What does the anemonefish say?: decoding Amphiprion percula’s acoustic communication

What does the anemonefish say?: decoding Amphiprion percula’s acoustic communication | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL Ecology and Evolution This is a preprint and has not been peer reviewed. Data may be preliminary. 30 May 2025 V1 Latest version Share on What does the anemonefish say?: decoding Amphiprion percula’s acoustic communication Authors : Lucia Yllan 0000-0002-5992-1133 [email protected] and Theresa Rueger 0000-0002-2105-6412 Authors Info & Affiliations https://doi.org/10.22541/au.174861208.89315690/v1 334 views 321 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Acoustic communication is crucial in many animal taxa for mate selection, foraging coordination, and predator avoidance. While fishes represent a significant portion of vertebrate biodiversity, their acoustic communication remains understudied. Anemonefishes (Amphiprion spp.) exhibit complex social behaviours and vocalisations, yet their acoustic signals have primarily been studied in laboratory settings, limiting our understanding of their natural communication. In this study, we examined the vocal behaviour of wild Amphiprion percula in Papua New Guinea, collecting in situ audio and video recordings to assess the role of vocalisations in social interactions. Our findings suggest that A. percula produces distinct vocalisations across multiple social contexts, with variations in acoustic parameters influenced by behaviour and social rank. We highlight the importance of acoustic communication in interspecific and intraspecific interactions and its role in territorial defence, and the classification of vocalisations in behavioural categories. Additionally, we discuss the potential impact of anthropogenic noise, such as motorboat activity, on anemonefish communication through acoustic masking. These results emphasize the importance of in situ studies for understanding fish bioacoustics and suggest that acoustic signals play a key role in maintaining social hierarchy and cohesion in anemonefish groups. Future research should explore species-specific vocal diversity in situ and assess the ecological implications of noise pollution on coral reef fish communication. Introduction Animals use acoustic signals to find potential reproductive partners, coordinate foraging strategies or alert each other of the presence of predators (1–3). Despite fish representing 35% of the total chordate biomass and half of vertebrate species (4), knowledge about acoustic communication in fishes is limited compared to other taxa. Fish are known to rely on acoustic cues at embryonic, larval and adult stages (5–7) Free-swimming larvae of coral reef fishes use acoustic cues to localise and discriminate habitats for settlement which is key for survival (8–10). Acoustic signalling is also important in adult fish for foraging, partner selection, coordination of gamete release, territory defence or social interactions (11). For example, in midshipmen fish, Porichthys sp ., courtship behaviour involves humming sounds from nesting males to attract females (3,12), and the use of acoustic signals in agonistic interactions is thought to reduce conflicts in group living species (13). However, how fishes use acoustic signals in natural interspecific and intraspecific interactions, and how these interactions may be disturbed by anthropogenic noises is still relatively unknow. Understanding the importance of acoustic signalling is key to predict the potential impact of noise pollution within social fish groups. Anthropogenic noise is defined as the sounds produced by humans or human activities. It is referred to as noise as it is unwanted sound that interferes with acoustic signals and serves no biological function (14). Anthropogenic noise pollution is ubiquitous and negatively affects fish in diverse ways (15) For example, noise pollution interferes with larval settlement (16); reduces fish attention which increases mortality risk by predation (17,18); and decreases foraging behaviour effectiveness (19). The overlap and interference between anthropogenic noise and biological acoustic signals produce a phenomenon known as masking, which disrupts acoustic communication between individuals (20,21). However, most of what we know about acoustic signalling in fish comes from laboratory studies, which greatly limits our understanding of how these signals may be disrupted by anthropogenic noise occurring in wild environments. Most studies of fish acoustic communication rely heavily on controlled laboratory settings, which may not capture the complexity of fish behaviour and vocalisations in natural environments. Using captive fishes to study acoustic signalling can be problematic, since sounds are potentially distorted through reflection off aquarium walls (22). Dominant frequency and sound duration are altered when in tank settings compared to open water recordings (23). In addition, laboratory settings often constrain animals’ ability to interact with their environment and alter their natural social dynamics (24). Behavioural studies conducted in the wild provide a more accurate understanding of the intricate interactions between individuals, their social structures, and their environment, capturing the complexity of these relationships in natural contexts (25–27). In situ studies of wild fish acoustic signals, associated behaviours and social and ecological context are needed. A good model system for how fish use acoustic signals in the wild are anemonefishes (genus Amphiprion ), which can produce vocalisations and display complex social and interspecific behaviours (28,29). Anemonefishes have an obligate mutualistic relationship with their sea anemone host, which makes them extremely site attached (30,31), allowing for repeated measures and understanding of the specific social and ecological context that vocalisations are produces in. Anemonefish groups have a strict size-based social hierarchy where individuals queue for the breeding position (32–35). The largest anemonefish is the dominant female and the second largest is the sub-dominant male, the breeding pair, which can cohabit non-reproductive subordinates (36,37). Acoustic cues may be key to maintain the strict hierarchical social structure within groups (38), by signalling dominance/submission and allowing group members to be differentiated. Anemonefishes can hear sounds between 75 and 1800 Hz (39) and produce sounds using their jaw teeth and vibrations from their rib cage, which generates variation in their vocalisations related to their body size (13,39–42). Previous studies in A. akallopisos have found that pulse duration increases with body size while dominant frequency decreases (39,41). The vocalisations of anemonefishes are formed by short duration “pulses” or “pops”, which in combination create a “train” of sounds. For example, A. clarkii produces eight consecutive pulses of sound that range between 450 and 800 Hz (40,41). Anemonefish sounds have been classified into two main categories: aggressive sounds accompanied by threat postures and charge-chases, and submissive sounds accompanied by body shakes (39,41,43). Colleye and Parmentier (2012) also noted no presence of sounds during reproductive or spawning activities. While broad categories such as agonistic and submissive sounds have been identified, there remains a lack of fine-scale understanding of how acoustic signals vary across specific behaviours and social interactions. This understanding will be particularly important to predict the impact of anthropogenic disturbance such as motorboat noise pollution, which increasingly threatens small bodied, site attached coral reef fishes (44,45). Therefore, a comprehensive classification of anemonefish sounds, and their associated biological functions, is essential to gain a deeper understanding of their communication and behaviour. In this study, we aim to fill the knowledge gap on the role of sound in wild anemonefish social groups and behaviour, using groups of orange anemonefish ( Amphiprion percula ). For this study we collected video and audio data in situ of a wild population. Based on video and audio data from 9 wild groups of A. percula in Papua New Guinea, we suggest changes to the classification of anemonefish vocalisations and test the following hypotheses: i) Vocalisations are used in a range of different behavioural contexts such as in interspecific interactions; ii) There are distinct differences between vocalisations related to certain behaviours, we predict that different behaviours may differ in the number of pulses of the vocalisations that accompany them (46); iii) The frequency of the vocalisations is determined by individuals body size (41); iv) anemonefish vocalisations can be masked by anthropogenic noise, as both overlap in frequencies (15) Methods Study site The study was conducted from February to April 2023 in Kimbe Bay, Papua New Guinea (5°30’ S, 150°05’ E). A total of 142 vocalisation trains (594 individual vocalisations) were recorded in situ from 9 wild A. percula groups of two ( N= 1), three ( N= 3), four ( N= 3) and five ( N= 2) individuals, associated with the anemone Heteractis magnifica . The groups were distributed across seven different inner reefs in Kimbe bay. Behavioural and acoustic data collection Each anemonefish group was recorded for 1 hour using GoPro Hero 9 action cameras and AudioMoths version 1.7.1 (acoustic logger) which were placed by SCUBA divers in front of groups on stationary tripods set close to the anemone to obtain video and audio data, respectively. After set-up, SCUBA divers would leave the area to ensure undisturbed recordings. The set of cameras and acoustic loggers is non-invasive and has very low impact on anemonefish and other surrounding species welfare and behaviour. After the recordings, all individuals of the group were caught with hand nets and measured underwater with callipers to obtain their standard length (SL) and total length (TL) in mm. During data collection in the field, motorboat noise was also recorded by the Audiomoths deployed in the proximity of the anemonefish groups. Behavioural and acoustic analysis For the analysis, we selected 12 minutes of continuous footage and audio from the raw recordings, excluding sections where anthropogenic noise was present (such as diver or boat noise) and 5 minutes after noise disturbance if any was present. The 12 minutes footage of the groups was watched to score individual behaviours using BORIS (v. 7.12) (47), and audio data was listened to classify individual vocalisations using Audacity, both open-source software. The behaviours scored were divided into categories (based on (28)): i) aggression; ii) submission, iii) neutral interactions. Vocalisations were labelled in relation to the behaviours that were displayed by the vocalising fish, their rank and the rank the interacted with in the case of intragroup interactions. The target individual receiving that behaviour was also recorded in intragroup interactions. Aggressive behaviours were divided into intragroup aggression and aggression towards heterospecific, which was also subdivided into defence (aggressive displays towards egg predators such as Labridae and anemone predators such as Chaetodontidae) and competition (aggressive displays towards food competitors such as other Pomacentridae) (29). Neutral interactions are defined as non-agonistic interactions between individuals that are within one body length of each other, such as swimming together, meeting, following or touching (28). Individuals were visually recognised by natural markings and size, a method widely used in anemonefish research (48). This genus has unique patterns of colour bands that vary between individual, and strict size ratios between ranks that allows easy identification of individuals in the hierarchy. Unwanted noise interferences present in the acoustic data were removed using a low-frequency filter in Audacity. Labelled acoustic data in format .wav was cut in smaller clips with Audacity and imported to Raven Pro v 1.6 to extract the spectrogram parameters of all vocalisations, which were exported in selection tables. Statistical analysis All statistical analyses were performed in R v. 4.4.2 (49). Acoustic data visualization, including the creation of spectrograms and oscillograms, was done using the seewave package (50). To analyse the spectrogram parameters, we imported selection tables from Raven using the warbleR and Rraven packages (51,52), which allowed us to extract and quantify specific acoustic features like frequency, pulse duration, and entropy for further analysis. To explore the covariation of vocalisation parameters and identify underlying patterns in the data, we conducted principal component analysis (PCA) using the factoextra package (53), which allowed us to reduce the dimensionality of the data by grouping correlated acoustic features (such as sound frequency, duration, pause duration, and pulse number) into principal components, making it easier to examine overall trends across different behavioural contexts. To examine the effects of social rank, behavioural category, and standard length (SL) on vocalisation characteristics, we fitted generalized linear mixed models (GLMMs) using the lme4 package (54). Separate models were fitted for each response variable, including mean frequency, dominant frequency, peak frequency, pulse duration, entropy, and number of pulses. The following predictors and their two-way interactions were included: social rank, behavioural category, and SL. Group ID and pulse train were included as random effects to account for non-independence of behaviours within the same group and of sound pulses within the same vocalisation train. The distribution of each response variable was tested, and the most appropriate distribution was selected for each variable. Akaike’s Information Criterion (AIC) (55) was used for model selection, and log-likelihood ratio tests were applied when ΔAIC was less than 2 to determine the best-fit model. Pairwise comparisons were conducted using the emmeans package (56) with Tukey’s method for p-value adjustment, which allowed us to compare the effects of different ranks and behavioural categories on vocalisation characteristics while controlling for multiple comparisons. Finally, the performance package was used to calculate the marginal R² (variance explained by fixed effects) and conditional R² (variance explained by both fixed and random effects), following Nakagawa’s R² method (57). The context of vocalisations in A. percula We identified five behaviours that were associated with vocalisations of A. percula : Interspecific aggression (aggressive displays to other fish species), intragroup aggression (aggressive displays to other fish within the social group), neutral interactions (non-agonistic interactions between members of the social group), submission (submissive displays to other members of the social group) and territorial displays, in which individuals would get close to the edge of the anemone and produce vocalisations without a direct interaction with another fish (see Table 1). Table 1. Table showing the mean (+/- standard deviation (sd)) of frequency, peak frequency, dominant frequency, entropy, duration, pause duration and pulse number of all the behaviours scored in Amphiprion percula. Mean ± sd Mean ± sd Mean ± sd Mean ± sd Mean ± sd Mean ± sd Mean ± sd Interspecific aggression 0.549 ± 0.141 0.576 ± 0.172 0.526 ± 0.162 0.579 ± 0.115 0.037 ± 0.027 0.102 ± 0.156 8.189 ± 5.346 Intragroup aggression 0.702 ± 0.074 0.720 ± 0.143 0.684 ± 0.120 0.566 ± 0.130 0.045 ± 0.024 0.246 ± 0.449 3.444 ± 2.007 Neutral interactions 0.568 ± 0.108 0.581 ± 0.185 0.550 ± 0.147 0.602 ± 0.114 0.044 ± 0.019 0.169 ± 0.313 5.846 ± 3.804 Submission 0.619 ± 0.205 0.595 ± 0.218 0.570 ± 0.214 0.563 ± 0.092 0.037 ± 0.013 0.156 ± 0.303 9.123 ± 5.545 Territorial displays 0.580 ± 0.141 0.599 ± 0.170 0.553 ± 0.157 0.570 ± 0.096 0.041 ± 0.017 0.104 ± 0.182 5.500 ± 2.930 We found that A. percula vocalisations are composed of pulses of sounds with a frequency range between 0.200 and 1.080 kHz and a duration between 0.007 and 0.361 ms (Fig.1). The pulses of sound can be produced individually or in a train of consecutive pulses that can range from 2 to 20 pulses, with an average of 0.036 ms of pause between pulses. Figure 1. Spectrograms of A.percula vocalisation trains. The y-axis represents frequency (in kHz), while the x-axis represents time (in seconds). The spectrogram in the left (a) correspondence with a submissive vocalisation while the spectrogram in the right (b) corresponds with a aggressive vocalisation. Characterisation of vocalisations used in different contexts and by different individuals Overall, A. percula used a range of different vocalisations associated with different behaviours and social rank. Vocalisations could be differentiated through frequency characteristics (PC1, 37.2%) and investment characteristics (PC2, 22.2%) (Fig.2). PC1 decreases with sound frequency (peak frequency, dominant frequency, and mean frequency) and PC2 increases with sound duration and entropy. Pause duration decreases the third component while pulse number strongly increases the fourth component. Figure 2. Plot showing the relationship between the response variables (arrows) across the first two principal components (PC1 and PC2). The arrows represent the loadings of each response variable onto the corresponding principal components, with the length and direction of each arrow indicating the strength and direction of the variable’s contribution to the axes. The colour represents the percentage of contribution of each variable. Individuals differed in the frequency characteristics of their vocalisations depending on their social rank. The dominant male, rank 2, had significantly lower PC1 scores (estimate (± standard deviation): -0.377 ± 0.431) than the dominant female, rank 1 which also had low PC1 scores (estimate: -0.417 ± 0.214). In contrast, rank 3 and rank 4 showed a significant increase in PC1 scores than the dominant ranks (rank 3: 0.229 ± 0.284; rank 4: 1.927 ± 0.575). The fixed effects explained 5.9% of the variance (R 2 c = 0.764, R 2 m = 0.059). Vocal characteristics, like sound duration and entropy, varied based on both the individual’s rank and the behaviour. There was a significant interaction between rank and behaviour on PC2 (χ 2 9 = 27.83, p < 0.001) indicating that the characteristics of vocal investment, such as sound duration and entropy, varied depending on both an individual’s rank and the behaviour being performed. The fixed effects explained 13.4% of the variance (R 2 c = 0.597, R 2 m = 0.134). Effect of rank on sound frequency Frequencies of vocalisations varied depending on the social rank of individuals in clown anemonefish groups. Dominant individuals had different dominant and mean frequencies compared to subordinates, but no difference was found in peak frequency (Fig.3). There was a significant difference between rank in the dominant frequency (χ 2 3 = 16.94, p < 0.001, R 2 c = 0.693, R 2 m = 0.063) and mean frequency (χ 2 3 = 11.96, p value < 0.001, R 2 c = 0.544, R 2 m = 0.139), but not peak frequency (χ 2 3 = 7.35, p = 0.061, R 2 c = 0.418, R 2 m = 0.077). Rank 2 had higher dominant and mean frequency than rank 1 (dominant frequency estimate: 0.039 ± 0.022; mean frequency estimate: 0.083 ± 0.046). In contrast, both rank 3 and 4 had lower dominant and mean frequencies than rank 1 (Rank 3: dominant frequency estimate: –0.017± 0.029, mean frequency estimate: -0.093 ± 0.062; Rank 4: dominant frequency estimate: –0.200 ± 0.058, mean frequency estimate: -0.293 ± 0.138) When comparing ranks, rank 4 consistently differed from the other ranks, particularly in dominant and mean frequency, while rank 2 showed notable differences from rank 3 and rank 4 for mean frequency. Rank 2 (males) used higher mean frequencies in their vocalisations than rank 3 (estimate = 0.176, p < 0.05), and rank 4 subordinates (estimate = 0.375, p < 0.05), with no significant differences observed between rank 1 and other ranks. Rank 4 used higher dominant and peak frequencies than rank 1 (estimate = 0.200, p < 0.01; estimate = 0.328, p = 0.159, respectively), and significant higher dominant frequencies than rank 2 (estimate = 0.238, p < 0.001), and rank 3 (estimate = 0.182, p 0.05). Figure 3. Pairwise comparisons of frequency measures across rank categories. The plot shows the estimated contrast values with error bars representing standard errors. Each point represents a contrast between two ranks, with significant differences highlighted in red and an asterisk, and non-significant contrasts shown in black. The x-axis represents the rank contrasts, and the y-axis shows the estimated contrast value (on the log scale). The plot uses Tukey’s method for p-value adjustment. Effects of standard length and behaviour on vocalisations Body size and behaviour influence the acoustic characteristics of vocalisations in clown anemonefish. The entropy, pulse number and pulse duration of the vocalisations differ between body sizes and behaviours. There were significant effects of the interaction between standard length (SL) and behaviour on the entropy (χ 2 5 = 21.42, p < 0.001, R 2 c = 0.475, R 2 m = 0.106), number of pulses (χ 2 5 = 16.10, p < 0.01, R 2 c = 0.647, R 2 m = 0.141) and pulse duration (χ 2 5 = 13.67, p < 0.05, R 2 c = 0.700, R 2 m = 0.056). The entropy of the pulse significantly increased with standard length (χ 2 1 = 5.67, p < 0.05), however no significant effect of behaviour or standard length were found for the other response variables. The pulse number of vocalisations differ across behaviours (See Fig. 4). Submissive vocalisations were significantly different from aggressive (estimate = -1.899, p < 0.01), competitive (estimate = -1.094, p < 0.05), territorial (estimate = 1.240, p < 0.01) and neutral vocalisations (estimate = 1.240, p < 0.01). However, no significant differences were found between vocalisations associated with other behaviours and pairwise comparison showed no significant differences between behavioural categories for pulse duration and entropy. These results suggest that acoustic communication associated with submission differs significantly from other behaviours. Figure 4. Pairwise comparisons of the effects of different behaviours on pulse number. The plot displays the estimates (log scale) for each pairwise contrast, with error bars representing the standard error (SE). Red points indicate statistically significant differences (p < 0.05), while black points represent non-significant differences. All contrasts pairs are contained in three categories: Intragroup (pairs of behaviours observed between individuals of the same group), Interspecific (pairs of behaviours observed from members of the group towards other species) and Intragroup vs Interspecific (pairs that compare the vocalisations of intragroup and interspecific behaviours). Anthropogenic noise The frequency range of anemonefish vocalisation (0.200-1.080 kHz) was found to overlap with the bandwidth of motorboat noise, which was 0.050-1.100 kHz (Fig. 5). In the recording in which motorboat was present, anemonefish vocalisations were completely undetectable until the noise disturbance stopped. This shows the potential for masking of anemonefish acoustic communication and disturbance of their group dynamics. Figure 5. Spectrogram of motorboat noise recorded underwater by Audiomoths placed next to anemonefish groups. The y-axis represents frequency (in kHz), while the x-axis represents time (in seconds). The spectrogram shows the temporal variation in frequency of the motorboat noise, with higher intensity indicated by darker shading. The recorded motorboat noise overlaps with the frequency range of the anemonefish vocalisations, highlighting the potential for acoustic masking that could disrupt fish communication. Discussion Our results highlight the diversity of sounds produced by A. percula and the critical role of acoustic signals in their communication, both within and outside their social groups. We demonstrated that specific behaviours are associated with distinct vocalisations, that vocalisations differ depending on individuals social rank and that they play a major role in interspecific interactions and territorial defence. These findings help bridge the knowledge gap in fish acoustic communication and reinforce the idea that acoustic signalling plays a pivotal role in establishing and maintaining social dynamics in fish communities. Our observations showcase that clown anemonefish, in their natural habitat, use vocalisations frequently and in a wide range of social contexts. Vocalisations, alongside visual displays, were used by group members to signal to each other, not only during agonistic encounters, but also during neutral/social interactions. This is congruent with previous observations (Parmentier & Lecchini, 2022), however, most studies have focused on agonistic sounds, and few have mentioned the presence of social sounds. The difference may lie in the definition for certain behaviours, as we classified vocalisations produced during neutral interactions, such as fish meeting up and swimming in the same direction within a body length of each other, as a separate category (28,29). We have also observed anemonefish produce vocalisations when interacting with other species and producing vocalisations on the edge of the anemone without directly interacting with other fish, something that has not been described previously in anemonefish acoustic studies. The laboratory settings used in previous studies greatly limit the social interactions observed. Social and ecological context is an important factor to consider to understand the full scope of social behaviour and acoustic communication (Campbell et al. 2009). Social rank influences frequency parameters of A. percula vocalisations. Interestingly, rank 2 had higher frequencies than rank 1 while the subordinates (rank 3 and 4) had lower frequencies. Social rank has been shown to be an important predictor of individuals behaviour in anemonefishes, with higher ranks displaying higher rates of aggression and cooperation (28,29). Contrary to our prediction and previous studies that suggested a relationship between body size and vocalisation frequency decreases (41) we didn’t find a significant effect of standard length on the frequency parameters of the vocalisations. We believe this discrepancy may be due to the specific methods used in our study, since larger and more dominant individuals tend to be more vocal, and their vocalisations, including jaw movements, are easier to detect compared to those of smaller individuals. Consequently, our data may be biased towards larger individuals, preventing us from finding a strong connection to body size. We suggest that the effect of rank on vocalisations may provide valuable information about an individual’s social status to other members of the group, serving as a mechanism to monitor individuals within the hierarchy and assess potential competitors. However, further studies need to be undertaken. These findings underscore that social status may shape acoustic communication in A. percula , highlighting the complexity of their behavioural ecology. Anemonefish use sounds to accompany several types of behaviour, and some sound characteristics differed with type of behaviour and body size. We found that acoustic characteristics of A. percula sounds such as entropy, duration of the pulse and number of pulses were influence by the interaction between behaviour and size, and the number of pulses varied between behavioural categories. These results suggest that, although vocalisations are influence by body size due to physiological constrains, A. percula can create a small degree of variation within their vocal repertoire to produce different acoustic signals that, accompanied by visual displays, are use in their social communication. Specifically, we found that sounds associated with submissive behaviours were very different to the sounds associated with behaviours such as intragroup aggression, territorial displays, and competition with heterospecifics (which all fall under the umbrella of aggressive/threatening behaviours). This is consistent with the classification of anemonefish sounds made by Colleye and Parmentier (2012). However, we suggest the inclusion of a third category of vocalisations that are produce without a direct interaction with another individual (conspecific or not), as we believe the function and context of these vocalisations is different from those related to aggressive/submissive sounds. As a social animal, communication between individuals is a key element to maintain the hierarchy and cohesion of the group (58). Acoustic signals may help reduce or avoid conflicts from escalating and be used to maintain their social hierarchy (13,43). Future studies should use multispecies approach as anemonefish species have been shown to differ in cooperative behaviour rates, aggressiveness, dispersion, and dependence on their host anemone (28,29,59). The frequency range of anemonefish vocalisations overlaps with the noise produced by motorboats, which might suggest that anemonefish is vulnerable to masking. Masking can disrupt communication between individuals by either preventing the auditory system from detecting sounds (energetic masking) or by interfering with cognitive processing (informational masking) (Branstetter et al., 2016; Rosa and Koper, 2018). Our findings highlight that anemonefish rely heavily on vocalisations when interacting with their environment and conspecifics, making them particularly susceptible to the adverse effects of sound masking. Impacts of masking have been observed in the Lusitanian toadfish ( Halobatrachus didactylus ), where ship noise reduced auditory sensitivity and impaired the detection of conspecific vocalisations (60). The widespread use of motorboats and jet skis in touristic coastal areas, including shallow waters near coral reefs, can significantly affect coral reef fish behaviour (61). For anemonefish, exposure to motorboat noise not only increases cortisol and androgen levels but also elevates aggression and hiding behaviours (62). Further research on the masking effects of anthropogenic noise on anemonefish vocalisations needs to be conducted to understand its impacts and inform mitigation strategies. In conclusion, our findings highlight the complexity and importance of acoustic communication in A. percula , demonstrating how vocalisations are shaped by both physiological constraints, such as body size, and social factors, such as rank and behaviour. These vocalisations likely play a critical role in maintaining social hierarchy and cohesion within groups, and their diversity underscores the sophistication of fish communication systems. Also, the overlap between the frequency range of A. percula vocalisations and motorboat noise raises concerns about the potential impacts of acoustic masking on these critical behaviours. Advances in acoustic monitoring technologies and analytical tools, such as machine learning, offer exciting opportunities to deepen our understanding of these mechanisms and assess the effects of noise pollution. Future research could explore interspecies differences in vocalisations, providing insights into the evolution of acoustic communication in anemonefish and other taxa. Additionally, understanding how A. percula and similar species respond to environmental changes, including noise pollution, could inform conservation strategies to protect the social and ecological stability of fish populations, particularly in noisy, human-impacted habitats such as coral reefs. Ethics Ethical clearance was given by Newcastle University Ethics committee. Data Accessibility Statement The data set used in this study is available at Mendeley Data, V1, https://data.mendeley.com/datasets/7m9v6hxzrr/1, DOI: 10.17632/7m9v6hxzrr.1 Author contribution statement Lucia Yllan: conceptualization, data curation, investigation, formal analysis, investigation, methodology, writing – original draft, review & editing; Theresa Rueger: supervision, funding acquisition, writing - review & editing. Competing Interests Statement We declare we have no competing interests. Acknowledgements We thank all the people that assisted with data collection, the Mahonia Na Dari Research and Conservation Centre (Kimbe Bay, Papua New Guinea) for supporting the fieldwork for this study and the Tamare-Kilu communities which granted us access to work in their local coral reefs. This study was funded by the School of Natural and Environmental Sciences (Start-up fund) of Newcastle University and the Leverhulme Trust (grant no. RPG-2023-186). References 1. Benoit-Bird KJ, Au WWL. Phonation behavior of cooperatively foraging spinner dolphins. J Acoust Soc Am. 2009;125(1):539–46. 2. Zuberbühler K. Chapter 8 Survivor Signals. The Biology and Psychology of Animal Alarm Calling. Adv Study Behav. 2009;40(09):277–322. 3. Ibara RM, Penny LT, Ebeling AW, van Dykhuizen G, Cailliet G. The mating call of the plainfin midshipman fish, Porichthys notatus. 1983;(May):205–12. 4. Bar-On YM, Phillips R, Milo R. The biomass distribution on Earth. Proc Natl Acad Sci U S A. 2018;115(25):6506–11. 5. Schwarz AL. The behavior of fishes in their acoustic environment *. Environ Biol Fishes. 1985;13(1):3–15. 6. Leis JM, Siebeck U, Dixson DL. How nemo finds home: The neuroecology of dispersal and of population connectivity in larvae of marine fishes. Integr Comp Biol. 2011;51(5):826–43. 7. Simpson SD, Yan HY, Wittenrich ML, Meekan MG. Response of embryonic coral reef fishes ( Pomacentridae : Amphiprion spp .) to noise. 2005;287:201–8. 8. Radford CA, Stanley JA, Simpson SD, Jeffs AG. Juvenile coral reef fish use sound to locate habitats. 2011;295–305. 9. Simpson SD, Radford AN, Tickle EJ, Meekan MG, Jeffs AG. Adaptive Avoidance of Reef Noise. PLoS One. 2011;6(2):2–6. 10. Parmentier E, Berten L, Rigo P, Aubrun F, Nedelec SL. The influence of various reef sounds on coral-fish larvae. J Fish Biol. 2015;86:1507–18. 11. Putland RL, Montgomery JC, Radford CA. Ecology of fish hearing. J Fish Biol. 2019;95(1):39–52. 12. McKibben JR, Bass AH. Peripheral encoding of behaviorally relevant acoustic signals in a vocal fish: Harmonic and beat stimuli. J Comp Physiol A. 2001;187(4):271–85. 13. Colleye O, Parmentier E. Overview on the Diversity of Sounds Produced by Clownfishes (Pomacentridae): Importance of Acoustic Signals in Their Peculiar Way of Life. PLoS One. 2012;7(11). 14. McKenna MF, Shannon G, Fristrup K. Characterizing anthropogenic noise to improve understanding and management of impacts to wildlife. Endanger Species Res. 2016;31(1):279–91. 15. Popper AN, Hawkins AD. An overview of fish bioacoustics and the impacts of anthropogenic sounds on fishes. Vol. 94, Journal of Fish Biology. Blackwell Publishing Ltd; 2019. p. 692–713. 16. Holles SH, Simpson SD, Radford AN, Berten L, Lecchini D. Boat noise disrupts orientation behaviour n a coral reef fish. Mar Ecol Prog Ser. 2013 Jun 27;485:295–300. 17. Simpson SD, Purser J, Radford AN. Anthropogenic noise compromises antipredator behaviour in European eels. Glob Chang Biol. 2015 Feb 1;21(2):586–93. 18. Simpson SD, Radford AN, Nedelec SL, Ferrari MCO, Chivers DP, McCormick MI, et al. Anthropogenic noise increases fish mortality by predation. Nat Commun. 2016 Feb 5;7. 19. Voellmy IK, Purser J, Flynn D, Kennedy P, Simpson SD, Radford AN. Acoustic noise reduces foraging success in two sympatric fish species via different mechanisms. Anim Behav. 2014 Mar;89:191–8. 20. Branstetter BK, Bakhtiari K, Black A, Trickey JS, Finneran JJ, Aihara H. Energetic and informational masking of complex sounds by a bottlenose dolphin ( Tursiops truncatus ) . J Acoust Soc Am. 2016;140(3):1904–17. 21. Rosa P, Koper N. Integrating multiple disciplines to understand effects of anthropogenic noise on animal communication: Ecosphere. 2018;9(2). 22. Duncan AJ, Lucke K, Erbe C, McCauley RD. Issues associated with sound exposure experiments in tanks. In: Proceedings of Meetings on Acoustics. Acoustical Society of America; 2016. 23. Banse M, Lecchini D, Bertucci F, Parmentier E. Reliable characterization of sound features in fishes begins in open-water environments. J Acoust Soc Am. 2023 Jul 1;154(1):270–8. 24. Campbell DLM, Weiner SA, Starks PT, Hauber ME. Context and control: behavioural ecology experiments in the laboratory. Vol. 46, Methods in Ecological Research. 2009. 25. Turko AJ, Firth BL, Craig PM, Eliason EJ, Raby GD, Borowiec BG. Physiological differences between wild and captive animals: a century-old dilemma. J Exp Biol. 2023 Dec 1;226(23). 26. Archard GA, Braithwaite VA. The importance of wild populations in studies of animal temperament. Vol. 281, Journal of Zoology. 2010. p. 149–60. 27. Calisi RM, Bentley GE. Lab and field experiments: Are they the same animal? Vol. 56, Hormones and Behavior. 2009. p. 1–10. 28. Yllan L, Heatwole S, Wong M, Rueger T. Effect of social context on behaviour in anemonefish hierarchies. Anim Behav. 2024 Mar 1;209:83–93. 29. Rueger T, Heatwole SJ, Wong MYL. Social rank and species-specific differences influence cooperation and conflict in anemonefish societies. Anim Behav. 2022 May 1;187:105–16. 30. Gail Fautin D. The Anemonefish Symbiosis: What is Known and What is Not. Vol. 10, Symbiosis. 1991. 31. Moyer JT, Nakazono A. Protandrous Hermaphroditism in Six Species of the Anemonefish Genus Amphiprion in Japan. Japanese Journal f Ichthyology. 1978;25(2). 32. Buston PM, Cant MA. A new perspective on size hierarchies in nature: Patterns, causes, and consequences. Oecologia. 2006;149(2):362–72. 33. Buston P, Hall SGM. Forcible eviction and prevention of recruitment in the clown anemonefish. Behavioral Ecology. 2003;14(4):576–82. 34. Wong MYL, Buston PM, Munday PL, Jones GP. The threat of punishment enforces peaceful cooperation and stabilizes queues in a coral-reef fish. Proceedings of the Royal Society B: Biological Sciences. 2007;274(1613):1093–9. 35. Buston PM, Bogdanowicz S, Wong A, Harrison RG. Are clownfish groups composed of close relatives ? An analysis of microsatellite DNA variation in Amphiprion percula. Mol Ecol. 2007;16:3671–8. 36. Fricke H, Fricke S. Monogamy and sex change by aggressive dominance in coral reef fish. Nature. 1977;266(5605):830–2. 37. Moyer JT, Nakazono A, Species S. of the Anemonefish Amphiprion in Japan. Japanese Journal of Ichthyology. 1978;25(2):101–6. 38. Rueger T, Branconi R, Froehlich CYM, Heatwole SJ, Wong MYL, Buston PM. The Next Frontier in Understanding the Evolution of Coral Reef Fish Societies. Front Mar Sci. 2021;8(May):665789. 39. Parmentier E, Colleye O, Mann D. Hearing ability in three clownfish species. Journal of Experimental Biology. 2009;212(13):2023–6. 40. Parmentier E, Colleye O, Michael L, Vandewalle P, Herrel A. Sound Production in the Clownfish. 2012;316(May 2007):2007. 41. Colleye O, Frederich B, Vandewalle P, Casadevall M, Parmentier E. Agonistic sounds in the skunk clownfish Amphiprion akallopisos: Size-related variation in acoustic features. J Fish Biol. 2009;75(4):908–16. 42. Colleye O, Nakamura M, Frédérich B, Parmentier E. Further insight into the sound-producing mechanism of clownfishes: What structure is involved in sound radiation. Journal of Experimental Biology. 2012;215(13):2192–202. 43. Parmentier E, Lecchini D. Sound Communication. In: Evolution, Development and Ecology of Anemonefishes. CRC Press; 2022. p. 95–102. 44. de Jong K, Amorim MCP, Fonseca PJ, Heubel KU. Noise Affects Multimodal Communication During Courtship in a Marine Fish. Front Ecol Evol. 2018 Jul 31;6. 45. Nedelec SL, Radford AN, Gatenby P, Davidson IK, Velasquez Jimenez L, Travis M, et al. Limiting motorboat noise on coral reefs boosts fish reproductive success. Nat Commun. 2022 Dec 1;13(1). 46. Amorim, M. C. P. Vasconcelos RO, Fonseca PJ. Fish sounds and mate choice. Sound communication in fishes. How Animals Communicate. 2015. 1–33 p. 47. Friard O, Gamba M. BORIS: a free, versatile open-source event-logging software for video/audio coding and live observations. Methods Ecol Evol. 2016 Nov 1;7(11):1325–30. 48. Laudet V, Ravasi T, editors. Evolution, Development and Ecology of Anemonefishes: Model Organisms for Marine Science (1st ed.). CRC Press; 2022. 49. RStudio Team. RStudio: Integrated development for R. RStudio, PBC; 2023. 50. J. Sueur, T. Aubin, C. Simonis. Seewave: a free modular tool for sound analysis and synthesis. Vol. 18. Bioacoustics; 2008. p. 213–26. 51. Araya-Salas M. Rraven: connecting R and Raven bioacoustic software. R package version 1.9; 2020. 52. Araya-Salas M SVG. warbleR: an r package to streamline analysis of animal acoustic signals. Methods in Ecology and Evolution; 2017. 53. Alboukadel Kassambara and Fabian Mundt. factoextra: Extract and Visualize the Results of Multivariate Data Analyses. . 2020. 54. Bates D, Mächler M, Bolker B, Walker S. Fitting Linear Mixed-Effects Models using lme4. 2014 Jun 23; Available from: http://arxiv.org/abs/1406.5823 55. Sakamoto Y., Ishiguro M, Kitagawa G. Akaike information criterion statistics. Dordrecht, The Netherlands: D Reidel. 1986;81(10.5555):26853. 56. Lenth R. emmeans: Estimated Marginal Means, aka Least-Squares Means. 2024. 57. Nakagawa S, Johnson PCD, Schielzeth H. Appendices S1-S5 and S7 for “The coefficient of determination R 2 and intra-class correlation ICC from generalized linear-mixed effects models revisited and expanded” Appendix S1 Deriving the observation-level variance σ 2 ε using log-normal approximation for the negative binomial distribution with log link. 2017. 58. Margoliash D, Hale ME. Neuroscience: Vertebrate vocalizations. Vol. 321, Science. 2008. p. 347–8. 59. Cleveland A, Verde EA, Lee RW. Nutritional exchange in a tropical tripartite symbiosis: Direct evidence for the transfer of nutrients from anemonefish to host anemone and zooxanthellae. Mar Biol. 2011 Mar 1;158(3):589–602. 60. Vasconcelos RO, Amorim MCP, Ladich F. Effects of ship noise on the detectability of communication signals in the Lusitanian toadfish. Journal of Experimental Biology. 2007 Jun;210(12):2104–12. 61. McCormick MI, Allan BJM, Harding H, Simpson SD. Boat noise impacts risk assessment in a coral reef fish but effects depend on engine type. Sci Rep. 2018 Dec 1;8(1). 62. Mills SC, O’Donnell JL, Bernardi G, Beldade R. Natural endocrine profiles of the group-living skunk anemonefish Amphiprion akallopisos in relation to their size-based dominance hierarchy. J Fish Biol. 2018 Mar 1;92(3):773–89. Information & Authors Information Version history V1 Version 1 30 May 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Collection Ecology and Evolution Keywords behavioral ecology description marine statistical vertebrate Authors Affiliations Lucia Yllan 0000-0002-5992-1133 [email protected] Newcastle University - Newcastle Campus View all articles by this author Theresa Rueger 0000-0002-2105-6412 Newcastle University View all articles by this author Metrics & Citations Metrics Article Usage 334 views 321 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Lucia Yllan, Theresa Rueger. What does the anemonefish say?: decoding Amphiprion percula’s acoustic communication. Authorea . 30 May 2025. DOI: https://doi.org/10.22541/au.174861208.89315690/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.174861208.89315690/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'a00846f879094193',t:'MTc3OTU4MzUyMg=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();