A descriptive-analytical design with a predictive focus was deemed appropriate for the study’s goal of analyzing psychosocial aspects, risk perception, and pilgrims’ crowd behavior in holy places. This approach allowed for the study of the phenomenon as it occurred in reality and of the relationships between its variables without researcher intervention.

To meet the study goals, all Muslim pilgrims who undertook the Hajj pilgrimage at Mina, Muzdalifah, Arafat, the Jamarat Bridge, and the Grand Mosque in Makkah throughout the study period were included. The group was varied in sex, age, education, country, and Hajj experience. Hajj is a pilgrimage event that attracts international pilgrims, such that its variation allows for studies on pilgrims’ psychology, sociology, risk perception, and crowd behavior. The pilgrim population’s sex, age, education, and culture were represented by 407 male and female pilgrims who partook in the research. The research required people with direct and meaningful Hajj or Umrah experience and who had gone to the holy sites; hence, it utilized purposive sampling. Only those who met the requirements were sampled to ensure data accuracy and relevance. Participants were required to have performed Hajj or Umrah within five years (1442–1446 AH) to guarantee that their experiences were current and aligned with the holy site’s organizational and crowd management procedures.

The sample includes foreigners (i.e., Saudi, Arab, and non-Arab pilgrims) to ensure that the data reflects the Hajj and Umrah culture and demography, pilgrims’ mental, social, and communal traits, and secure the cultural and national variety characteristic of these international pilgrimages—hence upholding the validity and applicability of the findings to the wider pilgrim population. Participation in the study was optional. Hajj and Umrah pilgrims from 1442–1446 AH took a Google Forms quiz instead of seeing the holy sites. This retroactive online technique was employed for logistical and safety reasons, with the downside being lower external validity, as replies may have been based on recalled experiences. The questionnaire’s participant information statement outlined the study’s goals, data confidentiality, and scientific use of the responses, and Digital written informed consent was obtained from all participants prior to participation. Participants were required to review the study information and provide their consent electronically before accessing the questionnaire., as voluntary participation was essential for the study’s goals. Table 1 shows participants’ demographic characteristics. Regarding nationality, 76.4% of sample members were non-Saudis, and the most representative age group was 30–39 years (44.5%), followed by 20–29 years (31.9%), and 40+ years (23.6%). Most of the sample had a bachelor’s degree (59.2%), followed by diplomas (24.3%) and postgraduate studies (16.5%). The highest proportion regarding economic level was medium (70.0%), followed by high (18.7%) and low (11.3%). These characteristics show that most of the sample was non-Saudi (76.4%), has 30–39 years old, holds a bachelor’s degree, and in the middle class. This multinational sample ensures that our study population accurately depicts the heterogeneity of the pilgrim community, supporting cross-national comparisons.

To achieve the study’s objectives, the survey’s questionnaire included items on demographic variables (e.g., sex, age, educational level, and economic status) and three other instruments, which are outlined below.

Psychological and Social Factors Scale

A Psychological and Social Factors Scale based on Cheng et al. (2019) was used to assess the psychosocial factors of pilgrimage. The study team translated the English instrument into Arabic and had an independent bilingual expert back-translate to ensure conceptual comparability. The modified scale comprised 19 items on spirituality (6 items), learning (4 items), sharing (3 items), help (3 items), and dissatisfaction/negativity (3 items). Items were responded on a five-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree), with higher scores implying an increased connection between pilgrims’ experiences and psychosocial characteristics.

Validity and reliability

In a confirmatory factor analysis, the five-dimensional, 19-item scale model fit the data well. The construct validity of the scale was confirmed through all fit indices being below the statistical thresholds. There was convergence between composite reliability and average variance extracted. The expert review established content validity for this research, with 80%–90% rater agreement. Regarding construct validity, Pearson’s correlation coefficients between the scores for the subdimensions and overall scale were significant at the 0.01 level. Cronbach’s alpha coefficients and split-half reliability scores were over 0.80 for all subdimensions and 0.93 for the total scale, demonstrating the scale’s good internal consistency. These data indicate that the scale was reliable and consistent, making it suitable for this study.

Perceived Risk Scale Among Pilgrims

To meet the study’s goals, we first reviewed the available scales in the literature on pilgrims’ risk perception ( Azmi et al., 2021; Cheng et al., 2019; Nautiyal et al., 2025; Sharifi-Tehrani & Esfandiar, 2018), disease risk perception, risk preventive effectiveness, the health belief model, and general risk perception scales. This led the researchers to develop the Perceived Risk Scale among Pilgrims. This is a comprehensive 23-item instrument created for assessing risk perception among Hajj pilgrims within the context of Arab culture and society, with items being divided into the six subdimensions of perceived probability, severity, anxiety, personal readiness, knowledge of risks, trust in institutions, and community support. Items were responded on a five-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree), with higher scores implying a higher risk perception among pilgrims .

Validity and reliability

Expert assessment demonstrated the scale’s content validity through an interrater agreement of over 80%, proving the usefulness of the questions for evaluating risk perception among Hajj pilgrims. Pearson’s correlation coefficients between the scores for the subdimensions and overall scale were significant at the 0.01 level, confirming its construct validity. Exploratory factor analysis found that the data were suitable (KMO = 0.91) and Bartlett’s test of sphericity was significant. Seven variables explained 68.4% of the variation, showing 0.54–0.83 factor loadings. All fit indices in the confirmatory factor analysis were within acceptable limits, verifying the construct validity of the seven-factor model. High internal consistency indicates the scale’s dependability. Cronbach’s alpha values for the subdimensions were 0.81–0.88 and for the whole scale, it was 0.93. The split-half reliability values were also high, with a total scale coefficient of 0.91. These results demonstrated the reliability and internal consistency of the scale, making it suitable for use in this study.

Hajj Crowd Behavior Scale

Abd Rahman et al. (2019) developed the Hajj Crowd Behavior Scale, which measures pilgrims’ cognitive, emotional, and behavioral behaviors at holy sites and was subsequently translated and culturally adapted for the Arab environment. In Arabic, the scale has 52 items divided into the three subdimensions of cognitive (1–18), emotional (19–35), and behavioral (36–52). Items were responded on a five-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree), with higher scores implying a higher occurrence of crowd behavior. Individual scores for each subdimension were added to obtain a total scale score, which indicated low, moderate, or high crowd behavior.

Validity and reliability

An examination by social psychology and crowd behavior experts confirmed the scale’s content validity, with an interrater agreement of 85%, showcasing that the items were culturally acceptable and reflected the three crowd behavior subdimensions. Pearson’s correlation coefficients between the scores for the subdimensions and total scale indicated appropriate construct validity and coherence across the scale dimensions, varying from 0.72 to 0.81 (p < 0.01). Additionally, the scale had high internal consistency, with the Cronbach’s alpha for the cognitive, emotional, and behavioral subdimensions being 0.87, 0.89, and 0.84, respectively, and that of the total scale alpha being 0.91. The split-half reliability coefficients, ranging from 0.81 to 0.86 for subscales and 0.88 for the total score, confirmed the instrument’s robustness and stability. These results indicated that the Scale is a robust and reliable tool for assessing pilgrimage crowd behavior in field-based research.

Data management and statistical analyses were performed using IBM SPSS version 26 and AMOS version 22. The initial data screening process assessed data correctness, identified missing values and outliers, and verified distributional assumptions such as normality. Means, standard deviations, frequencies, and percentages were used to describe the sample and key research variables. Reliability and validity assessments were used to investigate the psychometric qualities of the instruments. Cronbach’s alpha and split-half reliability coefficients were used to assess the internal consistency of the scales. Construct validity was assessed using exploratory factor analysis with principal component extraction and varimax rotation. Confirmatory factor analysis was performed to test the model’s data fit, using standard goodness-of-fit indices, including χ2/df, GFI, AGFI, CFI, TLI, RMSEA, and SRMR.

Pearson’s correlation coefficients were used to examine psychosocial characteristics, risk perception, and crowd behavior. The predictive ability of independent factors was assessed using multiple regression analysis, with crowd behavior as the dependent variable. All statistical tests were conducted at a significance level of p < 0.05 to ensure data interpretation accuracy and rigor.

Figure 1 shows that pilgrims’ risk perception was moderate to high, with an average total scale score of 72.862 and a standard deviation (SD) of 18.286. The subdimension of trust in institutions ranked the highest, with an average of 13.877 (SD = 3.263), indicating pilgrims’ strong self-reported faith in Hajj organizers; the subdimension of personal readiness followed closely (M = 13.344, SD = 3.968), showing that the pilgrims understood the necessity of personal preparation and prevention. Pilgrim anxiety averaged at 9.310 (SD = 3.199). The perceived probability (M = 8.037, SD = 3.343) and severity (M = 7.956, SD = 3.424) subdimensions demonstrated balanced risk perception, while knowledge of risks averaged at 9.742 and community support at 10.597, highlighting the significance of knowledge awareness and solidarity in pilgrims’ safety.

For psychosocial factors, the total score was 64.408 (SD = 13.205), indicating an acceptable balance of scores for psychosocial factors. The spirituality subdimension showed the highest average scores (M = 20.383, SD = 4.876), indicating that the pilgrims drew positive psychological energy from their faith and spirituality. The learning subdimension followed with averages of 13.418 (SD = 2.890) and 11.091 (SD = 2.414), depicting that the pilgrims had prior training and experience in behavioral management amid pilgrimages and related rituals. The scores for help (M = 10.015, SD = 2.533) and dissatisfaction/negativity (M = 9.501, SD = 2.933) were lower, showing individual variances in cooperation and contentment, but within acceptable ranges. The total mean score for crowd behavior during pilgrimage was 213.025 (SD = 22.866).

Figure 2 shows that the cognitive subdimension had the highest mean score (M = 74.779, SD = 8.253), reflecting a high awareness and understanding of pilgrimages. The emotional subdimension followed closely (M = 74.472, SD = 9.304), indicating a strong emotional engagement during pilgrimage. The behavioral subdimension showed a considerably lower average score (M = 63.774, SD = 10.805), suggesting that the transfer of cognitive and emotional feelings into practical behavior may occasionally be influenced by crowds or environmental pressures.

These descriptive findings show that pilgrims have moderate-to-high risk awareness and institutional trust, a balanced psychosocial profile with spirituality as the dominant subdimension, and predominantly positive cognitive and emotional crowd behavior. The behavioral subdimension showed somewhat lower scores, likely because of the physical and environmental constraints of high-density pilgrimage settings.

Table 2 shows that significant relationships (p < 0.01) exist between psychosocial characteristics and pilgrims’ risk perception. Spirituality was moderately correlated with all aspects, including perceived probability (r = 0.440), dissatisfaction/negativity (r = 0.555), and risk perception (r = 0.569), demonstrating how spirituality increases risk awareness.

The learning subdimension was one of the most influential, being associated with high values for most subdimensions, including perceived probability (r = 0.516) and risk perception (r = 0.607), demonstrating the importance of knowledge and learning in cognition. The sharing subdimension was weaker, revealing some non-significant correlations (e.g., r = 0.259 with perceived probability) and meaningful associations, such as with trust in institutions (r = 0.473) and total score (r = 0.450).

The help subdimension showed intermediate and significant associations with most subdimensions, including perceived probability (r = 0.445) and risk perception (r = 0.538), highlighting the importance of pilgrims’ mutual support in raising awareness. Dissatisfaction/negativity was evident and scored high for severity (r = 0.571) and total scores (r = 0.595), demonstrating that unpleasant experiences may increase risk perception. The scores for psychosocial factors had the highest correlation, ranging from the score for personal readiness (r = 0.494) to that of risk perception (r = 0.661), confirming that the total interaction of these subdimensions’ shapes pilgrims’ risk awareness more clearly and deeply.

As shown in Table 3, the spirituality subdimension was associated with the scores for behavioral dimension (r = 0.400) and total crowd behavior (r = 0.414), whereas the learning subdimension was associated with scores for cognitive behavior (r = 0.271) and total crowd behavior (r = 0.399). Psychosocial characteristics and the crowd behavior were marginally linked (r = 0.423). Regarding crowd behavior, the cognitive subdimension was strongly correlated with emotional (r = 0.740) and overall crowd behavior (r = 0.846). The emotional subdimension was substantially correlated with total score (r = 0.835), whereas the behavioral subdimension was correlated with value (r = 0.751). These findings confirm the internal coherence of the Hajj Crowd Behavior Scale and the coherent model of crowd behavior.

Based on multiple regression analysis, the suggested model with risk perception and psychosocial variables as predictors was statistically significant (F = 44.533, p < .001). The model explained 18.1% of the crowd behavior variation (R ² = 0.181; adjusted R ² = 0.177), indicating limited explanatory power. The model’s overall correlation coefficient (R = 0.425) and RMSE (20.749) demonstrated good prediction accuracy. The standardized regression coefficients showed that psychosocial variables significantly predicted crowd behavior (β = 0.391, p < .001). In the model including psychosocial variables, risk perception did not substantially predict crowd behavior (β = 0.049, p = .413). According to Table 4, psychosocial characteristics significantly predicted crowd behavior (β = 0.391, p < .001), but risk perception did not (β = 0.049, p = 0.413). This model suggests that psychosocial dynamics have a more significant effect than risk perception, although the low explained variance (R ² = 0.181) suggests caution when generalizing beyond this group. Demographic variables may function as contextual markers reflecting cultural and resource differences rather than as direct behavioral determinants.

Table 5 shows the results of the intergroup impact analysis of the demographic variables on risk perception, psychosocial factors, and emotional behavior. Regarding nationality, it had a strong significant effect on all dependent variables; significant differences appeared in risk perception (F = 20.781, p < 0.001), psychosocial factors (F = 37.844, p < 0.001), and crowd behavior (F = 19.970, p < 0.001). This suggests that nationality is a crucial variable in explaining the differences between groups. Pertaining to age, there was no significant effect on any of the dependent variables, and the p-values were not significant (all p > 0.05). Regarding educational level, it did not significantly affect any of the dependent variables, as the F values did not reach statistical significance. Economic level showed a significant effect only on psychosocial factors (F = 3.215, p = 0.041), while no effect was observed on risk perception or crowd behavior.

Accordingly, nationality had the greatest impact on the three dependent variables, whereas economic level had little impact on psychosocial elements, and dimensional difference tests revealed no significant differences by age and education for all dependent variables. However, the psychosocial factor showed significant variations according to economic level (low to medium), with the middle income group showing a higher influence. The other factors showed no significant differences.

Pilgrims evaluated the Hajj pilgrimage’s risks as moderate to high in this study. Risk perception was dominated by an influence of trust in institutions and personal readiness, whereas anxiety, perceived probability, and severity matched. Pilgrims tend to be more mentally balanced and spiritual, such that the skillfully organized and spiritually important Hajj may present risks regarded as manageable rather than scary. This finding complements those of previous studies, which found that institutional trust, shared religious identity, and spirituality affect risk evaluations during large religious gatherings. Alnabulsi and Drury (2014) discovered that pilgrims who trusted organizational authorities and strongly identified with the crowd felt safer in large crowds. According to collective pilgrimage studies, a shared spiritual identity and emotional synchrony decrease anxiety and enhance adaptive responses to perceived risk by increasing psychological comfort, social cohesiveness, and emotional stability. Samarkandi et al. (2025) also showed that institutional trust and prior pilgrimage experience moderate pilgrims’ risk perceptions, resulting in greater readiness and compliance rather than fear-based reactions. These findings differ from those of research that found pilgrims underestimating health risks or weakly translating risk awareness into preventive behaviors. Almehmadi et al. (2021) observed that pilgrims were aware of health hazards, but rarely took precautions, while Khan et al. (2016) discovered that first-time pilgrims, particularly those unaccustomed to large crowds, reported increased tension, anxiety, and adjustment concerns. These differences in evidence may owe to divergences in organizational readiness, health communication tactics, pilgrims’ experience levels, and data collection scheduling (relative to important global health crises) among these studies.

These results can be analyzed through the lenses of social identity theory and contemporary crowd psychology. When people see themselves as part of a religious group and trust the legitimate organizational structures, risk perception is controlled by society rather than improved by individuals ( Alnabulsi & Drury, 2014). Durkheim’s collective effervescence theory posits that shared rituals and emotional synchronization offer reassurance, moral regulation, and social order within collective settings. Indeed, spirituality enables individuals to reinterpret peril as a significant and controllable aspect of communal religious experiences, enhancing psychological resilience and fostering balanced crowd behavior in high-density pilgrimage settings ( Alnabulsi & Drury, 2014; Páez et al., 2015).

The study indicated that pilgrims exhibited stronger cognitive and emotional crowd behavior, but lower behavioral activity. Although pilgrims appear alert, understanding, and emotionally immersed in rites, other aspects of pilgrimages such as congestion, limited mobility, and weariness may also affect them. These findings confirm those of earlier studies indicating that shared religious knowledge and emotional synchrony boost pilgrims’ communal awareness and emotional coherence, especially in crowds ( Alnabulsi & Drury, 2014; Wlodarczyk et al., 2020). As described in religious crowd studies, cognitive and emotional convergence precedes behavioral coordination, particularly when the circumstances limit agency; emotional engagement and behavioral cooperation may be higher in less crowded or regulated settings ( Alnabulsi et al., 2018). This is explained by crowd psychology theories that divide cognition, emotion, and behaviors. Collaboration is based on shared identity and emotions, while external factors influence behavior, entailing that situational restrictions—not desire or awareness—can explain the gap observed in the current study between cognitive–emotional participation and behavioral expression, making environmental and crowd control essential to ensure adaptive crowd behaviors.

This study discovered substantial positive relationships among psychosocial characteristics, risk perception, and crowd behavior. Spirituality, learning, help, and psychosocial balance affected the crowd cognitive, emotional, and behavioral adaptations. Therefore, specific psychosocial dynamics can increase cognitive, emotional, and behavioral crowd adaptation during religious pilgrimages like the Hajj. These descriptions complement those of prior studies, where a balanced risk perception in a supportive psychological and spirituality context improves adaptive group behavior over fearful or panicky behavior ( Alnabulsi et al., 2018). In high-density religious meetings, shared identification and emotional synchronization foster collective attentiveness and social norms ( Wlodarczyk et al., 2020). Health-related risk perception research shows that awareness, trust, and social support are more likely to affect cooperative and preventive behavior than risk awareness alone ( Almehmadi et al., 2021). This research supports social identity–based crowd behavior theories that imply shared identity, emotional cohesiveness, and authoritative legitimacy motivate collective action. These findings support theThe idea that psychological resources modify or mediate risk perception and adaptive behavior is supported, albeit formal mediation analyses are needed in the future to corroborate this influence pathway ( Alnabulsi et al., 2018; Wlodarczyk et al., 2020).

Psychosocial factors predicted crowd behavior better in the regression analysis. As a bivariate factor, risk perception was strongly correlated with crowd behavior, whereas psychosocial factors minimized its independent contribution, indicating mediation or suppression. The model explained 18.1% of the crowd behavior variance, with a small impact size (R ² = 0.181). This difference may owe to factors such as Hajj experience, group dynamics, environment, and situational demands. These data suggest substantial psychosocial pathways, not all collective behavior features. Research has shown that risk awareness alone cannot promote adaptive crowd behavior without trust, collective identity, learning, and social support. Researchers show that psychosocial resources enhance community awareness via risk knowledge, not terror ( Reicher et al., 1995; Wlodarczyk et al., 2020), while health and disaster psychology research suggests that risk perception drives action indirectly through mediating variables such as institutional trust, perceived efficacy, and social norms ( Wlodarczyk et al., 2020). Additionally, psychosocial factors may affect risk perception in highly structured religious pilgrimages such as the Hajj, where institutional presence and community connections matter. According to social identity-based and psychosocial crowd behavior theories, community meaning, emotional congruence, and normative expectations influence crowd behavior more than risk assessment, while institutional trust, learning, and solidarity affect adaptive crowd behavior in pilgrimages. These discussions underline how crowd control success requires risk communication and psychological interventions. Since the model in this study explained 18.1% of the variance, future research should uncover additional factors that can influence crowd behaviors ( Alafif et al., 2025; Knoke, 2025; Shah, 2025).

Nationality was the biggest demographic factor affecting risk perception, psychosocial factors, and crowd behavior. Age and education did not significantly affect any dependent variable, whereas economic level only affected some the psychosocial elements. These results find consonance in prior literature, indicating that age and education serve as background characteristics rather than direct predictors of crowd behavior in highly-regulated religious contexts ( Reicher, 2011). Standardized rituals, behavioral norms, and robust institutional governance may mitigate age- and educational-related behavioral disparities in large religious pilgrimages and assemblies, such as the Hajj, leading to more uniform behavior. Studies have indicated that nationality, culture, language, and previous exposure to significant religious events influence risk perception and psychological experiences ( Abd Rahman et al., 2019; Khan et al., 2016), as pilgrims from different nations may have distinct crowd management systems, authority trust, and risk interpretations that may affect their psychological responses and crowd behavior.

Based on the current findings, economic limitations may influence people’s psychosocial factors and cause resource, housing, and comfort inequities that indirectly harm Hajj pilgrims’ emotional well-being and social experiences. Still, religious affiliation and institutional governance may affect crowd behavior more than economic level. These results support demographic factors as contextual indicators, not causal ones. In highly-salient shared identity settings, such as the Hajj pilgrimage, social identity theory states that group-based norms and meanings matter more than demographics ( Norenzayan & Shariff, 2008; Petrica et al., 2024). Additionally, effective crowd management should prioritize culturally sensitive communication and psychological interventions that account for national and cultural diversity above age- or education-based distinctions ( Norenzayan & Shariff, 2008; Reicher, 2011). Many risk psychology and crowd behavior studies also emphasize psychosocial factors over risk perception, corroborating these descriptions findings. Studies on risk psychology and crowd behavior have shown that trust, shared identity, meaning-making, and collective effectiveness influence risk perception and action ( Wlodarczyk et al., 2020; Alnabulsi et al., 2018). Wlodarczyk et al. (2020) specifically proposed a dynamic risk perception model in which identity processes, perceived control, and the social environment influence behavior more than perceived probability or severity. Alnabulsi and Drury (2014) also demonstrated that social norms, institutional trust, and collective meaning are more reliably associated with adaptive conduct than risk perception alone. Psychosocial resources, thus, transform risk awareness into a coordinated, regulated group effort, fostering cooperation rather than fear or avoidance during religious pilgrimages like Hajj.

Crowd science and behavior research support these findings. Drury and Reicher (2000) argued that, in dense crowds, collective organization and social regulation govern behavior more than individual cognition. That is, in structured situations such as the Hajj pilgrimage, similar norms, ritualized mobility, and institutional supervision decrease the behavioral relevance of risk assessment discrepancies. This proposal by Drury and Reicher’s (2000) is therefore consistent with the current findings that psychosocial factors are more proximally associated with crowd behavior than risk perception. Importantly. research on religion and resiliency also supports this study’s emphasis on spirituality. Regarding risk communication, Raj (2015) recommended prioritizing trust, cultural resonance, and shared meaning over technical risk statistics, while Taibah et al. (2018)—in their investigation of Hajj pilgrims’ communication choices—described that culturally resonant and spiritually themed communications outperformed technical danger information. This implies that trust building, culturally relevant messaging, and spirituality framing may improve adaptive crowd behavior more than quantitative risk knowledge.

Despite various contributions, this study is not without its methodological limitations. First, the cross-sectional methodology prevented causal inferences between risk perception, psychosocial factors, and crowd behavior. Second, social desirability or response bias may affect self-reported measurements, particularly in religious and highly controlled contexts, such as the Hajj pilgrimage. Third, the sample varied in country and demography, but it may not completely represent pilgrims’ temporal changes across Hajj seasons or organizational settings.

Regarding this study’s strengths, few empirical investigations have combined risk perception and psychological factors when conducting predictive analyses of crowd behaviors amid the Hajj pilgrimage. The use of valid and culturally tailored assessment tools, along with rigorous multivariate statistical analyses, also increases the dependability and believability of the results, which outline that the incorporation of psychological, social, and spiritual aspects into crowd management intervention designs can lead to more comprehensive strategies. With these qualities, this study advances crowd behavior theory and practice in large religious events.

Among risk perception factors, trust in institutions showed the highest scores. Therefore, crowd management authorities should make efforts to deliver clear, multilingual safety communication. Digital signs at pilgrimage locations and official cell phone alerts may help spread safety information quickly. Personal readiness ranked second among the risk perception subdimensions, entailing that brief pre-Hajj psychoeducational interventions, such as video-based instructional materials or structured workshops on crowd safety, first aid, and personal protective equipment, may improve readiness and reduce stress. Spirituality was the highest-scoring psychosocial subdimension; therefore, risk communication strategies that ground safety behaviors in the Islamic principle of self-preservation may increase pilgrims’ compliance with preventive measures. Finally, because learning is an important psychological feature, scalable digital technologies that supply pre-travel risk education in different languages may help the heterogeneous international Hajj community behave more informedly and adaptively.