Neuroticism, perceived stress, psychological resilience, and pre-dive state anxiety in professional divers: a cross-sectional study

Abstract

Background

Professional divers work in safety-critical environments that require sustained attention, rapid judgment, and stable emotional control. Neuroticism is a vulnerability trait related to negative affectivity, perceived stress reflects appraisal of recent demands, and psychological resilience may be associated with weaker stress-related associations. This study examined how these factors were associated with pre-dive state anxiety in professional divers.

Methods

This cross-sectional study used a de-identified, code-linked, two-component survey design in China from January to February 2026. A total of 300 professional divers completed the baseline web-based questionnaire, and 277 were included in the final analysis after eligibility and data-quality screening. The baseline questionnaire assessed demographic characteristics, neuroticism, perceived stress, and psychological resilience. The State scale of the State-Trait Anxiety Inventory (STAI-S) was completed separately on the day of a scheduled diving task, shortly before the dive. Pearson correlations, adjusted linear regression models, and exploratory interaction analyses were conducted controlling for age, education, and years of diving experience.

Results

Neuroticism was positively correlated with state anxiety (r = 0.77, p < 0.001) and perceived stress (r = 0.29, p < 0.001), whereas resilience was negatively correlated with neuroticism (r = -0.49, p < 0.001) and state anxiety (r = -0.58, p < 0.001). After adjustment, neuroticism was strongly associated with state anxiety (B = 0.81, SE = 0.04, β = 0.78, p < 0.001), and perceived stress was also associated with state anxiety after accounting for neuroticism (B = 0.16, SE = 0.05, β = 0.13, p = 0.001). Resilience did not significantly moderate the direct neuroticism-state anxiety association (B ≈ 0.000, p = 0.837), but it moderated the neuroticism-perceived stress and perceived stress-state anxiety associations (both B = -0.011, p < 0.001). These associations were weaker at higher resilience.

Conclusions

In this convenience sample of Chinese professional divers, higher neuroticism and perceived stress were associated with higher pre-dive state anxiety, whereas greater resilience was associated with lower state anxiety and with weaker selected stress-related associations. The mean STAI-S score was in a low-to-moderate range; therefore, the findings should not be interpreted as showing clinically elevated anxiety or as implying that all pre-dive arousal is maladaptive. Longitudinal and repeated-measures studies are needed to clarify temporal ordering, practical significance, and intervention implications.

Introduction

Professional diving places workers in environments characterized by high task load, environmental uncertainty, and limited tolerance for error. In underwater operations, breakdowns in emotional regulation may impair cognition, compromise task performance, and increase safety risk. Earlier work in scuba and diving medicine has shown that anxiety and panic are meaningful contributors to diving incidents, and experimental studies have suggested that anxiety can worsen underwater psychomotor and cognitive performance [1–4].

Within personality research, neuroticism reflects a relatively stable tendency to experience negative affect, heightened threat sensitivity, and increased emotional reactivity. A large body of work links neuroticism to internalizing problems, especially anxiety and depressive symptoms, and longitudinal meta-analytic evidence suggests that neuroticism is prospectively associated with common mental disorders even after substantial adjustment for baseline symptoms [5–7]. In diving-related contexts, this trait may be particularly relevant because underwater operations combine environmental uncertainty, physiological load, and high performance demands. Emerging evidence in special operation divers has shown that higher neuroticism is associated with both state and trait anxiety, while broader reviews of diver psychology suggest that personality characteristics are relevant to adaptation, behavior, and safety under water [8–10]. Against this background, neuroticism can be understood not only as a general marker of affective vulnerability, but also as a plausible psychological risk factor in occupational diving.

One plausible linking variable between neuroticism and state anxiety is perceived stress. According to transactional stress theory, stress is not determined solely by external demands but also by how individuals appraise those demands and their coping resources [11, 12]. The Perceived Stress Scale was developed to capture the extent to which situations are experienced as unpredictable, uncontrollable, and overwhelming [13]. Perceived stress and state anxiety overlap because both involve negative affect and threat-related appraisal, but they differ in time frame and specificity. Perceived stress reflects a recent appraisal tendency across life or work circumstances, whereas state anxiety reflects a current emotional condition in a specific moment. In a diving context, a diver who has recently perceived demands as uncontrollable or overloaded may enter a scheduled diving task with a lower threshold for appraising that task as threatening. This provides an association-based rationale for examining perceived stress together with neuroticism and immediate pre-dive state anxiety.

Resilience, in contrast, is generally understood as the capacity to adapt successfully under stress or adversity. Classic and contemporary work has emphasized that resilience is not simply the absence of distress; rather, it reflects adaptive psychological resources that help individuals recover, maintain function, and resist stress-related deterioration [14, 15]. Recent systematic reviews and meta-analyses further indicate that resilience-oriented interventions can improve mental health outcomes, although effect sizes vary across settings and populations [16, 17]. From this perspective, resilience may weaken the extent to which neurotic tendencies are translated into stress appraisal, and it may also weaken the extent to which perceived stress is converted into anxiety symptoms.

Although mental health in diving has received increasing attention, the combined roles of personality vulnerability, stress appraisal, and psychological resilience have not been adequately tested in professional divers. Recent moderated mediation studies in other populations have shown that resilience can alter both direct and indirect stress-related associations [18, 19], but comparable evidence remains limited in underwater occupational groups. The present study therefore had three aims. First, we examined whether neuroticism, perceived stress, and psychological resilience were associated with pre-dive state anxiety in professional divers. Second, we examined whether perceived stress contributed additional explanatory information regarding pre-dive state anxiety after accounting for neuroticism and covariates. Third, we explored whether psychological resilience moderated the neuroticism-perceived stress association, the perceived stress-state anxiety association, and the direct neuroticism-state anxiety association. Given the cross-sectional design, these aims were framed strictly as association-based and exploratory rather than causal.

Methods

Study design and setting

This cross-sectional study used a de-identified, code-linked, two-component survey design conducted in China from January to February 2026. Data collection consisted of two linked components. First, participants completed a baseline web-based questionnaire assessing demographic characteristics, neuroticism, perceived stress, and psychological resilience. Second, the STAI-S was administered separately on the day of a scheduled diving task, shortly before the dive, to capture immediate pre-dive state anxiety rather than generalized anxiety in daily life. Both components were completed within the same January-February 2026 data-collection period.

To protect participant privacy while allowing individual-level linkage, participants were not linked by name, telephone number, personal identification number, or other directly identifying information in the analytic dataset. Instead, each participant used a unique non-identifying study code in both components of data collection. This code was used only to match the baseline questionnaire with the corresponding pre-dive STAI-S assessment. After linkage and data-quality screening, the analytic dataset contained no directly identifying personal information. The exact interval between the baseline questionnaire and the STAI-S assessment was not recorded for every participant.

Participants

Participants were recruited using convenience sampling through professional diving-related work units, training groups, and occupational diver networks in China. The target population for the present study was therefore defined conservatively as Chinese professional divers who were currently engaged in diving-related work or training and who were reachable through these professional channels during the study period.

Survey access was provided electronically, and participation was voluntary. Because the invitation was disseminated through electronic professional channels and the exact number of potentially eligible divers who received or viewed the invitation was unknown, a valid response rate could not be calculated. The study did not systematically collect detailed information on specific diving branch, training versus operational task, planned depth, planned duration, solo versus team diving, environmental conditions, or task difficulty.

Inclusion criteria were: 18–50 years old; physically healthy by self-report; and currently engaged in professional diving-related work or training. Exclusion criteria were: no explicit consent; incomplete questionnaire responses; self-reported neurological impairment; and current use of medications or psychoactive substances known to influence neurological function and/or mood status. Eligibility and exclusion criteria were assessed by self-report screening items rather than by independent medical examination or register linkage.

A total of 300 questionnaires were returned. Responses were screened sequentially for consent, eligibility, completeness of key study variables, validity of the study code linkage, and invalid or implausible entries. After excluding 23 questionnaires because of incomplete responses or failure to meet eligibility criteria, 277 participants were included in the final analysis. The final sample included 251 men (90.6%) and 26 women (9.4%).

Ethical approval for this study was obtained from the Naval Special Medical Center (protocol code 2021070201). Participants provided informed consent before participation. The participant information and ethical notice indicated that participants could stop completing the survey if they experienced discomfort and could seek support from occupational medical staff or psychological support services if they experienced marked anxiety or distress. Questionnaire results were not used to make clinical diagnoses or operational fitness decisions. This research was conducted in accordance with the Declaration of Helsinki.

Measures

All scales were administered in Chinese.

Neuroticism (NEO-FFI neuroticism subscale)

Neuroticism was assessed using the neuroticism subscale of the NEO Five-Factor Inventory (NEO-FFI), a 60-item short form of the NEO personality inventories based on the Five-Factor Model. The NEO-FFI contains five domains, with 12 items for each domain, and responses are rated on a 5-point Likert scale ranging from strong disagreement to strong agreement. In the present study, only the 12-item neuroticism subscale was used. Reverse-keyed items were recoded before summation, and higher total scores indicated greater emotional instability, proneness to negative affect, and stress reactivity. The possible score range for the neuroticism subscale in the present study was 12–60. The NEO inventories are among the most widely used instruments for assessing the Big Five traits, and Chinese-language work has supported the psychometric adequacy of NEO-based neuroticism assessment in Chinese samples [20]. The Cronbach’s alpha coefficient of the neuroticism subscale in the present study was 0.83.

State anxiety (STAI-S)

State anxiety was measured with the State subscale of the State-Trait Anxiety Inventory (STAI-S). The STAI is a classic self-report instrument designed to distinguish state anxiety, defined as a transient emotional condition characterized by tension, apprehension, and autonomic arousal, from trait anxiety, which reflects a more enduring disposition. The STAI-S contains 20 items assessing how respondents feel “right now, at this moment.” In the present study, the STAI-S was administered before scheduled diving tasks, so the score was intended to reflect divers’ immediate pre-task anxiety level rather than a generalized long-term anxiety tendency. Each item is rated on a 4-point scale, and after reverse scoring of anxiety-absent items, responses are summed to produce a total score, with higher scores indicating greater current anxiety. The possible STAI-S score range was 20–80. The STAI has been widely used internationally in both clinical and occupational settings, and Chinese versions have shown acceptable validity and reliability in Chinese populations [21, 22]. In this study, the STAI-S score was treated as a continuous variable reflecting pre-dive state anxiety rather than categorized by a screening cutoff. The Cronbach’s alpha coefficient for the STAI-S in the present study was 0.91.

Perceived stress (PSS-14)

Perceived stress was assessed using the 14-item Perceived Stress Scale (PSS-14), originally developed to measure the degree to which individuals appraise situations in their lives as unpredictable, uncontrollable, and overloaded during the past month. The PSS-14 includes seven positively stated items that require reverse scoring. In the original version, items are rated from 0 to 4; in the present survey, the Chinese version was implemented using a 5-point response format (1–5). Item scores were summed after reverse coding so that higher total scores reflected greater perceived stress. The resulting score range in the present study was 14–70. Chinese psychometric evidence supports the use of the PSS in community samples, including the 14-item version [23]. Because of the response-format modification, PSS-14 scores were interpreted as continuous within-sample indicators of perceived stress rather than compared directly with absolute scores from studies using the original 0–4 format. The Cronbach’s alpha coefficient for the PSS-14 was 0.886.

Psychological resilience (RS-25)

Psychological resilience was measured using the 25-item Resilience Scale (RS-25) developed by Wagnild and Young to assess positive personality characteristics that facilitate successful adaptation to adversity. The original RS-25 uses a 7-point Likert format and captures broad resilience-related attributes such as perseverance, equanimity, meaningfulness, self-reliance, and existential aloneness. Chinese adaptations of the Resilience Scale have shown acceptable psychometric properties in Chinese samples [24]. In the present study, the Chinese RS-25 was administered using a 5-point response format (1–5); total scores were obtained by summing all items, with higher scores indicating greater psychological resilience. The score range in the present study was 25–125. Because this response format differs from the original 7-point version, RS-25 scores were interpreted as continuous within-sample indicators rather than directly compared with studies using the original format. The Cronbach’s alpha coefficient for the RS-25 was 0.912.

Covariates

Age (years), education (ordinal-coded as 1 = high school/technical secondary school or below, 2 = junior college, 3 = bachelor’s degree, and 4 = master’s degree or above), and years of diving experience.

Statistical analysis

All analyses were conducted using IBM SPSS Statistics 31.0.2.0 and Hayes’ PROCESS macro version 5.0. Pearson correlations were first used to examine bivariate associations among study variables. Multiple linear regression models adjusted for age, education, and years of diving experience were then used to estimate: (1) the association of neuroticism, perceived stress, and resilience with state anxiety; (2) the association of neuroticism with perceived stress; and (3) the association of perceived stress with state anxiety after additional adjustment for neuroticism and covariates. No missing-data imputation was performed; incomplete responses were excluded during data-quality screening before the final analytic dataset was generated. To improve interpretability, standardized beta coefficients (β) were added for the main regression paths in Table 3, while interaction coefficients in Table 4 are reported in unstandardized form.

Table 3.

Adjusted associations among neuroticism, perceived stress, and state anxiety, with exploratory indirect association estimates

Path	B	SE	β	t	p	95% CI
X→Y total	0.81	0.04	0.78	21.05	< 0.001	0.74, 0.89
X→M	0.27	0.05	0.30	5.23	< 0.001	0.17, 0.37
M→Y	0.16	0.05	0.13	3.49	0.001	0.07, 0.24
X→Y direct	0.77	0.04	0.74	19.42	< 0.001	0.69, 0.85
Indirect association (a × b)	0.04	0.02	—			0.01, 0.08

X neuroticism, M perceived stress, and Y state anxiety. B values are unstandardized regression coefficients; β values are standardized beta coefficients. All models were adjusted for age, education, and years of diving experience. The indirect association estimate is exploratory and should not be interpreted as evidence of causal mediation

Table 4.

Interaction effects of psychological resilience and conditional indirect associations at low, mean, and high levels of resilience

Effect	B	SE	t	p	95% CI
X × W → Y	-0.000	0.002	-0.21	0.837	-0.004, 0.003
X × W → M	-0.011	0.002	-4.45	< 0.001	-0.016, -0.006
M × W → Y	-0.011	0.002	-5.12	< 0.001	-0.015, -0.007
Low W (-1 SD)	0.24				0.154, 0.355
Mean W	0.07				0.032, 0.122
High W (+ 1 SD)	0.00				-0.010, 0.023

X neuroticism, M perceived stress, Y state anxiety, and W psychological resilience. Unstandardized coefficients are shown for interaction terms. Conditional indirect associations were estimated at low (-1 SD), mean, and high (+ 1 SD) levels of resilience using bootstrap confidence intervals based on 5,000 resamples. All models were adjusted for age, education, and years of diving experience. For small interaction coefficients and standard errors, three decimals were retained to avoid rounding values to zero

Because the conceptual framework involved stress-related association patterns, exploratory conditional process analyses were conducted to examine whether psychological resilience moderated the association of neuroticism with perceived stress, the association of perceived stress with state anxiety, and the direct association of neuroticism with state anxiety. Continuous predictors used in interaction terms were mean-centered before analysis. When interaction terms were statistically significant, simple-slope analyses were performed at low (-1 SD), mean, and high (+ 1 SD) levels of resilience.

To characterize whether the cross-sectional indirect association of neuroticism with state anxiety through perceived stress varied across levels of resilience, conditional indirect association estimates were calculated using bootstrap confidence intervals based on 5,000 resamples. Given the cross-sectional design, these estimates were interpreted as exploratory, association-based conditional indirect associations rather than evidence of causal mediation or temporal ordering. The interaction and conditional indirect association analyses were exploratory; no formal correction for multiple testing was applied, and p values from these analyses should therefore be interpreted cautiously. Multicollinearity diagnostics were acceptable, with all variance inflation factors below 2.0. Statistical significance was set at p < 0.05 (two-tailed).

Results

Participant characteristics

A total of 277 professional divers were included in the final analysis. The mean age was 32.89 years (SD = 7.36), and the mean years of diving experience was 3.26 years (SD = 1.55). Most participants were male (90.6%). In terms of education, 18.4% had a high school or technical secondary school education or below, 37.2% had a junior college education, 36.8% held a bachelor’s degree, and 7.6% had a master’s degree or above (Table 1).

Table 1.

Sample characteristics of the study participants (N = 277)

Characteristic	n (%) or Mean ± SD
Male	251 (90.6)
Female	26 (9.4)
Age (years)	32.89 ± 7.36
Education: high school/technical secondary school or below	51 (18.4)
Education: junior college	103 (37.2)
Education: bachelor’s degree	102 (36.8)
Education: master’s degree or above	21 (7.6)
Years of diving experience	3.26 ± 1.55

Descriptive statistics and correlations

Table 2 presents the descriptive statistics and Pearson correlations among the study variables. Neuroticism was positively correlated with state anxiety (r = 0.77, p < 0.001) and perceived stress (r = 0.29, p < 0.001), and negatively correlated with resilience (r = -0.49, p < 0.001). State anxiety was positively correlated with perceived stress (r = 0.37, p < 0.001) and negatively correlated with resilience (r = -0.58, p < 0.001). The correlation between perceived stress and resilience was not statistically significant (r = -0.07, p = 0.250). Given the high neuroticism-state anxiety correlation, this association should be interpreted with caution because shared negative-affect content, same-source measurement, and limited discriminant validity between trait-like vulnerability and immediate state anxiety may have contributed to its magnitude. These findings supported subsequent analyses examining perceived stress as an associated explanatory correlate and resilience as a moderator in the association patterns among neuroticism, perceived stress, and state anxiety.

Table 2.

Descriptive statistics and Pearson correlations among the study variables

Var.	M	SD	1	2	3	4	5	6	7
1. Age	32.89	7.36	—
2. Educ.	2.34	0.86	0.06	—
3. Exp.	3.26	1.55	0.62***	-0.07	—
4. N	26.97	9.09	-0.02	0.02	-0.08	—
5. Anx.	36.86	9.44	-0.12*	-0.06	-0.02	0.77***	—
6. PSS	37.08	8.15	-0.14*	-0.03	-0.03	0.29***	0.37***	—
7. RS	88.18	20.78	-0.00	0.07	-0.04	-0.49***	-0.58***	-0.07	—

Values are Pearson correlation coefficients. Educ. education; Exp. years of diving experience, N neuroticism, Anx. pre-dive STAI-S state anxiety, PSS 14-item Perceived Stress Scale, RS 25-item Resilience Scale. Education was treated as an ordinal variable. * p < 0.05; ** p < 0.01; *** p < 0.001

Adjusted associations among neuroticism, perceived stress, and state anxiety

After adjustment for age, education, and years of diving experience, neuroticism was positively associated with state anxiety (B = 0.81, SE = 0.04, β = 0.78, p < 0.001). Neuroticism was also positively associated with perceived stress (B = 0.27, SE = 0.05, β = 0.30, p < 0.001). When neuroticism and perceived stress were entered simultaneously into the model predicting state anxiety, perceived stress remained positively associated with state anxiety (B = 0.16, SE = 0.05, β = 0.13, p = 0.001), while neuroticism retained a strong association with state anxiety (B = 0.77, SE = 0.04, β = 0.74, p < 0.001) (Table 3). These findings indicate that perceived stress contributed additional explanatory information beyond neuroticism in relation to pre-dive state anxiety.

Moderation by psychological resilience

In interaction analyses, the Neuroticism x Resilience term predicting state anxiety was not statistically significant (B = -0.000, SE = 0.002, p = 0.837). However, the Neuroticism x Resilience interaction predicting perceived stress (B = -0.011, SE = 0.002, p < 0.001) and the Perceived stress x Resilience interaction predicting state anxiety (B = -0.011, SE = 0.002, p < 0.001) were both statistically significant.

Simple-slope analyses showed that the association between neuroticism and perceived stress was statistically significant at low resilience (B = 0.51, p < 0.001) and mean resilience (B = 0.29, p < 0.001), but not at high resilience (B = 0.06, p = 0.448). Similarly, the association between perceived stress and state anxiety was statistically significant at low resilience (B = 0.48, p < 0.001) and mean resilience (B = 0.25, p < 0.001), but not at high resilience (B = 0.03, p = 0.634). Thus, higher resilience was associated with weaker neuroticism-perceived stress and perceived stress-state anxiety relationships, but not with a weaker direct neuroticism-state anxiety relationship.

Conditional indirect association analyses further indicated that the cross-sectional indirect association of neuroticism with state anxiety through perceived stress was strongest at low resilience (estimate = 0.24, 95% CI 0.15 to 0.36), weaker at mean resilience (estimate = 0.07, 95% CI 0.03 to 0.12), and not evident at high resilience (estimate = 0.00, 95% CI -0.01 to 0.02) (Table 4). These findings suggest that resilience was linked to selected stress-related association patterns rather than to the overall neuroticism-state anxiety association. The final statistical model is presented in Fig. 1.

Fig. 1.

Final outcomes model. Coefficients are adjusted for age, education, and years of diving experience. Standardized beta coefficients are shown for main regression paths to aid interpretation, whereas interaction terms are shown as unstandardized coefficients. Solid arrows indicate statistically significant associations. Dashed arrows indicate moderation terms or the non-significant direct moderation term. ** p < 0.01; *** p < 0.001

Discussion

The present cross-sectional study examined how neuroticism, perceived stress, and psychological resilience were associated with pre-dive state anxiety in a convenience sample of Chinese professional divers. Three findings merit particular emphasis. First, neuroticism showed a strong positive association with state anxiety even after adjustment for age, education, and years of diving experience. Second, perceived stress was also associated with state anxiety after accounting for neuroticism and covariates, although the corresponding indirect association estimate was modest in absolute magnitude. Third, resilience was negatively correlated with both neuroticism and state anxiety and statistically moderated two stress-related associations: the neuroticism-perceived stress association and the perceived stress-state anxiety association. Importantly, resilience did not significantly moderate the direct neuroticism-state anxiety association. Taken together, the findings support an association-based framework in which dispositional vulnerability, recent stress appraisal, and adaptive psychological resources are jointly related to pre-dive anxiety, without establishing causal mediation or temporal ordering.

The strong association between neuroticism and state anxiety is consistent with a broad literature identifying neuroticism as a vulnerability marker for internalizing outcomes. Neuroticism reflects a relatively stable tendency toward negative affect, heightened threat sensitivity, emotional lability, and stronger reactivity to aversive events. Recent evidence continues to reinforce this position. For example, a systematic review and meta-analysis found a robust association between neuroticism and anxiety outcomes [25], while a prospective study in emerging adulthood showed that neuroticism was associated with later changes in broadband internalizing symptoms over 30 months [26]. A more recent mediation-oriented review likewise described neuroticism as a general antecedent of internalizing psychopathology, while emphasizing that its associations are not exhausted by direct trait–symptom correlations alone [27]. In the context of professional diving, these tendencies may be especially relevant because the operational environment combines environmental uncertainty, physiological load, vigilance, and limited margin for error. Anxiety regulation is therefore not merely a matter of comfort; it is closely tied to safe execution and cognitive performance under pressure.

Our findings also align with the emerging diver-specific literature. A recent cross-sectional study of special operations divers reported that neuroticism was strongly and positively correlated with both state and trait anxiety, whereas the other Big Five traits were negatively associated with anxiety [8]. That study was important in showing that personality is relevant to anxiety vulnerability in a diver population, but it remained largely descriptive. The present study extends that descriptive pattern by examining perceived stress and resilience in the same statistical framework. Specifically, the results suggest that pre-dive anxiety in professional divers may be better understood by considering trait-like vulnerability, recent stress appraisal, and adaptive psychological resources together, rather than focusing on personality or anxiety symptoms alone. Nevertheless, because the data are cross-sectional and the exact interval between questionnaire components was not recorded for every participant, these findings should be interpreted as correlational rather than temporal.

A further point is that the neuroticism-state anxiety relationship should be interpreted in light of the close conceptual relationship among neuroticism, trait anxiety, negative affectivity, and state anxiety. Spielberger’s formulation of trait anxiety explicitly describes stable individual differences in anxiety proneness and in the tendency to perceive stressful situations as threatening and respond with elevations in state anxiety [28]. This means that the observed correlation between neuroticism and STAI-S scores may partly reflect shared negative-affect content, same-source self-report measurement, and limited discriminant validity between overlapping affective constructs, not only a diving-specific process. The present study therefore does not claim that neuroticism uniquely causes pre-dive anxiety. Rather, it suggests that a trait-like vulnerability measure was strongly associated with immediate pre-dive anxiety in this occupational sample.

The role of perceived stress also requires careful interpretation. In this cross-sectional dataset, higher neuroticism was associated with higher perceived stress, and perceived stress was associated with higher state anxiety after accounting for neuroticism. Recent work outside diving supports the plausibility of this pattern. For instance, one study showed that neuroticism was linked to perceived stress through self-efficacy in younger and older adults, highlighting that personality-related vulnerability often co-occurs with intermediate psychological processes rather than with a purely direct trait-symptom link [29]. Likewise, recent personality-focused research in Chinese adults reported that neuroticism and related personality dimensions are meaningfully connected to perceived stress and its regulation [30]. At the same time, the standardized coefficients indicate that neuroticism showed a substantially larger association with pre-dive state anxiety (β = 0.78 for the total association and β = 0.74 after accounting for perceived stress) than perceived stress did (β = 0.13). Thus, perceived stress contributed statistically significant but comparatively modest explanatory information beyond neuroticism. This is theoretically sensible because neuroticism and anxiety share substantial variance in negative affectivity, and neuroticism may be related to acute anxiety through several additional processes, including threat interpretation, attentional bias, coping style, and emotion regulation. Thus, perceived stress should be interpreted here as one associated explanatory correlate rather than as a demonstrated mediator.

In addition, the distinction between perceived stress and state anxiety is important. The PSS was developed to assess the degree to which individuals appraise situations in their lives as unpredictable, uncontrollable, and overloaded, typically over the past month [13]. By contrast, the STAI-S assesses a transient emotional state characterized by tension, apprehension, and autonomic arousal, reflecting how respondents feel “right now, at this moment” [28]. Thus, perceived stress and state anxiety may overlap through negative affectivity and threat appraisal, but they are not identical constructs. From the perspective of transactional stress theory, recent stress appraisal may shape how individuals evaluate and emotionally respond to an imminent task demand. In the present study, perceived stress contributed additional explanatory information beyond neuroticism in relation to pre-dive state anxiety. However, the modest indirect association estimate indicates that perceived stress should not be regarded as the main pathway linking neuroticism with state anxiety. Instead, it is better understood as one stress-appraisal correlate among several possible psychological processes.

The interaction findings provide the most distinctive contribution of the study. Resilience did not significantly moderate the direct neuroticism-state anxiety relationship. Instead, the neuroticism-perceived stress and perceived stress-state anxiety relationships were weaker at higher levels of resilience. Recent work in other populations likewise suggests that resilience can condition the strength of stress-related associations [30, 31]. Although these studies involved different populations and outcomes, they converge on the idea that resilience is not a universal correlate of lower distress across all pathways. In this sense, the present results suggest that resilience was associated with weaker stress-related associations in the present sample, rather than serving as evidence of a demonstrated protective mechanism or as a direct modifier of personality-related anxiety vulnerability itself.

The meaning of the observed STAI-S level also requires nuance. The mean STAI-S score was 36.86 (SD = 9.44), which does not suggest uniformly severe pre-dive anxiety in this sample. In safety-critical work, a moderate level of pre-task arousal may be functional because it can support vigilance, preparation, and respect for task demands. Therefore, the findings should not be interpreted as implying that all pre-dive anxiety is harmful or that all divers would benefit from anxiety reduction. A more appropriate implication is that occupational mental health programs should distinguish functional readiness from excessive, persistent, or impairing anxiety. Interventions aimed at reducing anxiety should be considered primarily for divers with marked distress, repeated high anxiety, impaired functioning, or safety-relevant concerns, rather than for all divers with normal pre-task arousal.

These findings have several occupational health implications, but they should be framed cautiously. From a preventive perspective, diver mental health surveillance may benefit from paying attention not only to manifest anxiety symptoms but also to personality-linked vulnerability, perceived stress, and resilience. Higher neuroticism should not be viewed as a stand-alone exclusion criterion in diver selection. Rather, it may be considered one component of a broader risk-stratification framework, helping identify divers who may benefit from closer follow-up, stress-appraisal education, pre-task psychological preparation, or referral pathways if anxiety is persistently high. CBT-informed programs may help divers identify catastrophic appraisals, strengthen coping self-statements, and rehearse adaptive responses under pressure, while resilience-oriented or mindfulness-based stress-management training may improve emotion regulation, attentional control, and recovery from operational stress. Recent evidence indicates that resilience-oriented digital interventions and cognitive-behavioral approaches can improve resilience-related and mental health outcomes in adults [32, 33]. Nevertheless, the present data do not demonstrate that such interventions would reduce pre-dive anxiety or improve performance in divers; this remains a testable direction for future longitudinal and intervention studies.

Limitations

Several limitations should be acknowledged. First, the cross-sectional design precludes causal inference and does not allow temporal ordering among neuroticism, perceived stress, resilience, and state anxiety to be established. For this reason, the present analyses should not be interpreted as evidence of mediation. Second, all variables were assessed by self-report, which may have introduced common method bias and inflated same-source associations. This concern is particularly relevant given the strong correlation observed between neuroticism and state anxiety; the magnitude of this correlation may also reflect construct overlap and limited discriminant validity between neuroticism-related negative affectivity and state anxiety. Third, although the STAI-S was administered before diving tasks, pre-dive assessment conditions may not have been fully standardized across participants, and variability in task type or operational context may have influenced state anxiety scores. Fourth, participants were recruited through convenience sampling, which may limit representativeness and generalizability, especially to female divers and to diver populations outside the sampled settings. Fifth, the response formats of the PSS-14 and RS-25 differed from their original versions, which may affect comparability with other studies despite acceptable internal consistency in the present sample.

Future research should move beyond one-time self-report designs. Longitudinal or repeated-measures studies could assess stress appraisal and anxiety before, during, and after training cycles or operational exposures, thereby clarifying temporal sequencing. Future studies could also incorporate alternative personality measures, clinician-rated anxiety assessments, behavioral indicators, physiological stress markers, and supervisor or task-performance ratings to better establish discriminant validity, strengthen inference, and reduce mono-method bias. It would also be useful to examine whether resilience-related statistical interaction patterns differ across operational subgroups, exposure intensity, or experience level. Most importantly, intervention studies are needed to test whether strengthening resilience and improving stress appraisal can actually reduce pre-task state anxiety and improve safety-relevant performance outcomes in professional divers. Given the accumulating evidence that resilience-focused and cognitive-behavioral strategies can improve resilience-related outcomes in adults, this appears to be a realistic and testable translational direction rather than a purely speculative one.

Conclusion

In this cross-sectional study of professional divers, neuroticism and perceived stress were positively associated with higher pre-dive state anxiety, whereas psychological resilience was negatively associated with anxiety and was linked to weaker selected stress-related associations. Resilience did not significantly moderate the direct neuroticism-state anxiety association, suggesting that its role may be more closely related to stress appraisal and coping-related processes than to trait-linked anxiety vulnerability itself. Because the mean STAI-S score was in a low-to-moderate range and the design was cross-sectional, the findings should not be interpreted as evidence of clinically elevated anxiety, causal mediation, or a need to reduce all pre-dive arousal. Longitudinal, repeated-measures, and intervention studies are needed to clarify temporal ordering, practical significance, and occupational mental health applications in diving populations.

Abbreviations

NEO-FFI

NEO Five-Factor Inventory

NEO-FFI-N

Neuroticism subscale of NEO-FFI

STAI-S

State scale of the State-Trait Anxiety Inventory

PSS-14

14-item Perceived Stress Scale

RS-25

25-item Resilience Scale

SEM

Structural equation modeling

Authors’ contributions

Houyu Zhao, Xuhua Yu and Yiqun Fang designed study; Longfei Wang, Shi Zhang and Yan Wang collected data; Houyu Zhao, Wei Ding and Nan Zhao analyzed data; Houyu Zhao and Longfei Wang drafted manuscript; all authors reviewed and approved.

Funding

This work was supported by the Naval Weapon and Equipment Pre-research Unit (3020201020102), the Discipline Development Program in Diving Medicine (25XKQ0003).

Data availability

The dataset contains potentially identifiable occupational information and is not publicly available. De-identified data may be available from the corresponding author upon reasonable request, subject to ethics approval and institutional policies.

Declarations

Ethics approval and consent to participate

This study involves human participants and was approved by the Naval Special Medical Center (protocol code 2021070201). Participants gave informed consent to participate in the study before taking part. This research is in compliance with the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.