1 Background

An ever-growing number of people are indulging in sports in the mountains and in the wilderness environment, a trend that was accelerated by the COVID-19 pandemic (Hedenborg et al. 2024, Beery et al. 2021). In contrast to indoor sport activities, outdoor sport activities tended to increase after the pandemic (Beery et al. 2021, Lee et al. 2022). In particular, backcountry skiing, ski touring, and snowshoeing on mountainous terrain have become more popular in recent years (Procter et al. 2014). Physical activity in an alpine environment has a positive effect on mental and physical health (Ower et al. 2019). Studies have shown that even viewing an alpine environment already positively affects emotional analytics in patients with somatoform, depressive, and anxiety disorders as well as in healthy controls (Hufner et al. 2020). Despite this beneficial effect, winter sports off secured ski slopes are at significant risk for avalanches.

In the case of critical burial, that is, at least involving the head and chest, the chance of survival is highly time dependent (Falk et al. 1994, Procter et al. 2016). In the initial phase, when survival rates are the highest, the companions or bystanders initiate the search and rescue operation. Companion rescue comprises searching for the victim by using an avalanche transceiver, probing, and extracting the victim from the avalanche debris (Van Tilburg et al. 2024). Unfortunately, not all mountaineers are sufficiently trained in avalanche rescue (Procter et al. 2014), which can then lead to a disproportionate increase in mental stress and arousal, which in turn can negatively affect search performance (Rahman et al. 2012). High arousal in a traumatic situation is also a risk factor for posttraumatic stress disorder (PTSD) (Finnsdottir and Elklit 2002, Leonard et al. 2021, Salvotti et al. 2024). Recently, transceivers with automated verbal commands have been introduced to help the user locate a critically buried avalanche victim in the shortest possible time. The aim of the current mixed-methods field study was to assess the effect of automated verbal commands during avalanche transceiver search on measures of mental stress and affective response valence, arousal, and dominance.

2 Materials and Methods

This prospective mixed-methods crossover study investigated the impact of a new voice navigation technology and the effect of automated verbal commands on mental stress and affective responses during the course of an avalanche transceiver search in a simulated avalanche scenario. This was part of a larger study on the effect of verbal commands on avalanche rescue. The ethics committee of the Medical University of Innsbruck, Austria, was contacted and considered a rigorous ethical evaluation unnecessary (inquiry to the Ethics Committee on December 23, 2021). After being informed in detail about the study aims and procedures, all participants provided informed consent prior to study participation. The study was conducted over 2 days in March 2022.

2.2 Study Protocol

One week prior to the trials, a video with structured instructions on how to perform an avalanche search and how to adequately use a standard avalanche transceiver (not the ones used in the experiment) and probe was sent to each participant. Immediately prior to each trial, the participants were confronted with the hypothetical scenario of being part of a skiing team, and one of their teammates was buried by an avalanche to simulate a realistic scenario. All participating researchers were instructed to only interact with participants in a very professional and distant manner and to refrain from any encouragement or small talk with the participants in order not to influence stress levels through human interaction.

Each participant performed two trials, and the activation of voice navigation by the avalanche transceiver and the choice of transmitters (left or right, TM1 or TM2; see Figure 1) were randomized. The trials were complete when the participant hit the transmitter panel with the probe and called out to have a “hit.” The trial was stopped after 10 min or three incorrect “hit” callouts by the participants.

3 Results

The mean age of the 50 volunteers (50% female and 50% male) was 24.0 ± 3.6 years. All participants were medical students at the Medical University Innsbruck, Austria. None of the participants reported any previous experience using an avalanche transceiver or knowledge of avalanche rescue techniques. Ninety-five of 100 individual trials were completed, and five trials were terminated by the experimenters because the predefined termination criteria were met: in two trials, the maximum trial time of 10 min was exceeded (one using VOICE, one NO VOICE), and three trials were terminated because of probing failure after three erroneous hits (one using VOICE, two NO VOICE).

3.1 Mental Stress

Participants reported higher levels of mental stress after the trial than before the trial (p ≤ 0.001) (paired t-test; mean diff. −1.2 ± 2.3; p ≤ 0.001; |d| = −0.52) overall and when analyzing the use of VOICE (p = 0.006) (paired t-test; mean diff. −1.0 ± 2.5; p ≤ 0.006; |d| = −0.41) and NO VOICE (paired t-test mean diff. −1.4 ± 2.1; p ≤ 0.001; |d| = −0.66) for navigation (Table 1). NO VOICE (Δ1.4 points) navigation showed a greater but not significantly greater increase in stress level than did VOICE navigation (Δ1.0 points, p ≤ 0.329). When mental stress levels were directly compared between conditions (VOICE vs. NO VOICE), there was no significant difference either before or after each trial.

TABLE 1. Mental stress, arousal, and dominance before (PRE) and after (POST) each trial, as well as the difference in the POST–PRE score as a function of the mean and SD. All three parameters were assessed using a visual analogue scale (VAS) ranging from 1 to 10 points, with 1 indicating “no stress at all” and 10 indicating “the highest level of stress possible.”

Mental stress, arousal, and dominance - Paired samples test
	Paired differences (mean)	PRE	POST	POST − PRE (Mean ± SD)	T-test	Cohen's d
Mental stress	VOICE	5.9	7.0	1.0 ± 2.5	p = 0.006	−1.3
	NO VOICE	6.0	7.3	1.4 ± 2.1	p < 0.001	−1.9
	Δ NO VOICE – VOICE			0.5 ± 3.5	p = 0.329	0.5
Arousal	VOICE	2.4	1.2	−1.2 ± 3.2	p < 0.001	1.9
	NO VOICE	2.1	1.5	−0.6 ± 2.2	p = 0.009	1.0
	Δ NO VOICE – VOICE			−0.6 ± 2.2	p = 0.041	−0.8
Dominance	VOICE	4.0	4.2	−0.2 ± 1.4	p = 0.382	−0.3
	NO VOICE	3.6	4.3	0.7 ± 2.7	p = 0.017	−1.0
	Δ NO VOICE – VOICE			0.5 ± 2.9	p = 0.207	0.6

The median increase in mental stress level was 2.00 (= median difference POST − PRE). A large increase in mental stress level was defined as >2.00, and a small increase in mental stress level was defined as ≤2.00. An increase in mental stress had a significant impact on the duration of the coarse search (p = 0.046). Those participants reporting a small increase in mental stress (n = 74 of 100 trials) were faster in the coarse search, leading to shorter coarse search times (small increases in mental stress mean 59.0 ± 8.2 s; 95% CI 43.0–74.9) (Figure 2). Those participants who reported a large increase in mental stress (n = 26 of 100 trials) were slower in the coarse search (81.0 ± 11.5 s; 95% CI 58.6–103.5; p = 0.046). A further analysis of the correlation showed a direct correlation between mental stress and certain search times. A reduction in mental stress correlated with a decrease in the fine search (Cox model p = 0.005), in the probing time (Cox model p = 0.037), and in the total search time (Cox model p = 0.018).

4 Discussion

This study was the first to evaluate the impact of a search operation after a simulated avalanche accident on psychological parameters, namely, mental stress and affective responses. The study further analyzed the effect of verbal commands from a new generation of avalanche transceivers on those parameters and the performance of laypersons.

The most important findings of our study were as follows:

VOICE and NO VOICE showed an increase in mental stress, which was less pronounced when using VOICE navigation. Greater increases in mental stress were associated with longer coarse search times.

Both VOICE and NO VOICE showed a reduction in arousal following the trial, with a significantly greater reduction using VOICE navigation. Individuals who showed a decrease in arousal (after each trial compared to before) were faster in the coarse search than participants who did not show this effect. In the subgroup of participants who reported an increase in arousal following the search compared to prior to the search, the use of VOICE was associated with faster search times.

Participants using NO VOICE navigation reported significantly greater dominance after the trial than immediately prior to the trial.

In the individual semistructured interviews, more than two thirds of all participants described VOICE navigation to be helpful.

The participants who indicated that VOICE navigation was helpful in the interviews were faster in the coarse search, which they attributed to an improvement in concentration and a reduction in the perceived stress level.

Verbal commands are increasingly being used in medical devices. Automated external defibrillators (AEDs) are increasingly fitted with verbal commands and provide an impressive example of enhancing user interactions with a medical device. The latest-generation AEDs make use of verbal commands to guide CPR (cardiopulmonary resuscitation) providers through the required steps while performing successful CPR (Beckers et al. 2005). In contrast, verbal commands that do not transport an adequate amount of information or distract from the actual task could create confusion. Clearly, the use of voice prompts in an AED has been shown to be both helpful and reassuring during a stressful event (Beckers et al. 2005).

The mental stress observed in the participants was significantly greater after each test run than before. The effect may be explained by the nature of the simulated situation. The study team aimed to create an atmosphere that re-enacted the stressful situation after an avalanche accident. In addition to their immense responsibility, which was transferred to the participants, they had to run over avalanche debris, which is a strenuous physical exercise to be performed at altitude (2.450 m asl).

This increase in stress was less pronounced in the participants who used VOICE navigation. When analyzing the performance of the participants in each trial, we found that those participants who had lower levels of stress performed better and had shorter coarse search times. This finding is in accordance with various studies found in the literature. One study revealed that acute psychological stress provocation led to decreased performance in elite male swimmers and that prerace cortisol levels were positively associated with lactate (Rano et al. 2019). In a different study, floorball players’ physical and psychosocial stress impacted performance (van der Does et al. 2015). The correlation analysis of mental stress and search times shows a clear linear correlation. An increase in mental stress was associated with an increase in valence and significantly longer search times. The clearest correlation was shown in an increase in those times when focusing and concentrating on accuracy and precise work were of the greatest importance, namely, in the fine search and during probing.

We found that individuals who described VOICE navigation as helpful or calming reported lower levels of stress and were significantly faster in the coarse search. One possible explanation may be the fact that certain participants use VOICE navigation and utilize the device for their benefit. Other participants found it difficult to accept the information and described the VOICE navigation to be distracting. This finding is in accordance with other studies that state that in certain cases, verbal commands may cause confusion. A study analyzing the performance of laypersons operating AEDs concluded that simple devices should be developed with clear instructions, and it reiterates that design, construction, and visual aids have an impact on user performance (Eames et al. 2003). This statement was confirmed by the observation of Beckers et al. that even in the second evaluation, in the automatic group (including voice prompts), three students were unable to deliver a shock (Beckers et al. 2005).

The present results have additional implication for the design of future devices being used in search and rescue devices as well as human–machine interfaces (HMIs). One study on the development of HMI in disaster-purposed search robot systems found that these systems aim to extend the existing robotics and human interface technologies and to establish a novel human–machine interacting structure to enable effective use of semiautonomous robots in place of humans for searching survivors from rubble under human remote instructions (Zheng et al. 2004).

In contrast to an increase in mental stress, the analysis of arousal showed a different picture. The majority of participants (57%) reported a lower level of arousal following the trial than directly prior to the test. Nevertheless, 43% of participants described an increase in arousal. When the two conditions, VOICE and NO VOICE navigation, were analyzed separately, a significant overall reduction in arousal levels was identified in favor of VOICE navigation. The condition using VOICE navigation showed a significantly greater reduction in arousal than did the condition using NO VOICE. As previously described, humans experience emotional arousal when threats to life and limb are imminent, such as in high stakes and emergency situations that occur in sociotechnical systems such as firefighting, aviation, and combat (Rahman et al. 2012). Emotional arousal has both advantages and disadvantages. On the downside, it introduces perceptual distortions and biases and an inability to process symbolic information and alters motor abilities. For this purpose, various emergency HMIs have been designed to accommodate the human capacities that have been altered by danger-induced emotional arousal.

Another highly interesting aspect is the course of arousal during the trial. We observed a decrease in arousal sometime after the trial, during a period when mental stress levels were still high. One possible explanation might be the different changes in arousal and mental stress over time. The authors hypothesize that arousal decreases faster immediately after the trial, while mental stress persists at a higher level and decreases only at a later stage. This later decrease was then detected in the semistructured interviews where participants reported a stress reduction when using VOICE, which could not be found in the quantitative tests. This finding has already been described in a study that showed that subjective stress perception was significantly greater immediately after the acute stressor than before it and started to decrease only 20 min after the acute stressor (Becker and Rohleder 2019).

The only significant effect of dominance was observed using NO VOICE navigation, where participants reported higher levels of dominance after the test trial. This result shows that dominance is closely related to self-control and self-efficacy. Higher dominance ratings could indicate that the rescuer feels more in control. This can be beneficial for decision-making in difficult situations. A recent study analyzed the integration of self-efficacy and response efficacy in decision-making and found that their interaction in shaping individuals’ decisions can enhance our comprehension of how perceived control influences choices, coping strategies, and overall psychological health (Yang and Delgado 2025). Participants who did not receive any help in the form of voice commands stayed in control and had to determine more by themselves. Valence does not seem to be influenced by VOICE navigation or by the order of the test trials, and since valence is an inferred criterion from instinctively generated emotions, it can be deduced that the entire trial does not influence the extent to which emotions are pleasant or unpleasant.

One possible consequence of being involved in an avalanche rescue operation is the subsequent development of PTSD not only in possible victims but also in the rescuers. High levels of mental stress and increased arousal have been found to possibly be risk factors for the development of PTSD; therefore, addressing affective responses and mental stress during avalanche rescue could also have long-term positive health consequences (Salvotti et al. 2024, Sareen 2014, van der Kolk 2000, Tian et al. 2014). Stress and arousal reduction are among the few factors that predict the occurrence of PTSD.

5 Conclusion

This was the first study to evaluate the impact of search and rescue operation after a simulated avalanche accident on psychological parameters and discovered that the performance of a simulated avalanche transceiver search resulted in an increase in mental stress but a reduction in arousal. Greater increases in mental stress were associated with longer coarse search times. Individuals who showed a decrease in arousal (after each trial compared to before) were faster in the coarse search than participants who did not show this effect. In the subgroup of participants who reported an increase in arousal following the search compared to prior to the search, the use of VOICE navigation was associated with faster search times. The majority of participants rated VOICE navigation as helpful, and those who indicated VOICE navigation to be helpful in the interview were faster in the coarse search, which they attributed to an improvement in concentration and a reduction in the stress level. For future studies and more in-depth insights, it is recommended to include additional control groups and longitudinal follow-ups, as well as addition analysis to further clarify the causal pathways.

Author Contributions

Bernd Wallner: conceptualization, investigation, writing – original draft, methodology, validation, visualization, writing – review and editing, data curation, resources. Fabio Caramazza: conceptualization, investigation, writing – original draft, writing – review and editing. Simon Woyke: conceptualization, investigation, writing – original draft, writing – review and editing, methodology. Manuel Winkler: conceptualization, investigation, writing – original draft, writing – review and editing, project administration. Ivo B. Regli: conceptualization, investigation, writing – original draft, writing – review and editing. Peter Mair: conceptualization, investigation, writing – original draft, writing – review and editing, validation, supervision, project administration. Gabriel Putzer: conceptualization, investigation, writing – original draft, writing – review and editing. Giacomo Strapazzon: conceptualization, investigation, writing – original draft, writing – review and editing, project administration, supervision. Markus Falk: conceptualization, investigation, writing – original draft, writing – review and editing, software, formal analysis, methodology, data curation. Hermann Brugger: conceptualization, investigation, writing – original draft, writing – review and editing, project administration, supervision, resources, validation, methodology. Katharina Hüfner: conceptualization, investigation, writing – original draft, methodology, validation, writing – review and editing, project administration, supervision, data curation.

Ethics Statement

The ethical committee of the Medical University Innsbruck was contacted, and the patients were informed about the study. The need for ethical approval was waived. All participants provided written informed consent for their participation in the study and for the publication of the data.

Consent

All participants provided written informed consent for the publication of the data.

Conflicts of Interest

The company Ortovox supported this study with logistical support and provided the avalanche transceivers and the transmitters buried in the snow as well as catering to the participants of the study. A collaborator of Ortovox assisted in the planning and the implementation of the study. No financial support was provided, and none of the coauthors received any form of nonfinancial compensation for their work.