigger panic. These practices reveal that young audiences actively curate disaster information by weighing cognitive, affective, and social considerations, thereby reinforcing the argument that digital disaster communication is increasingly collaborative and dialogic in nature (Shahbazi & Bunker, 2024; Simamora et al., 2023). The implications of these findings for crisis communication are significant. Media organizations must expand their focus beyond speed and factual accuracy, ensuring that content framing and dissemination strategies are responsive to the social contexts in which audiences engage. Demographic and psychographic segmentation is essential, given that Generation Z is a heterogeneous group with diverse needs—ranging from those who prioritize factual accuracy to others who value emotional resonance or social validation. Ultimately, the effectiveness of crisis communication in a digital environment rests on the dynamic interplay between social context, institutional communication strategies, and audience engagement patterns, which collectively shape selective exposure to disaster information on Instagram within the framework of strengthening disaster information literacy. This research demonstrates that digital disaster communication functions not only as a vehicle for conveying information but also as a communicative space where psychological needs, message design, and social dynamics intersect to influence Generation Z engagement. These findings extend existing literature—which often prioritizes political communication or mainstream media contexts (Baltezarević & Baltezarević, 2022; Winter & Krämer, 2016; Zhu et al., 2024)—by showing how Selective Exposure Theory (Winter & Krämer, 2016) Uses and Gratifications Theory (Blumler, 1979; Sundar & Limperos, 2013), and Situational Crisis Communication Theory (Coombs, 2022) can be applied in tandem to explain young audiences’ selective interaction with crisis-related content. In the case of the Sumedang earthquake, the decision to follow specific local media accounts was not incidental but reflected an interplay between communication strategies, emotional reassurance, and socially embedded information validation processes within the disaster information literacy framework. From a communication perspective, the research highlights that credibility, speed, and clarity of information underpin the trust of young audiences. Media outlets such as @radarsumedang, which prioritize speed and factual accuracy, attract audiences seeking timely updates, while @inimahsumedang, with its emotionally resonant narratives and emphasis on community solidarity, engages those who value empathy and human-centered storytelling. Visual discourse analysis further revealed that message formats carrying practical value—such as evacuation maps, mitigation infographics, and empathetic explanations of local experiences—are highly effective in capturing attention and reducing uncertainty. This supports the argument by Salonen & Laaksonen (2023) and (Vázquez-Herrero et al. (2019) that message style not only facilitates communication flow but also shapes the perceived credibility of digital communication in the context of disaster information literacy. The study also confirms the significance of social context in disaster communication. Interviews and FGDs showed that Generation Z engagement is shaped as much by peer interaction, family influence, and school-based communities as by individual motivation. Practices such as peer recommendations, reposting content from trusted community accounts, and discussions in comment threads create a shared validation ecosystem that reinforces or undermines information credibility. This resonates with the argument by Shahbazi & Bunker (2024) and Seneviratne et al. (2024) that digital disaster communication has become increasingly dialogic and collaborative, with audiences actively participating as curators and co-constructors of meaning rather than passive recipients within the disaster information literacy framework. The implications of these findings for local communication strategies are substantial. Effective disaster communication in a digital environment requires a hybrid model that combines the speed of real-time updates with the narrative depth of explanatory content and the interactivity provided by social platforms. Sensitivity to demographic and psychographic diversity is also vital, as some members of Generation Z prioritize factual precision while others value emotional reassurance or social affirmation. By integrating these dimensions, local communication can move beyond the timely delivery of information to also foster disaster literacy and contribute to the building of collective resilience in crisis-prone communities through the strengthening of disaster information literacy. Furthermore, this study makes explicit theoretical contributions by extending Selective Exposure Theory to identify “emotional certainty” as a novel motivation beyond traditional cognitive dissonance reduction. Generation Z seeks not only information that confirms existing beliefs but also content that provides emotional reassurance during crisis uncertainty. The research enriches Uses and Gratifications Theory by demonstrating that gratifications are situational and dynamic during disasters, shifting from cognitive needs (accuracy) to affective needs (reassurance) as crises evolve. Finally, the study advances Situational Crisis Communication Theory by emphasizing that credibility in social media contexts is co-constructed through collective social validation mechanisms such as comment verification and peer endorsement, rather than solely determined by institutional authority. From a disaster information literacy perspective, this research underscores the critical role of librarians and information professionals in strengthening community preparedness. Libraries can serve as trusted intermediaries by curating verified disaster information, conducting digital literacy workshops focused on source evaluation during crises, and collaborating with local media to develop information literacy campaigns. Educational institutions should integrate disaster information literacy into curricula, teaching students not only how to access information but also how to critically evaluate credibility, recognize misinformation, and apply information toward protective actions. These interventions are essential for transforming Generation Z from passive information consumers into active, critical participants in disaster communication ecosystems. However, this study also identifies critical limitations of selective exposure in disaster contexts. While selective engagement protects individuals from information overload and anxiety, it may also create echo chambers where contradictory but important safety information is ignored. Participants’ preference for emotionally reassuring content could lead to avoidance of uncomfortable but necessary warnings about aftershocks or evacuation orders. This tension between psychological comfort and comprehensive information access represents a significant challenge for disaster communication practitioners, requiring strategies that balance emotional support with the delivery of potentially distressing but life-saving information (Nguyen & Chung, 2019; Ren & Yan, 2022).

This research examines how Generation Z engages with earthquake crisis information on Instagram through qualitative analysis of Sumedang earthquake coverage, revealing that selective exposure to disaster information is shaped by three interrelated dimensions: psychological dynamics, message design, and social context within a disaster information literacy framework. The study demonstrates that trust, affective reassurance, and practical motivations intersect with message format affordances, where Instagram Stories provide immediacy, feed posts offer narrative depth, and infographics enhance clarity, while social validation practices including peer recommendations, comment verification, and editorial strategies balancing speed with empathy reinforce engagement patterns that are situational, heterogeneous, and dialogic. These findings extend Selective Exposure Theory by identifying emotional certainty as a novel motivation beyond cognitive dissonance reduction, enrich Uses and Gratifications Theory by demonstrating situational gratifications that shift dynamically across disaster phases, and advance Situational Crisis Communication Theory by emphasizing collective social validation as a credibility mechanism in social media contexts. The qualitative orientation captures dimensions often overlooked in quantitative research, particularly the role of emotion in format preference and the significance of interactive features for collective meaning-making within disaster information literacy frameworks. Practically, the findings emphasize the need for hybrid communication strategies that integrate the speed of Stories with the narrative depth of feed posts, supported by visual tools such as infographics, evacuation maps, and safety checklists, while balancing timeliness and accuracy with empathetic framing that reassures rather than sensationalizes. Libraries and information professionals emerge as critical intermediaries who can curate verified disaster information, conduct digital literacy workshops focused on source evaluation during crises, and collaborate with local media and educational institutions to develop comprehensive information literacy campaigns that teach students to critically evaluate credibility, recognize misinformation, and apply information toward protective actions. However, practitioners must recognize that selective exposure may create echo chambers where contradictory but important safety information is ignored, necessitating strategies that balance emotional support with potentially distressing but life-saving information to ensure psychological comfort does not compromise comprehensive information access. This study is limited by its single case focus on one earthquake event in Sumedang and one social media platform, which constrains the generalizability of findings to other disaster types, geographical contexts, or demographic groups beyond Generation Z. Future research should explore diverse disaster scenarios involving varied demographics, integrate quantitative approaches to test relationships identified qualitatively, conduct longitudinal studies tracking information behavior changes across different disaster cycle phases, compare selective exposure patterns across multiple social media platforms with different affordances, and employ experimental designs to test causal effects of specific message features on selective exposure behaviors within disaster information literacy frameworks.