Neurobehavioral changes in response to trauma-like and contextual anxiety in zebrafish (Danio rerio): the modulatory influence of music

Exposure to acute stress events, trauma, and intense anxiety can cause cognitive and memory deficits. This study investigated whether exposure to structured acoustic stimulation (classical music) modulates stress-induced cognitive impairments in adult zebrafish exposed to conspecific alarm substance (CAS), using the Y-maze test. Fish were divided into groups, two of which were subjected to 15 days of Vivaldi’s classical music, with sound intensity between 65 and 75 dB SPL with predominantly measured frequencies between 330 and 506 Hz. One group was exposed to CAS before the acoustic stimulation (CAS + M), while another received the stimulation prior to CAS exposure (M + CAS). Additionally, two groups were exposed to CAS only, one on day 0 (CAS 0d) and the other on day 15 (CAS 15d), without any acoustic stimulation. A control group (CTL) was included without experimental manipulation. Spatial memory performance was assessed using the Y-maze test, and social behavior was evaluated through the shoaling test. Serum cortisol levels were measured as an index of physiological stress. Exposure to CAS impaired exploratory behavior, evidenced by reduced time spent in the novel arm of the Y-maze. Consider this parameter as the main test outcome, post-stress acoustic stimulation restored exploratory performance to levels comparable to control fish, whereas prior exposure had no protective effect. Importantly, the CAS 15d group, which experienced the same temporal interval without acoustic stimulation, showed no recovery, supporting a specific effect of post-stress acoustic exposure on behavioral outcomes. CAS exposure significantly increased cortisol levels compared to controls; however, acoustic stimulation did not significantly attenuate this endocrine response. Together, these findings indicate that prolonged exposure to structured acoustic stimulation can modulate behavioral and cognitive responses following acute stress independently of cortisol normalization. The results support the use of zebrafish as an experimental model to investigate stress-induced cognitive dysfunction and highlight the relevance of acoustic environmental factors in post-stress behavioral modulation. Further studies are needed to elucidate the underlying neurobiological mechanisms and to determine the specificity of acoustic parameters involved.

The word trauma originated from Greek and refers to a wound, not only in the physical sense but also in the psychological sense (1). Psychopathological symptoms resulting from trauma may develop in individuals exposed to imminent life-threatening situations. Such experiences can overwhelm adaptive coping mechanisms and elicit a range of emotional, physiological, and behavioral responses, which may lead to states of intense anxiety, intrusive re-experiencing, avoidance of memories, and stimuli related to the traumatic experience, or even post-traumatic stress disorder (PTSD), all of which compromise the individual’s well-being and functionality (1, 2).

In conditions of intense anxiety, fear—which is, from an evolutionary perspective, an adaptive response to danger—becomes disproportionate to the actual threat, resulting in states of hypervigilance and a generalization of aversive responses (3). Individuals exposed to traumatic experiences may develop hyper-reactivity to mild stressors and a greater propensity for the formation of new fears, especially when the stimuli resemble the original traumatic experience (3, 4). The dysregulation of memory processes and prolonged suffering represent the debilitating and persistent nature of post-traumatic stress disorder. “Pathogenic memory,” referring to recurrent and involuntary memories that trigger intense stress, is linked to this dysfunction (5). Traditional cognitive theories of PTSD associate the memory alterations manifested as dissociative flashbacks and intrusive, distressing memories with the development and persistence of the disorder, as well as its duration (5, 6).

Although the concept of psychological trauma is traditionally defined within clinical contexts, experimental research necessarily employs operational definitions that can be applied to non-human organisms. In this sense, experimental models focus on trauma-related or trauma-like phenotypes, including persistent anxiety, contextual fear conditioning, and maladaptive behavioral plasticity following exposure to intense stressors (7). In zebrafish, exposure to conspecific alarm substance (CAS) represents an ethologically valid stimulus capable of inducing both acute defensive responses and longer-lasting behavioral alterations, thereby providing a useful framework for investigating trauma-related processes rather than transient stress alone (2, 8).

Given the complexity of psychological trauma, various therapeutic approaches have been explored. Among the most used interventions are psychotherapy, pharmacological treatments, and, more recently, complementary non-pharmacological strategies (9, 10). Music, for example, has been widely recognized as a resource capable of inducing relaxation, modulating mood, and reducing levels of stress and anxiety (11–13).

Human studies demonstrate that music therapy can improve cognitive and emotional symptoms in different clinical populations, such as individuals with Alzheimer’s disease, in whom it promotes improved verbal fluency, evokes autobiographical memories, and reduces behavioral agitation (14, 15). Additionally, evidence indicates that music can induce beneficial changes in neural circuits involved in emotional regulation, cognitive functions, and the stress response, making it a promising non-pharmacological intervention for improving the well-being of affected individuals (16, 17). Despite this evidence, assessing the effects of music on neurobehavioral responses associated with trauma remains a challenge. In this context, animal models have proven to be valuable tools, allowing for the systematic investigation of behavioral, cognitive, and neuroendocrine effects of these interventions in severe stress contexts (12, 18, 19). The zebrafish (Danio rerio) is a small freshwater teleost native to South Asia. This species has become a relevant experimental model due to its low maintenance cost, ease of handling and development, and the high degree of homology with the human genome (approximately 70–80%). Based on these characteristics, zebrafish represent an excellent choice for translational research (20–23).

Studies have shown that zebrafish are sensitive to stressful environmental manipulations and exhibit robust changes in behavioral and neuroendocrine parameters comparable to those observed in anxiety- and trauma-related disorders (24–27). Stress has also been reported to impact cognition, particularly memory (28, 29). Evidence indicates that intense aversive experiences, such as exposure to CAS, can trigger responses similar to those seen in these states (30). CAS is released by specialized epidermal cells called club cells (CCs), which rupture during predatory attacks, releasing into the environment an alarm signal that causes escape and freezing behaviors in the fish that perceive it. These chemical signals act as unequivocal indicators of danger and are crucial for the survival of the shoal. Given its ability to induce acute defensive and long-lasting behavioral changes, CAS exposure in zebrafish has been increasingly used as a translational model to investigate the neurobiological mechanisms underlying post-traumatic stress, intense anxiety, and related affective disturbances (31–34).

Based on this context, the present study proposes investigating the effects of music on neuroendocrine and neurobehavioral parameters, focusing on the assessment of cognitive performance and memory in zebrafish exposed to CAS. Rather than modeling music therapy itself, this study aims to examine how prolonged exposure to structured acoustic environments influences stress-related behavior and memory in a vertebrate model. The aim is to explore the modulatory potential of music on states compatible with traumatic experiences and intense anxiety, contributing to the understanding of neurobehavioral mechanisms associated with non-pharmacological interventions.

Ethical considerations

All experimental procedures were conducted in accordance with the guidelines of the Brazilian National Council for the Control of Animal Experimentation (CONCEA). This study was approved by the Animal Use Ethics Committee of the University of Passo Fundo (CEUA), Brazil (protocol CEUA 10/2023).

Animals

A total of 150 wildtype (5D strain) zebrafish (Danio rerio), approximately 120 days old, of both sexes (male:female ratio of 1:1), were included in this study, in an experimental design depicted in Figure 1. From these 150 fish, 64 fish were used to assess CAS-induced fear reactions (Figure 2), 80 were allocated across the five experimental groups (Figure 3 and 4), and six were used for CAS extraction. The animals were obtained from a commercial breeder (Recanto dos Peixes, Marau, Rio Grande do Sul, Brazil) and acclimated for 20 days at the Fish Physiology Laboratory of the University of Passo Fundo (UPF) (including five days of acclimation and 15 days of testing). The animals were kept in groups of four in 3 L glass tanks (20 × 15 × 14 cm, height × depth × width), under constant aeration and a photoperiod of 15 hours (light)/9 hours (dark), lights on at 9:00 pm, lights off at 6:00 am. The photoperiod was identical for all experimental groups and remained constant throughout the study. Acoustic stimulation sessions were performed during the light phase and were not temporally aligned with transitions between light and dark phases, thereby minimizing potential circadian confounds. Water temperature was maintained at 27.5 ± 1.3 °C, with pH 7.7 ± 0.08, dissolved oxygen at 5.6 ± 0.5 mg/L, and ionized ammonia <0.022 ppm. Water was partially (30%) changing every two days throughout the experimental period.

Baseline noise levels in the experimental site ranged from 50–55 dB SPL (with frequencies varying from 240 to 420 Hz) and consisted mainly of sounds generated by fish-rearing equipment such as aerators and water pumps. The animals in the control group were kept away from the room used for musical exposure in the experimental cohort (“treated”). The control group was housed in a separate room located approximately 15 m away from the room used for musical exposure. The rooms were physically separated by solid walls and closed doors, preventing sound transmission. No other sounds were presented to the control group, and their only difference compared to the experimental group was the lack of music and CAS exposure during the experiments. Music-exposed groups were housed in a separate room from non-music groups to prevent unintentional auditory exposure. Both rooms were maintained under comparable conditions (photoperiod 15 h light/9 h dark, water temperature 27.5 ± 1.3 °C, pH 7.7 ± 0.08, dissolved oxygen 5.6 ± 0.5 mg/L, and identical feeding schedules). However, we recognize that environmental variables such as ambient noise, light intensity, and human activity may have differed between rooms and could have influenced behavioral outcomes.

The zebrafish used on this study were randomly distributed (using the app RANDOM.ORG) into five experimental groups: the first group, designated as control (CTL), was not exposed to music or CAS. The second group (CAS 0d) was exposed to the alarm substance on day 0, without any musical treatment, the third group (M+CAS) received music until day 15 and was exposed to the alarm substance on day 15, the fourth group (CAS+M) was exposed to CAS on day 0 and “treated” with music until day 15 and the fifth group (CAS 15d) was kept without contact with music until day 15, when it was exposed to the alarm substance. The groups were subsequently submitted to the Y-maze test with the aim of investigating the effect of music exposure on neurobehavioral processes, as well as evaluating the cognitive performance and memory of the animals.

Exposure to music

Fish were divided into groups of 16 individuals, kept in glass tanks at a ratio of four fish per tank (four tanks per group). Music sessions were carried out as per a previous study of our research group (18) and consisted of a two-hour selection of Vivaldi’s classical music, with sound intensity between 65 and 75 dB SPL with predominantly measured frequencies between 330 and 506 Hz, as assessed by real-time monitoring using a sound level meter application, although transient higher-frequency peaks were occasionally observed. While frequency range and intensity were controlled, we did not perform underwater acoustic validation (e.g., hydrophone measurements of sound pressure levels, tank resonance, or vibration artifacts). Therefore, the precise acoustic stimulus perceived by the fish remains uncertain.

The term ‘classical music’ is used here as an operational descriptor of a structured acoustic stimulus and does not imply semantic or cultural music perception in zebrafish. The intensity and frequency of the music, as well as the background noise, were measured out of the water using the Sound Level Meter app (available online at https://play.google.com/store/apps/details?id=com.bolshakovdenis.soundanalyzer) on a Samsung Galaxy S22 smartphone. Daily music exposures occurred at two times: in the morning at 8:30 a.m. and in the afternoon at 5:30 p.m. To minimize the effect of hunger on fish behavior, all individuals were fed twice daily, 30 minutes before each music session, including the control non-exposed fish.

Extraction and exposure to CAS

For CAS extraction, the procedure described by Maximino et al. (35) was followed. Briefly, 6 animals were collectively transferred to a test tank (25 × 15 × 6 cm, length × height × width) containing 1.5 L of distilled water, with a 10-minute acclimation period. Another donor animal was cold-anesthetized and euthanized by decapitation, and excess blood was removed with absorbent swabs. Fifteen shallow cuts were made in the trunk epidermis, and the area was submerged in 10 mL of distilled water; 7 mL of this suspension were added to the exposure tank water, where all 6 animals were exposed for 6 minutes. Control groups were exposed only to distilled water. The quality of CAS extraction was assessed by observing changes in shoaling behavior (ZBC 1.141) during exposure. Photos were taken every 15 seconds (a total of 20 shots per group), and the Delaunay distance between fish was calculated for each photo using the Fiji distribution of ImageJ (36). To calculate the proximity between fish (inter-fish distance, distance to nearest and farthest conspecific) and group dispersion (shoal area), the screenshots were calibrated to the tank size and each fish was marked to allow automated quantification. Possible concerns about possible variability were minimized, as all CAS-exposed groups were treated with extracts prepared by the same operator, ensuring consistency across exposures.

Aversive responses

Behavioral responses to CAS were evaluated in a separate prior experiment. A total of 32 zebrafish were tested in the same tank used for exposure, immediately after the addition of CAS. Another 32 fish were tested without CAS exposure and served as controls. The fear-related responses in the zebrafish groups were assessed by the presence or absence of alarm reactions. According to the zebrafish behavioral catalog (ZBC) (37), freezing is defined as the cessation of movement, except for the eyes, associated with a high frequency of opercular movements (ZBC term 1.68). Erratic movements consist of successive rapid accelerations with abrupt changes in the direction of movement (ZBC term 1.51). All aversive behaviors were observed and scored within the 5-minute exposure period in this tank. Behavioral categories (freezing, erratic swimming, bottom-dwelling, and shoal cohesion) were classified using predefined criteria. Interobserver agreement was assessed using Cohen’s kappa coefficient to evaluate the reliability of the qualitative observations.

These outcomes were observed according to the following scores during the 5-minute observation period: (1) absent; (2) a few episodes (1–3 episodes per group); (3) moderate responses (4–6 episodes per group); (4) increased responses (7–9 episodes per group); (5) exacerbation of fear-like behaviors (more than 10 episodes per group). This scale-based scoring system was chosen based on the values that can be detected when zebrafish are placed in a novel tank, and it is a simple method to detect aversive behaviors in a shoal of zebrafish (8, 38–41).

Time-dependent sensitization

Time-dependent sensitization (TDS) is a phenomenon in which the exposure of laboratory animals to an intense and inescapable stressor leads, after an incubation period, to a delayed amplification of responses to a distinct (heterotypic) stressor (2). To assess the incubation component of this process, after the administration of the stimulus and exposure to the alarm substance, the same animals were immediately transferred to a 4 L tank with fresh water (i.e., free from alarm substance), where they remained for 24 hours before the Y-maze test.

Spatial memory test – Y-maze

The Y-Maze behavioral test (Y-Maze) used to assess the effects of music on cognition and spatial memory in zebrafish was performed in all experimental groups on day 16, corresponding of after the 15 days of repeated music exposure. The position in the arm of the Y-maze was considered an index of spatial memory (42). The fish were tested in a tank with three arms measuring 25 × 8 × 15 cm (length × width × height). Different geometric shapes (squares, circles, and triangles) were used as visual stimuli and placed on the outer wall of each arm, and the remaining area was covered with black plastic. All arms were identical in size, illumination, and visual complexity, differing only in the geometric cues, in order to avoid positional or contextual bias. The arms of the Y-maze were randomly assigned: the initial arm, where the fish started the test; the novel arm (unavailable during the initial test but open during the second test); and the permanently open arm. Under control conditions, fish did not exhibit a consistent preference for any specific arm, indicating the absence of intrinsic arm bias. The center of the Y-maze was considered a neutral area and was therefore not included in the analysis. The task consisted of two phases with a one-hour interval between them. In the first phase (5-minute training), the fish could explore the initial arm and the permanently open arm. In the second phase (5-minute test), the animals were placed in the initial arm and were allowed to freely access all three arms of the maze, including the novel arm. Fish behaviors were recorded by a Logitech HD Webcam C525 camera positioned above of the test aquarium. To avoid interference from human activity, the operator left the experimental room immediately after the fish were released into the test arena. Additionally, a blinded design was employed, whereby the operator was unaware of the treatment assigned to each fish. After, videos analyzed using ANY-maze ® software (Stoelting CO, USA).

Determination of cortisol levels

Individually, after each behavioral trial, the fish were anesthetized in cold water and euthanized by decapitation. They were immediately frozen in liquid nitrogen and subsequently stored at -20 °C until tissue cortisol extraction was performed. The extraction was carried out following the protocol described by Sink et al. (43), using the whole body of the previously frozen samples. The extracts obtained were resuspended in PBS buffer, and cortisol levels were determined by enzyme-linked immunosorbent assay (ELISA) using commercially available cortisol kits (AccuBind Elisa Microwells, Monobind Inc, USA) following the manufacturer’s instructions.

Statistical analysis

The area in cm and the distance between the fish were calculated, and the endpoint results were expressed as means ± standard error and analyzed by the Student’s t-test. For the spatial memory test, the data were calculated as median ± interquartile range and analyzed by the Kruskal-Wallis test followed by Dunn’s multiple comparisons test, since all behavioral as well cortisol datasets did not follow a normal distribution verified by the Brown-Forsythe and Bartlett’s tests. All data were considered significant when p < 0.05. Effect sizes were estimated using eta squared (η 2), calculated as the proportion of total variance explained by the experimental factor. The magnitude of the effect was expressed as Cohen’s f, derived from η 2 using the formula f = η 2 / (1 − η 2). Statistical power (1 − β) was estimated based on sample size, number of groups, and effect size, with a significance level of α = 0.05. A power threshold of 0.80 was considered adequate to detect differences among groups.

The differences in group sizes in the behavioral evaluation (Y-maze) resulted from the videos lost during the behavioral analysis. These videos were lost due to insufficient contrast, which prevented the tracking software from reliably identifying and following the fish. As a result, the program could not generate accurate trajectories, and these videos were excluded from the analysis. For cortisol measurements, 16 fish from each group were euthanized and preserved for analysis. Variations in group sizes in this case were due to sample losses during preparation. All statistical analyses were conducted using GraphPad Prism version 8.0.0 for Windows (GraphPad Software, Boston, MA, USA; www.graphpad.com), and values of η 2 or f, and statistical power were reported for each dataset and expressed in the results description.

Shoaling behavior

Interobserver agreement analysis revealed substantial to almost perfect concordance for the qualitative assessment of aversive behaviors during CAS exposure, with Cohen’s kappa values ranging from 0.82 to 0.95 across behavioral categories. These results indicate high reliability of the observational data.

Acute exposure to CAS caused significant changes in the shoaling behavior of zebrafish, reducing the distance between fish in the test apparatus (A) as well as the shoal area (B), as can be seen in Figure 2. The Student´s t test revealed a significant difference between zebrafish exposed to CAS (day zero) and those in the control group (CTL), which were exposed to distilled water. Analysis of distribution in the tank indicated significant effects on the spatial organization of the exposed population, which remained deeper in the tank (Figure 2A; p < 0.0001, t = 8.079, Cohens’ d = 0.94, power > 0.99) and showed a significant reduction in the distance between fish (Figure 2B; p < 0.0001, t = 5.256, Cohens’ d = 1.39, power > 0.99).

Effect of music on alarm substance

Regarding the time spent in the novel arm (Figure 3A; p = 0.0003, H = 24.21, η² = 0.28; statistical power 90-95%), fish exposed to CAS spent significantly less time in the novel arm compared to the control group (p < 0.01). Fish treated with music after CAS exposure (CAS+M) spent more time in the novel arm than fish exposed only to CAS (p < 0.001) as well as fish exposed to music before CAS (M+CAS) (p < 0.05). Fish in the CAS-15d group did not show this reversal in time spent in the novel arm.

Regarding the latency to first entry into the novel arm (Figure 3B; p = 0.8123, H = 1.522, η² = 0.12; statistical power 70-75%), no significant differences were observed among the groups.

Regarding the number of entries into the novel arm (Figure 3C; p = 0.0184, H = 11.87, η² = 0.08; statistical power 60-65%), fish treated with music before CAS exposure (M+CAS) showed a significantly higher number of entries compared to fish exposed only to CAS (p < 0.01).

Finally, regarding the distance traveled in the novel arm (Figure 3D; p = 0.0279, H = 10.89, η² = 0.18; statistical power 80-85%), fish treated with music both before and after CAS exposure traveled a greater distance in the novel arm than negative control fish (p < 0.05), a pattern similar to that observed in the CAS 15d group (p < 0.05).

Effect of music on cortisol levels

Analysis of cortisol levels revealed statistically significant differences between the experimental groups (Figure 4; p = 0.0002, H = 21.74, η² = 0.20; statistical power 85-90%). Fish exposed to CAS showed a significant increase in cortisol levels compared to the negative control group (p < 0.001), which exhibited the lowest concentrations of the stress hormone. Animals previously exposed to music for 15 days before CAS also showed significantly higher cortisol levels compared to the negative control group (p < 0.05). Although a tendency toward reduced cortisol levels was observed in the groups exposed to music after CAS, compared with the CAS-0d group, the difference was not statistically significant, a pattern also identified in CAS-15d. However, the cortisol values in these groups were indistinguishable from those of the control group, suggesting that music partially modulated the stress response.

In this study, we aimed to explore the modulatory potential of music on states compatible with traumatic experiences and intense anxiety. We show that exposure to the alarm substance (CAS) caused behavioral changes in the Y-maze test, markedly reducing exploration of the new arm, as indicated by the shorter time spent in that segment, and that treatment with music after exposure to CAS reversed the damage caused by the substance (Figure 3A), restoring exploration levels to patterns similar to those of control fish not exposed to CAS. This effect of music is reinforced by the time spent in the new arm by fish in CAS-15d, which underwent the same interval between CAS and the test, but without music. This result is important to rule out a possible extinction of the behavior caused by CAS unrelated to exposure to music, since in these fish without musical treatment, the reversal of the effect was not observed.

Importantly, while the immediate behavioral responses observed following alarm substance exposure are consistent with acute stress and anxiety, the persistence of altered exploratory patterns and contextual avoidance in the Y-maze, as observed under the present experimental conditions, suggests the involvement of memory-dependent processes (5). These processes extend beyond transient stress responses and are typically associated with trauma-like states, in which aversive experiences lead to maladaptive behavioral plasticity and persistent alterations in cognitive and emotional processing.

Although the present design does not allow discrimination between music-specific effects and non-specific acoustic mechanisms such as masking or frequency sensitivity, the results demonstrate that prolonged exposure to structured acoustic stimulation can modulate behavioral and cognitive responses following acute stress in zebrafish. Importantly, this behavioral and cognitive improvement occurred in the absence of significant cortisol normalization, indicating a dissociation between endocrine and behavioral responses to stress.

One plausible alternative mechanism is noise masking. Laboratory environments are characterized by continuous mechanical noise generated by pumps and aerators, which may act as a chronic stressor. Structured acoustic stimulation at higher and more regular sound pressure levels may attenuate the impact of this background noise, thereby reducing baseline stress responses. In addition, the temporal predictability inherent to structured acoustic stimuli may contribute to stress reduction by decreasing environmental uncertainty and facilitating habituation. Predictable auditory patterns have been shown to modulate behavioral and physiological stress responses in fish, independently of semantic content (18, 44). Finally, differences in frequency composition between the acoustic stimulus and baseline noise may also play a role, either through novelty or attentional modulation. Because frequency-matched control stimuli were not included, the contribution of this factor cannot be isolated in the present study. Future experiments incorporating white noise, naturalistic soundscapes, or frequency-specific controls will be necessary to disentangle these mechanisms (18, 45).

On the other hand, music administered before exposure showed no effect on the recovery of cognitive impairments caused by CAS. These findings suggest that music may have an anxiolytic or restorative effect (by increasing the time spent in the new arm) in zebrafish after an episode of acute stress, in this case caused by exposure to CAS. In fact, CAS has been reported as a stressor (8, 46) and triggers anxiety-like behaviors (8, 25, 40, 46).

Stress responses, both central and peripheral, can be activated by any new or potentially threatening environmental stimulus, triggering behavioral changes that promote adaptation and aim to maintain and restore balance (47–50). Exposure to uncontrollable stressors, whether homotypic, heterotypic, or chronic, can amplify behavioral and neuroendocrine responses through a process of sensitization (2). When the intensity of the stressor is high, this sensitization can expand to stimuli not directly related to the initial event and remain for long periods. Even a single brief exposure may be sufficient to trigger this generalized response (51, 52).

As a defense strategy to increase its chances of survival in adverse situations, zebrafish tend to intensify aversive behaviors, such as episodes of freezing, erratic swimming, remaining at the bottom of the environment, and greater social cohesion (8, 25, 37). Although these behaviors were not subjected to parametric statistical analysis, the high interobserver agreement supports the reliability of the qualitative behavioral assessment and reinforces the interpretation that CAS exposure elicits robust aversive responses in zebrafish. Treatment with music fifteen days prior to exposure to CAS did not reduce the aversive responses observed during the period of contact with the substance, since exposed fish had more episodes of freezing and greater shoal cohesion (Figure 2), indicating a clear aversive response, which highlights the potential for CAS to cause anxiety and stress in zebrafish (8, 40). Notably, Amaral et al. (53) reported that zebrafish maintained under prolonged environmental enrichment, a condition characterized by sustained sensory and structural stimulation, displayed exacerbated anxiety-like behaviors in the novel tank test following acute stress exposure. These findings support the hypothesis that prior exposure to highly comfortable or enriched conditions, such as continuous musical stimulation, may sensitize individuals to subsequent aversive events, potentially amplifying the perception of threat and resulting in intensified defensive responses, including increased freezing and social cohesion.

Time-dependent sensitization involves two processes: initiation and consolidation (54, 55). Initiation occurs in the initial peritraumatic events, while consolidation represents the “incubation” of the stressor’s effects. These processes involve distinct neurochemical mechanisms: initiation depends on NMDA and CRF1 receptors (56, 57), while consolidation involves β-adrenergic, glucocorticoid, mineralocorticoid, 5-HT1A, 5-HT2, CCKB, and CRF1 receptors (56). In the present study, the experimental design was focused on the incubation phase of this process, assessed through behavioral testing 24 h after CAS exposure. This evidence suggests an experimental approach in which pharmacological interventions before stress affect initiation, evaluating the permissive role of the system under study, while post-stress interventions influence consolidation, allowing investigation of their suppressive or stimulatory functions (2).

Since the trauma initiation process refers to the initial events that occur shortly after exposure to an intense or traumatic stressor, it is possible that music may not be able to reduce the risk of developing post-traumatic disorders. However, in our study, treatment with classical music after exposure to CAS showed a pronounced effect on the memory consolidation process after the traumatic event. Previous studies indicate that Vivaldi’s classical music can positively modulate neuronal activation in the hippocampus and enhance spatial cognition in rodents, possibly through the regulation of BDNF (58).

In studies already conducted by our group, zebrafish showed significant effects of music exposure on the brain and behavior. For example, adult fish exposed to Vivaldi’s music (65–75 dB) for two hours, twice a day, for 15 days, exhibited reduced anxiety-like behavior in the novel tank task and light-dark test. In addition, lower peripheral levels of pro-inflammatory cytokines, such as interferon-gamma (INF-γ), were observed (18).

It is important to emphasize that exposure to traumatic experiences are necessary for the development of post-traumatic disorders, but trauma alone is not sufficient for their development (59, 60). Only 25% to 30% of individuals exposed to a traumatic event develop PTSD (61). This individual variability is extremely relevant for the creation and validation of translational models of PTSD (60). The evaluation of the behavioral response of the animals in the experiment in isolation and the classification into extreme and minimal behavioral response is important to better modulate experiments that simulate the effects of PTSD and is one of the objectives of future studies by our group.

When exposed to stimuli that evoke fear-like or anxiety-like, zebrafish may exhibit a wide range of well-defined and quantifiable behaviors, including reductions in exploratory behavior, increased scototaxis (preference for darkness), geotaxis (preference for substrate), thigmotaxis (preference for peripheral areas), freezing (complete cessation of movement, except for the opercula and eyes), and erratic movements (zigzag swimming pattern with unpredictable direction and rapid, sudden changes in direction) (62, 63). These behavioral phenotypes are analogous to those of rodents and humans (37). Among the behavioral phenotypes in zebrafish (37) relevant to the translational modeling of clinical and symptomatic criteria for post-traumatic disorders, changes in memory task performance may be correlated with mnemonic deficits associated with cognitive dysfunction and exacerbated fear characteristic of the disorder, as described in the DSM-5 (64).

Still in this context, in our study we were able to assess the cognitive effects of acute stressors using the Y-maze test, as performed by Kim et al. (65), who used an adapted version of the cross maze as a way of exploring spatial memory. The reversal of deficits and restoration of exploration levels observed with musical treatment after exposure to CAS is valued when compared to the findings of Gaikwad et al. (66). These authors reported a significant impact on target arm preference in a cue-based memory task, suggesting that psychological stressors increase the time required to associate a learned cue with the location of food and cause confusion in the search. In the behavior observed in the spatial memory task, stress induced by a predator reduced the time and frequency of entries into the new arm, while fish exposed to the alarm pheromone swam less. In contrast, in the present study, fish treated with classical music showed the opposite pattern.

Furthermore, we clearly demonstrated that exposure to CAS elevates cortisol levels in zebrafish, as evidenced by significantly higher values in the CAS-0d group compared to the control group. This increase reflects the activation of the hypothalamic-pituitary-adrenal (HPA) axis, functionally analogous to the HPA axis in mammals, which is responsible for coordinating the neuroendocrine response to stress. Activation of the HPA axis induces a signaling cascade that culminates in the release of cortisol, allowing it to mobilize the body’s physiological resources in response to a stressor. Under normal conditions, elevated cortisol induces a negative feedback mechanism that inhibits further release of the hormone, regulating the stress response and promoting the restoration of homeostasis (67, 68). Furthermore, CAS has already been used as a stressor in zebrafish by other authors, who also demonstrated elevated whole-body cortisol levels through exposure to this substance (69).

Although a downward trend in cortisol levels was observed in zebrafish exposed to music before and after exposure to CAS, compared to the group exposed only to CAS, this difference did not reach statistical significance. This suggests that music, under the conditions tested, was not sufficient to prevent or completely reverse the physiological stress triggered by this aversive stimulus. Marchetto et al. (44) evaluated the effects of music as auditory environmental enrichment in zebrafish socially isolated for 24 hours and observed no significant differences in cortisol levels between treated and untreated groups. It is important, however, to consider the nature and intensity of the stressors involved in each study. In our protocol, CAS functions as a model of acute threat with ecological relevance, features commonly associated with trauma-related and trauma-like conditions, rather than transient stress alone. In contrast, the stressor used by Marchetto et al., although valid and ethologically relevant, was social isolation, which, despite affecting highly social species such as zebrafish, does not directly involve the perception of physical threat or imminent danger. However, the authors reported behavioral changes associated with reduced anxiety, indicating that music can modulate emotional aspects without necessarily affecting the HPA axis directly. These findings indicate that behavioral and cognitive modulation by music can occur independently of full HPI axis normalization, a phenomenon previously reported in zebrafish and other vertebrate models.

In the present study, we used the same musical style and frequency as the aforementioned authors, which reinforces the hypothesis that, although music may have modulating effects on behavior, it may not significantly affect the endocrine response to stress under the conditions evaluated. Additionally, it is possible that mechanisms other than the neuroendocrine pathway are involved in this response, such as those mediated by neuroimmune processes (18). It is also known that different musical styles, sound frequencies, and exposure times can activate different circuits in the central nervous system, affecting more than just animal behavior (12, 18).

There are other limitations to our study. For example, we used the same strain (wild short-finned fish) from a mixed population with a 1:1 male/female ratio of the same age group (adults around 120 days old). Thus, the present study does not evaluate individual differences in sex (specifically referring to possible sex differences in behavioral outcomes), strain, or age in the cognitive responses of zebrafish to music and exposure to alarm substances, which may limit the translational projection of the results to human phenotypes. However, these possible limitations do not diminish the relevance of the data presented here, since zebrafish is already an established model for neurochemical and behavioral studies (20) and Vivaldi’s classical music already has a considerable body of evidence regarding its effects on the central nervous system and behavior (18, 70, 71).

Overall, the results reinforce the potential of zebrafish as an experimental model for studying the impacts of intense trauma and anxiety on cognition, memory, and cortisol levels, and suggest possible music-based therapeutic approaches. A detailed underwater acoustic characterization, including particle motion and spatial variability, was not performed in the present study and represents an important direction for future investigations aimed at elucidating the specific acoustic mechanisms underlying the observed behavioral effects. Additionally, no direct neurobiological markers (e.g., BDNF, cytokines, monoamines) were assessed in the present study. Therefore, the involvement of neuroplastic or neuroimmune pathways remains hypothetical and should be directly investigated in future studies, which may further elucidate the underlying neurobiological mechanisms and the translational applicability of these effects.

Although our findings suggest potential cognitive benefits associated with the interventions tested, we recognize important limitations that constrain causal inference. Specifically, the study cannot fully distinguish whether the observed effects are specific to music, related to general acoustic stimulation, or due to noise masking. In addition, some outcomes presented relatively low statistical power, which weakens the strength of the conclusions, However, in the main endpoint of the Y-maze test (time spent in the new arm) and in cortisol data, the statistical power exceeded 90%, which is considered ideal. Taken together, these issues call for cautious interpretation of our results. Rather than definitive evidence of restoration of cognitive function, our study should be viewed as opening a promising avenue for future investigations that require larger sample sizes and mechanistic approaches to better substantiate these effects.

In this study, we demonstrated that treatment with sound intensity between 65 and 75 dB and frequencies ranging from 330 to 506 Hz of Vivaldi’s classical music, when administered after exposure to CAS, was able to reverse the cognitive deficits induced by acute stress. On the other hand, prior exposure to music did not result in evident cognitive benefits. These findings reinforce the hypothesis that music can act as, at least in part, a modulator of stress responses, especially by positively influencing memory consolidation after traumatic experiences.