Neurobiological stress markers in educational research: A systematic review of physiological insights in health science education

Background: Traditional self-reported measures in health science education often overlook the physiological processes underlying cognitive and emotional responses. Purpose: This review aims to analyze the frequency, sensitivity, and utility of physiological markers in understanding cognitive and emotional dynamics in learning environments. Methods: A systematic PubMed search identified 156 records, with 13 studies meeting inclusion criteria. Markers analyzed included heart rate (HR), heart rate variability (HRV), cortisol, alpha-amylase, testosterone, s-IgA, blood pressure, oxygen saturation, and respiratory rate. Main Findings: HR and HRV were sensitive to educational stressors. Cortisol and alpha-amylase showed mixed results, while testosterone and s-IgA showed limited utility in directly assessing stress responses. No consistent link was found between any marker and immediate learning success. Conclusion: Physiological markers in learning environments can offer valuable insights into emotional and cognitive dynamics but should not be misconstrued as direct indicators of learning outcomes.


Introduction
Traditional educational research methodologies, predominantly tethered to self-reported measures, have offered valuable insights into subjective experiences but fall short in capturing the rich, real-time biological interplay that accompanies and modulates memory and learning processes.The brain, an organ highly responsive to both external and internal stimuli, orchestrates responses through a series of neurobiological mechanisms that are not always consciously perceived or accurately reported by individuals [1].Since education fundamentally transpires within this exquisitely sensitive organ, understanding these mechanisms requires an integrative approach that goes beyond self-report surveys.Incorporating physiological metrics in modern learning environments can serve as a complementary approach, providing a window into the autonomic nervous system's engagement and the physiological undercurrents that signify deeper cognitive and emotional involvement in learning tasks [2][3][4].Building on this understanding, the aim of this review is to systematically explore and synthesize existing literature on the use of physiological stress markers-such as heart rate (HR), heart rate variability (HRV), cortisol, alpha-amylase, and others-in health science education.By summarizing the methodologies, findings, and limitations of current research, this review seeks to highlight the potential of these physiological measures to enrich our understanding of the neurobiological dimensions of learning.The review aspires to establish a foundation for future research that integrates these metrics with traditional methods, thereby advancing the field and potentially leading to more effective educational strategies tailored to the physiological and emotional needs of learners.
Research indicates that the relationship between stress and learning is complex: low stress can lead to disengagement and reduced motivation, while high stress can overwhelm cognitive functions, impairing memory and focus [2,5].However, moderate stress levels have been shown to enhance attention and support memory consolidation, which are crucial for effective learning and improved academic performance [6].Contrary to the common perception that stress is purely detrimental, substantial evidence has long demonstrated its beneficial role in learning and memory processes.Findings in human studies corroborate that stress, when within manageable limits, can act as a potent enhancer of cognitive functions [7][8][9][10][11].For example, a study examining the effects of stress-induced cortisol increases found that during acute stress, retrieval of previously learned emotionally negative and neutral word pairs was impaired, particularly for items learned five weeks prior.This impairment was associated with elevated cortisol levels and heightened sympathetic activity, suggesting that high arousal during stress can hinder memory retrieval [12].Conversely, another study demonstrated that post-learning stress, induced via the cold pressor stress (CPS) procedure, enhanced memory consolidation for emotionally arousing material.This enhancement was observed in participants who had elevated cortisol levels following the stress procedure, indicating that stress can strengthen the consolidation of memories [7].These findings highlight the complex role of stress, where it can disrupt recall under high arousal conditions but also bolster the consolidation of memories, particularly those with emotional significance.
Upon encountering stress, two primary physiological systems are activated: the autonomic nervous system (ANS) and the hypothalamopituitary-adrenal (HPA) axis.
The ANS, through its sympathetic branch, rapidly mobilizes energy by releasing adrenaline and noradrenaline, collectively known as catecholamines.
Concurrently, the HPA axis triggers a prolonged response by releasing corticotropin-releasing hormone (CRH) from the hypothalamus, prompting the pituitary gland to secrete adrenocorticotropic hormone (ACTH).ACTH then prompts the adrenal cortex to produce glucocorticoids, such as cortisol in humans and corticosterone in rodents.These hormones orchestrate a body-wide response to stress, preparing the organism for sustained alertness and adaptive changes [13].
Noradrenaline, primarily secreted by the sympathetic nervous system and also present in the brain, plays a vital role in modulating learning during stress.[14].It bolsters the consolidation of emotionally charged experiences by influencing the amygdala, a crucial area of the brain involved in processing emotional memories.This effect is largely mediated through β-adrenoceptors in the basolateral amygdala (BLA).When these receptors are activated by noradrenaline, they aid in the consolidation of memories tied to significant emotional events [14][15][16].Experimental evidence supports this by showing that infusions of noradrenaline into the BLA improve memory consolidation, whereas blocking its action impairs this process [14,17,18].Thus, noradrenaline serves to prioritize the retention of important information during stressful or emotional situations, enhancing learning outcomes [15].
Glucocorticoids like cortisol and corticosterone also significantly influence learning and memory.These hormones enter the brain and bind to glucocorticoid receptors (GRs) located in many areas, including the BLA [3,19].Activation of these receptors after stressful learning tasks can enhance memory consolidation-a process crucial for converting short-term memories into long-lasting ones [6,20].For instance, administering a GR agonist into the BLA immediately after learning enhances memory retention, whereas blocking GR impairs this consolidation [14,21,22].Furthermore, the interplay between glucocorticoids and the noradrenergic system exemplifies a synergistic enhancement of memory consolidation, where glucocorticoids amplify the signaling pathways initiated by noradrenaline [20].This enhancement is thought to be mediated by a rapid potentiation of the noradrenaline signaling cascade, which underscores the interplay between these stress-responsive systems.
While stress can enhance the consolidation of memory, it also has a documented negative impact, particularly on the retrieval of memories [23][24][25].High levels of stress, especially during memory retrieval, can impair the ability to access stored information, primarily due to the deactivation of key brain regions such as the hippocampus.A research approach using functional MRI has shown a profound deactivation of the hippocampus and other limbic structures, including the hypothalamus and medio-orbitofrontal cortex, in response to acute stressors [26].This deactivation correlates strongly with the release of cortisol, the stress hormone known to inhibit hippocampal function.Given the hippocampus' crucial function in navigating memories [27], its diminished activity under stress results in challenges in memory retrieval.This demonstrates the complex, often contradictory roles of stress in memory processes: while facilitating initial memory encoding and consolidation, stress can simultaneously hinder the retrieval of these stored memories, highlighting the necessity for a nuanced understanding of stress responses to optimize cognitive outcomes.
Exploring the stress response with scientific rigor primarily involves tracking key physiological indicators.Researchers frequently measure cortisol levels to assess HPA axis activation [28][29][30] and blood pressure, heart rate or heart rate variability (HRV) to gauge SNS activity [31][32][33][34][35].These markers not only reveal the body's response to stress but also illustrate the dynamic interplay between stress and physiological regulation.Additionally, the field continues to evolve with numerous other potential biomarkers, expanding our capabilities to validate and enhance our understanding of stress responses.
Alpha-amylase, for instance, is another valid stress marker increasingly recognized by researchers for its role in reflecting the activity of the sympathetic nervous system [36][37][38].Unlike cortisol, which signals the response of the HPA axis, alpha-amylase is an enzyme found in saliva that increases during stress-related sympathetic activation [39][40][41][42][43].Its measurement provides a non-invasive, real-time indicator of stress response, complementing cortisol by offering insight into the body's immediate reactions to stressful stimuli.
Blood oxygen saturation and salivary immunoglobulin A (IgA) are also important markers used by researchers to assess stress responses, each providing unique insights into physiological adaptations under stress.Blood oxygen saturation, measured via pulse oximetry or hyperspectral imaging [44], indicates how well oxygen is being delivered to parts of the body furthest from the heart, reflecting the body's efficiency in adapting to stress and maintaining vital functions [45,46].Salivary immunoglobulin A (IgA) has been shown to react dynamically to stress, with its secretion patterns offering insights into the body's immunological response under varying conditions [47].Research indicates that IgA levels can initially rise in the immediate aftermath of stress exposure, potentially as a protective immune response [48][49][50].However, with prolonged or repeated stress, there is often a subsequent decline in IgA secretion, which may reflect a depletion of resources or downregulation of immune functions over time [47].These dual effects suggest that salivary IgA is a potentially good marker for stress-related immune changes, offering valuable information, but its significance is strongly context-dependent.
Building on the discussion of traditional stress markers such as cortisol and heart rate variability, it's important to also consider the role of testosterone in stress modulation.The acute effects of stress on testosterone levels show variability across studies; some report an increase [51][52][53], while others note a decrease [54][55][56] following stress exposure.This variability highlights testosterone's complex regulatory influence on the body's stress responses.Particularly in social settings, testosterone may reduce anxiety and tailor physiological reactions to stress, emphasizing its vital role in adaptive behavior [57].The hormone's capacity to modulate stress responses adaptively is pivotal, reflecting its key function in the management of stress across different scenarios.
Continuing this examination of physiological markers used to measure stress, respiratory rate is another significant indicator whose variations under stress can provide insightful data [58].Like other biomarkers, respiratory rate often changes in response to stress, typically increasing as part of the body's rapid attempt to enhance oxygen intake and prepare for 'fight or flight' responses.This adjustment in breathing patterns can be particularly telling in acute stress situations, where the need for increased oxygen to support heightened brain and muscle function becomes critical [59,60].Monitoring changes in respiratory rate can thus serve as a reliable measure of stress reactivity, offering a direct glimpse into the autonomic nervous system's engagement during stressful episodes.
The impact of stress on learning and memory is complex, exhibiting both facilitating and impairing effects.This duality can be understood through a unifying theory which suggests that stress enhances learning and memory when it coincides both temporally and contextually with the learning event [3].Specifically, the efficacy of stress in enhancing memory is contingent on the alignment of stress-induced hormonal and neurotransmitter activities with the brain circuits engaged by the learning experience.Such convergence ensures that stress-related neurochemical changes enhance the attention and retention of relevant information.Understanding these interactions is crucial, especially in designing educational environments that effectively leverage stress responses to foster better learning outcomes.As we navigate the digital age, with its unique stressors and learning modalities, a holistic grasp of these stress-inducing characteristics becomes essential to developing engaging and sustainable learning environments.
Despite the growing interest in understanding the neurobiological underpinnings of learning in health science education, significant gaps remain in the literature.While previous studies have explored the use of physiological stress markers, such as heart rate (HR), heart rate variability (HRV), and cortisol, in educational settings, the frequency with which these markers are used within health science educational research remains underexplored.Additionally, there is a need for more detailed information regarding their sensitivity across different learning formats and their correlations with constructs relevant to the teaching-learning context.The present study aims to address these gaps by systematically reviewing the existing literature on physiological stress markers in health science education.Specifically, we seek to analyze the frequency of use and the sensitivity of these markers, as well as their utility in providing insights into the cognitive and emotional dynamics of learning environments.By synthesizing these findings, this review aims to offer a more comprehensive understanding of the role these markers play in educational research and to propose directions for future studies that could enhance educational strategies through the integration of physiological measures.

Search strategy
A systematic search of PubMed (MEDLINE® database; National Institutes of Health, United States National Library of Medicine, Bethesda, MD) was performed in order to identify records for the present review.
The following search operator was used:

Study selection and inclusion criteria
The process of study selection is shown in Fig. 1. 156 records were identified from PubMed and screened by reading the title and the abstract of the respective records.Each record was screened for meeting the following inclusion criteria: 1. Human trial with a healthy study population 2. Peer-reviewed empirical study published in an academic open access journal 3. Analysis of physiological parameters 4. Context of higher healthcare education This review specifically targets higher health science education due to the distinctive stressors in this field, such as high-stakes clinical simulations and the intense academic environment, which present unique opportunities to study the impact of stress on learning.While the fundamental neurobiological mechanisms of stress are likely to be similar across various educational domains, the context of health science provides a particularly relevant backdrop for exploring these dynamics in detail.This focused approach allows us to explore the specific stressrelated challenges faced by students and educators in this field.
One hundred and fifty-three items were screended by reading the title and the abstract.The eligible articles from the data research were independently screened by the authors, who then compiled a list of potentially relevant articles according to the inclusion criteria.Any discrepancies regarding the inclusion or exclusion of articles identified during the abstract screening process were resolved through consensus reached via full-text review.
117 records were removed after the screening process because they did not meet the inclusion criteria, for instance due to an unsuitable study population or study design.The eligibility for inclusion of the remaining 36 records was further evaluated by the authors through fulltext screening.After this, 12 records were found to be included into the present review.Furthermore, one additional study was identified through the review of the reference lists of included records.

Data extraction
Information on the names of the authors, the year of publication, the country, the study design, the study objective and the relevant findings of the final included records were summarized in Table 1.The main focus of the present systematic review was the analysis of physiological stress parameters in the context of higher health science education.

Study characteristics
After the screening of 153 records, 12 articles met full criteria for inclusion in the review.One additional article was identified through the review of reference lists.In total, 13 articles were included in the review.The articles were published between the years 2016 and 2023.Apart from the year 2022, there was an increase in number of publications per year with the majority of articles published in 2023 (n = 4) (Fig. 2A).
Due to the inclusion criteria all study populations in the present review derived from heath science areas.Most studies included medical students (n = 6) (Fig. 2C).The other study populations consisted of early career physicians (n = 3), physical therapy students (n = 3) and nursing students (n = 2).Note that the populations in some records included more than one profession.

Physiological findings
The main part of the following physiological findings are summarized in Table 1.

Heart rate
The heart rate (HR) was mostly used as a physiological stress marker for measurements in the context of neurobiological arousal and health science education.For HR, various studies reported a positive association with educational stressors such as clinical simulation events or more interactive sessions [62][63][64][65][66][67] Cavaleri et al. (2023) found, that the low-stress group in the context of simulation-based medical education exhibited superior performance in comparison with higher-stress groups alongside with higher mean HR [64].It was also shown, that HR correlated with general joy of life and emotional exhaustion [65,67].Trainee baseline HR was also demonstrated to correlate with stress scores [65].On the other hand, it was observed by Gouin et al. that repeated simulation courses did not reduce maximum HR [68].Moreover, James et al. (2021) showed that increased HR during non-technical training was associated with gender and postgraduate medical study [65] (Table 1).

Heart rate variability
Heart rate variability (HRV) analyses were assessed in several studies and interpreted in terms of leading sympathetic or parasympathetic activity.Bakhsh et al. (2019) found that team simulation generated a higher sympathetic activity in comparison to individual simulation, demonstrated through an increase in the low frequency/high frequency ratio (LF/HF) [62].Moreover, subjective workload scores correlated with the sympathetic activity in early career surgeons [62].Gellisch et al. (2022) showed, that HRV was lower in face-to-face learning in comparison with online learning, indicating an increase in sympathetic activity for the face-to-face session [69].This activation of the sympathetic nervous system was also shown to positively correlate with the emotion of enjoyment in the context of face-to-face learning [69].In line with this, the parasympathetic activity assessed via SDNN was highest in passive online learning compared with interactive online learning and face-to-face learning.It was also publicized, that SDNN and general joy of life were positively associated [67].Moreover, Gellisch et al. ( 2023) demonstrated in another study, that the decrease in HRV, was more prominent in an interaction-enhanced environment compared to a passive online learning session [70].Hundertmark et al. (2019) found, that in the context of a medical student peer-teaching program, the HRV parameter RMSSD increased over individual course sessions and over measurement days [71].For this context, the parameter RMSSD correlated with the personality trait of extraversion [71] (Table 1).

Blood pressure
For blood pressure (BP), Bialka et al. ( 2021) observed an increase during a high-fidelity critical care simulation [63].Another study by Liaw et al. (2023) described a similar effect of an increase in systolic BP during a team-based simulation training in managing clinical deterioration.This effect was compared between a virtual reality simulation and a face-to-face simulation, but no difference was found between the groups [72].Moreover, no association was detected between the stress response and performance scores in this study [72] (Table 1).

Salivary cortisol
In the context of transitioning from face-to-face to online medical education, salivary cortisol was higher in face-to-face session [69].For digital medical education, salivary cortisol levels were increased in an interaction-enhanced environment in comparison with passive online learning [70].Subjectively perceived stress correlated positively with salivary cortisol levels, but only in the interactive online teaching condition [70].Hundertmark et al. (2019) described, that over the measurement period of repeated medical peer-teaching tutor sessions the salivary cortisol levels of the tutors declined, though there were no correlations with course quality ratings [71].On the contrary, Ferreira et al. (2020) did not observe significant correlations between cortisol levels, anxiety and performance in the context of an objective structured clinical examination (OSCE) [73].In the context of high-fidelity critical care simulations, there were also no significant changes in salivary cortisol levels during the simulation, but more experienced team leaders showed lower salivary cortisol levels than other team members [63].Moreover, Cavaleri et al. (2023) did not detect significant changes in cortisol level during simulation-based education [64] (Table 1).

Salivary alpha-amylase
For the comparison of interaction-enhanced online learning with passive online learning in the medical field, an increase in salivary alpha-amylase was observed for the interactive condition [70].Moreover, Brodersen et al. (2021) found that alpha-amylase was higher after a nursing exit exam in comparison to a homework condition [74].The alpha-amylase level as a physiological stress parameter was not found to be associated with the exit exam score though [74].For the context of a high-fidelity critical care simulation, the changes in salivary alpha-amylase levels during the simulation were overall insignificant, but more experienced team leaders showed lower alpha-amylase levels than other team members [63] (Table 1).Abbreviations: HRheart rate; HRVheart rate variability, BPblood pressure.

Other physiological parameters
The study performed by Bialka et al. (2021) on stress levels during the high-fidelity critical care simulation used other additional physiological markers.For salivary IgA, there were no significant changes during the simulation [63].The changes in blood oxygen saturation were also insignificant [63].The salivary testosterone levels were higher after the simulation in both male and female medical students, though the authors suggest a careful interpretation regarding gender-dependent differences [63].Overall, the authors could not detect clinically relevant correlations between the physiological parameters [63] (Table 1).
James et al. ( 2021) also used respiratory rate as a physiological stress marker [65].The authors described gradual significant increases in respiratory rate for lectures, non-technical skills training and skills simulation [65].The authors did not find any correlations between respiratory rate and other collected data [65] (Table 1).The main findings from all physiological markers investigated and their correlations are comprehensively summarized in Fig. 3.This figure provides a visual overview of the key results for nine physiological stress markers across the 13 studies included in this systematic review.

Discussion
This comprehensive analysis of the systematically selected studies offers a detailed overview of the varied physiological markers employed across different learning environments and educational strategies.The present review highlights the diverse ways in which physiological responses are measured and interpreted, underscoring their critical role in understanding the interplay between stress and learning.Within discussing these findings, the aim was to delineate the informative value of different physiological markers in reflecting the neurobiological impacts of educational settings, thereby providing insights that can inform the development of more adaptive and effective educational practices tailored to manage stress and enhance learning outcomes effectively.
In the present systematic review, Heart Rate Variability (HRV) and Heart Rate (HR) emerged as predominant among the physiological markers used to assess stress responses in health science education.These markers were featured in a significant proportion of the studies analyzed, highlighting their robustness and sensitivity for detecting physiological variations.Notably, the frequency of significant-or at least biologically relevant-findings reported with HR and HRV underscores their sensitivity to educational stressors within the studies reviewed [62][63][64]67,69,70,71,75,76].However, it is important to clarify that this observation is based on the reported outcomes across studies rather than on a direct, quantitative comparison between HR/HRV and other physiological markers within the same study.As such, while HR and HRV appear to be particularly sensitive, no formal statistical comparison was made to determine whether these markers are superior to others in influencing educational outcomes.. Nevertheless, this observation is particularly salient, as it suggests that HRV, in particular, can effectively detect subtle yet educationally relevant physiological changes across different learning scenarios.HR and HRV are especially valuable in assessing learning environments, as it has been consistently demonstrated that these measures are closely related to workload, providing direct insights into the physiological impacts of educational stressors [77].These findings were further substantiated by investigations demonstrating that HRV measures are particularly sensitive markers for mental stress and mental workload [78,79], revealing not only general trends but also individual differences in bodily responses to specific aspects of learning environments.This level of sensitivity is crucial for tailoring educational strategies to effectively manage cognitive load and enhance individual student engagement and performance, accommodating diverse physiological profiles [80].The ability of HRV and HR to consistently reveal significant physiological responses where other markers might not, confirms their critical role in enhancing our understanding of the neurobiological underpinnings of stress and learning interactions within health science educational environments.As part of our results, we observed that the LF/HF marker of Heart Rate Variability (HRV), a frequency-domain measure that represents the ratio of the low-frequency (LF; 0.04 to 0.15 Hz) to high-frequency (HF; 0.15 to 0.40 Hz) bands, in some cases appears to be positively correlated with joy during the learning experience [69].This finding is particularly significant, as positive emotions have been shown to be associated with improved academic performance, knowledge acquisition, and health [81], highlighting the potential of HRV, specifically the LF/HF ratio, as a valuable tool for assessing the emotional quality of learning environments.As part of the analysis of the systematically selected publications, it was observed that markers such as blood pressure, oxygen saturation, and respiratory rate, which are intimately connected to cardiovascular function, often changed in tandem with fluctuations in HR and HRV [63,75].This co-variation underscores a physiological interdependence where these markers collectively respond to stressors in learning environments.Blood pressure and HR, for example, are both influenced by autonomic nervous system activity, which regulates cardiovascular responses to stress.Similarly, oxygen saturation and respiratory rate are critical in maintaining adequate oxygen levels in the blood, essential for sustaining physiological homeostasis under stress.The changes in these parameters alongside HR and HRV not only validate the sensitivity of HR and HRV as indicators of stress response but also enhance our understanding of the broader cardiovascular adjustments that occur during stressful educational activities.This correlation suggests that while HR and HRV provide direct insights into autonomic activity, the associated changes in blood pressure, oxygen saturation, and respiratory rate offer additional validation and a more comprehensive view of the body's integrated response to educational stressors.
Cortisol, widely recognized and frequently measured as a biological marker of stress [28,30], presents a complex picture in the context of health science education.Despite its scientific validity and extensive use in stress-related studies [28,32,50,82,83], our review indicates that cortisol may not always serve as a reliable indicator of stress in educational settings.A significant portion of the studies reviewed reported inconclusive or insignificant findings when correlating cortisol levels with educational stressors [63,64,76,73].This suggests that while cortisol is a well-established marker in broader psychological and physiological research, its utility and sensitivity in detecting stress specific to learning environments can be unexpectedly limited.Despite the challenges of using cortisol as a stress marker in educational settings, as noted, several studies within this review did report significant changes in cortisol levels [69][70][71].This variability in findings can be contextualized by considering that cortisol release is particularly Fig. 3.This figure provides a visual summary of the key results for nine physiological stress markers across the 13 studies included in this systematic review.The biomarkers summarized include Heart Rate (HR), Heart Rate Variability (HRV), Cortisol, Alpha-Amylase, Blood Pressure, Oxygen Saturation, Respiratory Rate, Salivary Immunoglobulin A (IgA), and Testosterone.Each marker's sensitivity and response to educational stressors are depicted, highlighting significant findings and trends across the different studies.The pictograms are taken from the following source: https://icons8.com/.
responsive to uncontrollability and social-evaluative threats, which are among the most potent stressors for inducing cortisol responses in psychological stress scenarios [28,82].It is plausible that the studies reporting significant findings employed teaching frameworks that strongly elicited these specific stressors, thereby more effectively triggering a cortisol response.The studies reviewed in this manuscript may not have consistently subjected participants to such high levels of stress, which could explain the observed variability in cortisol responses.This suggests that while cortisol can be a valuable biomarker in educational settings, its effectiveness may be more pronounced under conditions of exceptionally heightened stress.However, even when significant changes in cortisol levels are observed, interpreting these measurements within teaching contexts remains complex.Interpreting cortisol measurements within teaching contexts can be challenging due to the multilayered nature of cortisol responses.Cortisol levels can be influenced by a wide range of factors, including individual differences in stress reactivity, the specific nature of the educational stressors, and the timing of cortisol measurements [30,34].These variables make it difficult to draw straightforward conclusions about the relationship between cortisol levels and educational outcomes, highlighting the need for careful consideration of context and methodology when using cortisol as a biomarker in educational research [84].Nonetheless, the fundamental importance of cortisol cannot be overlooked.Its role in modulating learning and memory processes is well-established, making it a crucial marker for investigating how stress impacts educational outcomes [4,6,16,85].The ability of cortisol to influence such neurobiological functions underscores its potential utility in enhancing our understanding of the physiological dimensions of learning environments.
Alpha-amylase has been recognized as a valuable physiological marker for assessing stress responses, underpinned by its scientific validation and role in reflecting sympathetic nervous system activation [36,37,39,40].Despite its physiological relevance, this review indicates that alpha-amylase is not frequently chosen by researchers, with only three records of the included dataset reporting its use [70,63,74].The findings from these studies present a mixed picture: while some reported significant differences in alpha-amylase levels in response to educational stressors [70,74], others did not observe such changes [63].This variation suggests that while alpha-amylase can provide valuable insights into the intensity of the stress response, it may not consistently perform across different contexts or study designs.Consequently, the decision to incorporate alpha-amylase into research on stress in educational settings should consider the balance between the marker's potential benefits and the logistical or financial costs associated with its measurement.
Turning to stress markers less commonly employed in educational stress research, both testosterone and salivary immunoglobulin A (IgA) were infrequently featured in the studies we reviewed.The inconsistent use of IgA may stem from the mixed perspectives in the literature regarding its reliability as a stress biomarker.While some sources advocate for IgA as a reliable indicator of acute stress [86,47], others contest its efficacy, pointing to empirical findings suggesting that IgA may not be sensitive enough to provide meaningful insights into the actual stress response [87].Additionally, the circadian patterns of salivary IgA, which could be altered by stress, introduce potential biases that further complicate its use as a stress marker [86].On the other hand, literature regarding testosterone as a direct stress marker is scarce, yet interestingly, there are studies indicating that testosterone may influence cortisol release, suggesting a more indirect role in the stress response [88].Although testosterone might not be ideal for assessing immediate stress reactions, its involvement in modulating broader hormonal responses could provide valuable insights into the overall dynamics of stress physiology.In the study that utilized testosterone and salivary immunoglobulin A (IgA), significant changes were reported for testosterone, though the interpretation of these data remains incomprehensible, highlighting the complexity and variability of these measures in educational stress research [63].Thus, while both salivary IgA and testosterone present certain limitations in directly quantifying stress, their inclusion in research could enhance our understanding of the more complex aspects of the stress response in educational settings.
A common initial assumption among a broad readership may be to directly correlate specific physiological responses, such as variations in stress markers, with academic performance or enhanced memory processes.However, the findings from our review clearly challenge this simplistic view.The authors found no consistent link whereby physiological stress, or its absence, could predict immediate learning success.Instead, this review underscores the complexity of the learning process, emphasizing that it is the intricate interplay of various mediatorssuch as emotional responses within the learning context [81,89] that are shaped by physiological stress responses [70].These mediators may significantly influence long-term learning outcomes.It is essential to recognize that while physiological markers provide valuable insights into the bodily states during educational activities, they are part of a broader system where emotional and cognitive factors interact, potentially steering sustained educational success.This understanding calls for a more holistic approach to interpreting how stress impacts learning, moving beyond simple cause-and-effect to embrace the multifaceted nature of human cognition and emotional regulation in educational settings.

Conclusion
Our review reveals significant variability in the sensitivity of physiological markers and their effectiveness depending on the context, highlighting the need for a careful and thoughtfully selected approach when integrating these measures into educational research.These findings underscore the importance of selecting appropriate biomarkers based on the specific educational environment and stressors being studied, as well as the necessity of considering the context in which these markers are applied.Further, it is important to recognize that physiological stress markers, while insightful, should not be expected to directly correlate with immediate academic performance or learning success.Our results clearly indicate that these markers are not direct performance-related measurements.However, they play a crucial role in monitoring the learning process, particularly because of their association with specific emotional responses that have been linked to academic success over the long term.Additionally, these physiological stress measurements can offer valuable insights into the effectiveness of different learning environments or teaching methods.For instance, while stress may hinder memory retrieval, it can be beneficial for memory consolidation.Understanding how certain stimuli within learning environments contribute to stress levels enables the design of educational strategies that optimize learning by aligning stress-inducing aspects with the targeted memory phase, leveraging the benefits of stress during consolidation while minimizing its impact during retrieval.Therefore, while physiological measurements may not predict immediate outcomes, they provide essential insights into the emotional and cognitive dynamics of learning and the underlying memory processes, all of which are vital for fostering sustained academic achievement.
While this review focuses on health science education, the findings have broader implications for understanding stress in other educational contexts.The physiological markers discussed here could serve as a foundation for future studies in different subject areas, thereby expanding the applicability of this research beyond the health sciences.
As educational science advances, it is imperative that the frameworks not only incorporate these physiological insights but also critically assess the interdependencies of stress markers with the holistic educational experience.This understanding is relevant for developing targeted interventions that can enhance both the physiological and psychological well-being of students, ultimately fostering more effective learning environments.By recognizing the complex nature of stress and its impact on learning, educators and researchers can better tailor their approaches to support and enhance student learning in health science education.

Fig. 1 .
Fig. 1.Flow chart after Page et al. (2021) [61] illustrating the process of record identification and screening for the present systematic review.

Fig. 2 .
Fig. 2. Diagrams illustrating the study characteristics of the included articles.A shows a line graph of the number of included publications published per year.B shows the distribution of publication countries.C shows the study populations.D shows the physiological parameters assessed.Note, that some records included several professions in their study population and assessed various physiological parameters.

Table 1
The table summarizes the reviewed articles with special regards to their study objective and physiological findings.The articles are sorted in alphabetical order of the authors.