Vocalization during agonistic encounter in Mongolian gerbils: Impact of sexual experience.

Behaviors and vocalizations associated with aggression are essential for animals to survive, reproduce, and organize social hierarchy. Mongolian gerbils (Meriones unguiculatus) are highly aggressive and frequently emit calls. We took advantage of these features to study the relationship between vocalizations and aggressive behaviors in virgin and sexually experienced male and female Mongolian gerbils through the same-sex resident-intruder test. Both sexes of resident gerbils exhibited aggressive responses toward intruders. Multiparous females exhibited the most aggressive responses among the four groups. We also confirmed two groups of vocalizations during the encounters: high-frequency (>24.6 kHz) and low-frequency (<24.6 kHz). At the timing of high-frequency vocalizations observed during the tests, the vast majority (96.2%) of the behavioral interactions were non-agonistic. While, at the timing of low-frequency vocalizations observed during the tests, around half (45%) of the behavioral interactions were agonistic. Low-frequency vocalizations were observed mainly during encounters in which multiparous females were involved. These results suggest that high- and low-frequency vocalizations relate to non-agonistic and agonistic interactions, respectively. In addition to affecting aggressive behavior, sexual experience also affects vocalization during encounters. These findings provide new insights into the modulatory effects of sex and sexual experience on vocalizations during agonistic encounters.

Introduction

In most species, aggression is an adaptive response to increase survival and reproduction rates. Territorial aggression is one of the most common aggressive behaviors displayed by rodents [1-3], and it consists of multiple complex behaviors, such as biting, boxing, and chasing. In addition, unique vocalizations associated with territorial aggression have been reported in males of many rodents [4]. As vocal communication is important for the survival and formation of a colony, rodents emit vocalizations in various behavioral contexts, including aggressive encounters [5,6]. Furthermore, past experiences have also been found to modulate vocalizations in rodents [7-9]. However, it is not well known how behavioral contexts and past experiences modulate vocalizations during aggressive encounters.

The Mongolian gerbil (Meriones unguiculatus) is a highly social and territorial rodent species [10-12]. These animals exhibit elevated territorial aggressiveness by which they attack, or in some cases even kill, unfamiliar conspecifics [13]. As it is well known, laboratory rodents such as mice or rats rarely exhibit this extent of aggression to kill the other conspecifics [3,6]. Aggressiveness in Mongolian gerbils has been observed during same-sex and opposite-sex interactions [14]. Furthermore, sexually naïve males and even females exhibit aggressive behavior toward unfamiliar same-sex conspecifics [15,16], which demonstrates that aggressiveness toward unfamiliar conspecifics is a prominent feature of both sexes of Mongolian gerbils. In addition, sexually experienced gerbils are more aggressive than virgin subjects [13,17].

Although Mongolian gerbils are aggressive toward unfamiliar conspecifics, they communicate with each other, even with unfamiliar conspecifics, frequently through vocalization. The vocalizations of Mongolian gerbils have been confirmed and investigated in various behavioral contexts [17-19], including encounters with conspecifics [20], mother-infant interactions [21-23], and mating [24]. Moreover, the call rates of these vocalizations in Mongolian gerbils are sexually dimorphic [24,25].

Given that both sexes are aggressive and emit calls frequently, Mongolian gerbils are one of the best laboratory animals to investigate vocalizations during agonistic encounters. Although several studies cited in the previous paragraph have reported aggression-specific vocalizations, it remains unclear how sex and sexual experience affect the call rate of various vocalizations during agonistic encounters in Mongolian gerbils. Thus, we analyzed vocalizations between four different pair-tests (virgin male–virgin male; experienced male–virgin male; virgin female–virgin female; multiparous female–virgin female) of Mongolian gerbils through the same-sex resident-intruder test, to clarify the difference of vocalizations and relation between vocalizations and agonistic interactions.

Materials and methods

All experiments were conducted in accordance with the guidelines and protocols approved by the Animal Care Committee of Kanazawa Medical University (2019–22, 2022–6) and the Animal Experimental Committee of Doshisha University (A21052-1).

Subjects

Fifty-two Mongolian gerbils (24 males and 28 females, aged 3–6 months; purchased from Sankyo Lab Service, Tokyo, Japan) were used. Five males were sexually experienced, and seven females were multiparous. These gerbils were mated at our facility. The other gerbils were virgins. All gerbils were bred and maintained in the laboratory at 22–23°C, and approximately 50% humidity. Two to four animals were housed together in plexiglass cages (20 cm × 40 cm × 17 cm; white paper beddings) under a 12 h light/dark cycle. Food and water were available ad libitum.

Same-sex resident-intruder test

We conducted 26 resident-intruder tests to assess the extent of territorial aggressiveness in the gerbils. Each subject was singly housed in a plexiglass cage (20 cm × 40 cm × 17 cm; resident cage) for at least one week before undergoing the resident-intruder test. The residents were divided into four groups: virgin females (VF; n = 7), multiparous females (MF; n = 7), virgin males (VM; n = 7), and experienced males (EM; n = 5). All tests were performed during the light cycle of the day. At the test, each resident’s home cage was placed in a sound-attenuated room one by one. Three minutes later, an intruder was introduced into the resident’s cage. All intruders were virgin, of the same sex as the resident and unfamiliar to the resident (12 VMs and 14 VFs). Each subject (resident and intruder) underwent the resident-intruder test for a single time. Most of the residents (24/26) expressed aggressive behaviors toward the intruder within several minutes. Aggressive behaviors were initiated by residents in the vast majority (20/26) of tests, thus basically we focused on the behaviors of residents in this study. Behavioral interactions between the resident and intruder were recorded using a CMOS camera (D435, Intel Corporation, Santa Clara, CA, USA) placed in front of the cage. For behavioral analysis, we converted the frame rate of the recorded files to 10/s, and visually identified the types of behaviors of the resident and intruder by following the classification described in a previous report [17] with a small modification. The behaviors were classified as follows by observers manually: alert posture, ano-genital sniff, approach, attack, boxing, chase, dig, explore, fight, flee, jump, move away, nasal sniff, push, self-groom, stop moving, and watch. Details of each behavior are shown in Table 1. ‘attack’, ‘boxing’, ‘chase’, ‘fight’, ‘flee’, and ‘push’ were scored as aggressive/agonistic behaviors. The rest was scored as non-agonistic interaction. The latency of aggressive behavior (start time of the first aggressive behavior) and the total duration of aggressive behavior were recorded. The total duration of aggressive behavior per minute was defined as ‘normalized duration of aggressive behavior’. The intruder was removed from the resident cage after 10 min of behavioral testing. If no aggressive behavior was observed within 10 min, the latency of aggressive behavior was regarded as 600 s, and the duration of aggressive behavior was scored as 0 s for the analysis. The tests were halted immediately when aggressive behaviors escalated to physical injuries. In this study, 3 tests (three female pairs) were halted at 3–5 min. These cases were included in this study, by using each index for behaviors and vocalizations normalized by time (duration or number of calls per minute).

Table 1

Descriptions of behaviors that were scored.

Behaviors	Description
Attack	One animal biting the another
Fight	Two animals gripping each other’s flanks, biting, and rolling over
Flee	Running away from the another
Boxing	Two animals facing each other with physical contact
Chase	One animal running after another more than one body length
Push	Using paws or body to make another move away
Jump	Pushing itself off the ground with the hind legs in a vertical movement
Alert	Standing on the hind legs alone and watching the surroundings
Watch	Both animals keeping motionless, face-to-face at a close distance
Move away	Making a distance from the another
Dig	Removing the beddings with its front paws moving quickly back and forth
Grooming	Licking, nibbling, and scratching of one’s own body
Explore	Actively moving around the cage like investigating
Stop	Staying still
Approach	Moving nearer to another within one body length
Nasal sniff	Nose-to-nose contact
Ano-genital sniff	Nose contact to the ano-genital region of another animal

Descriptions of each behavior, following the scoring in the previous reports [17,26] with small modifications.

Recording and analysis of vocalizations

A recording microphone (Ultrasound Microphone CMPA-P48/CM16 SN34, Avisoft Bioacoustics, Berlin, Germany) was placed 20 cm above the top of the cage, and vocalizations were recorded using a sound card (UltraSoundGate, 116Hb, Avisoft Bioacoustics, Berlin, Germany) at a sample rate of 125 kHz at 16 bits/sample. To identify calls, Adobe Audition software (Adobe, CA, US) was used. Each call was detected and cut out from the audio files manually by experimenters. All recorded calls with a signal-to-noise ratio higher than 10 dB were used for offline analysis and classified into different syllable types. This analysis was based on a 512-point fast Fourier transform (Hamming window) with an 80% overlap. Spectrograms were calculated using a MATLAB custom program at a frequency resolution of 244 Hz and a temporal resolution of 0.82 ms. Call envelopes were obtained from the original waveforms. The spectro-temporal parameters of vocalizations were quantified using the following parameters: total duration, frequency of fundamental frequency (F0) at which the call began, frequency of F0 at which the call ended, maximum frequency attained by F0, minimum frequency attained by F0, maximum frequency location (percentage of total duration), minimum frequency location (percentage of total duration). Frequency modulation (FM) sounds were defined if the maximum frequency was >20% of the minimum frequency. Noise burst (NB) sounds were defined if the sounds showed no clear F0 and no clear harmonic structure in spectral components (the “NB” does not indicate Gaussian noise in the sound signal). For classifying call types, we used criteria established by Kobayasi [18]. Briefly, calls were classified manually as being either simple syllables or composites. A simple syllable consisted of a single predominant sound element, such as an FM segment, a constant frequency (CF) segment, or an NB segment. We used a prefix to define secondary features of a call’s spectrogram, e.g., AFM for arched FM and UFM for upward FM. We also used a postfix to define the duration of syllables: s = duration <75 ms and l >300 ms. A composite syllable consisted of two or more types of distinct components each representing a simple CF, NB, or FM segment combined without an interval; for example, Quasi CF-NB (QCF-NB) would indicate a QCF segment combined with an NB segment. Adding to these, we defined ‘U-shape call’, which is similar to UFMs but starts with downward FM. Thus, we identified nine call types in this paper (Fig 1); U-shape, UFM-s, UFM, AFM, DFM-l, DFM, QCF, NB, and QCF-NB.

Fig 1

Representative spectrograms of nine call types of vocalizations.

U shape: syllables with downsweep followed by upsweep frequency change

UFM-s: syllables with upsweep frequency change and short duration (<75 ms)

UFM: syllables with upsweep frequency change

AFM: syllables with upsweep followed by downsweep frequency change

DFM-l: syllables with downsweep frequency change and long duration (>300 ms)

DFM: syllables with downsweep frequency change

QCF: syllables with the maximum frequency were <20% of the minimum frequency.

NB: syllables with no clear F0 and no clear harmonic structure in spectral components

QCF-NB: syllables with QCF followed by NB

After calls were recorded, we categorized calls into two groups, the high-frequency and the low-frequency vocalizations. The maximum fundamental frequency was used to categorize. We calculated the fitting functions of two normal distributions for high- and low- vocalization distributions. The intersection of the two functions is used as the objective frequency border between the high-frequency and the low-frequency vocalizations. The timing of behaviors and vocalization was synchronized with the sound in the video and the audio file.

Statistical analysis

The Chi-square test (χ2) was used to compare the ratio of behaviors at the timing of vocalizations. To compare the medians of latency of aggressive behavior and normalized duration of aggressive behavior, we applied the Mann-Whitney U test using SPSS (IBM, Armonk, NY, US). Spearman’s rank-order correlation coefficient was calculated using SPSS for each measurement combination: latency of aggressive behavior and normalized duration of aggressive behavior; latency of aggressive behavior and the number of high-frequency vocalizations per minute (the number of calls divided by the whole test duration (min)); latency of aggressive behavior and the number of low-frequency vocalizations per minute (the number of calls divided by the whole test duration (min)); normalized duration of aggressive behavior and the number of high-frequency vocalizations per minute; and normalized duration of aggressive behavior and the number of low-frequency vocalizations per minute. The Kruskal-Wallis test, followed by Dunn’s test, was used for multiple comparisons. Non-paired t-test was used for comparing the mean of each call parameter before and after the first agonistic behavior in the test. Statistical significance was determined using a threshold of p < 0.05. Ranges enclosed within [] indicate the interquartile ranges. The standard deviations are indicated as ‘± numbers’.

Results

We tested the effect of sexual experience on aggressive behavior in the gerbils during the same-sex resident-intruder tests. 71.4% percent (5 of 7 pairs) of VM residents and 100% (5 of 5 pairs) of EM residents exhibited aggressive behavior. The median latencies of aggressive behavior were not significantly different between these two groups of resident males (Fig 2A male; VM: 127 [interquartile range: 71–388] s, EM: 188 [48-230] s, Mann-Whitney U test: U = 17.0, p = 1). Also, the median normalized durations of aggressive behavior were not significantly different between these two groups of resident males (Fig 2B male; VM: 1.5 [0.3–2.0] s, EM: 1.2 [0.1–3.4] s, U = 21.0, p = 0.639). Female pairs also showed aggressive interactions; 71.4% (5 of 7 pairs; VF resident) and 100% (7 of 7 pairs; MF resident). The median latencies of aggressive behavior were not significantly different between these two groups of resident females (Fig 2A female; VF: 252 [107-556] s, MF: 43 [29-147] s, U = 13.0, p = 0.165), while the median normalized duration of aggressive behavior of MF resident was significantly longer than that of VF residents (Fig 2B female; VF: 0.7 [0.1–3.6] s, MF: 21.4 [15.5–33.1] s, U = 45.0, p = 0.007). These results suggest that females, but not males, become more aggressive after sexual experience.

Fig 2

Latency of aggressive behavior and normalized duration of aggressive behavior of gerbil groups during resident-intruder tests.

(A) latency of aggressive behavior in VM-VM, EM-VM, VF-VF, and MF-VF pairs. Thick horizontal bars represent medians. (B) normalized duration of aggressive behavior in VM-VM, EM-VM, VF-VF, and MF-VF pairs. Thick horizontal bars represent medians. (●, virgin male; ▲, experienced male; ○, virgin female; △, multiparous female). N.S.: Not significant, ** represents p < 0.01.

In addition, we examined the correlation between the latency of aggressive behavior and the duration of aggressive behavior. The duration of aggressive behavior in all groups decreased gradually with an increase in the latency of aggressive behavior. This correlation was statistically significant (Fig 3; Spearman’s correlation coefficient, rs = -0.669, n = 26, p < 0.001). There was no statistically significant correlation within each pair-tests.

Fig 3

Relation between latencies of aggressive behavior and durations of aggressive behavior.

Latencies of aggressive behavior and durations of aggressive behavior were negatively correlated (●, virgin male; ▲, experienced male; ○, virgin female; △, multiparous female; rs = -0.669; p < 0.001).

Vocalizations of gerbils during the same-sex resident-intruder tests

We quantified 2413 vocalizations in male-male interactions and 1949 vocalizations in female-female interactions. We detected nine call types of vocalizations (Figs 1 and 4A; U-shape, UFM-s, UFM, AFM, DFM-l, DFM, QCF, NB, and QCF-NB). Spectro-temporal features of these calls are shown in Table 2. The contour shapes of calls (Fig 1; except the U-shape call, which was newly defined in this study) in all pairs during the resident-intruder test were similar to those reported in previous studies [18]. The percentages of each vocalization among the total calls during the tests are shown in Fig 4B. Higher percentages of DFM/QCF/NB and a lower percentage of UFM-s were observed in MF (resident)—VF (intruder) pairs than in the other pairs. Obviously, as shown in Fig 4A, the nine call types could be categorized into two groups. From the fitting functions of two normal distributions, the separation at 24.6 kHz was determined. Thus, for further analysis, we categorized the quantified nine types of vocalizations into two groups: high-frequency with short duration (U-shape, UFM-s, and UFM; 32.7 ± 2.0 kHz, 25.9 ± 14.7 ms; n = 3329), and low-frequency with long duration (AFM, DFM-l, DFM, QCF, NB, and QCF-NB; 12.1 ± 2.9 kHz, 160.4 ± 73.3 ms; n = 1033).

Fig 4

Vocalizations of gerbils during resident-intruder tests.

(A) Maximum fundamental frequency and duration of each vocalization. High-frequency groups are represented by bluish symbols and low-frequency groups are represented by reddish symbols. (B) Percentages of each call type emitted during the resident-intruder test are shown.

Table 2

Spectro-temporal features of nine call types.

	Initial freq. [kHz]	Terminal freq. [kHz]	Max. freq. [kHz]	Min. freq. [kHz]	Duration [ms]	Max. freq. location [%]	Min. freq. location [%]
U-shapen = 70	30.91 ±2.08	32.80 ±1.97	33.27 ±1.82	26.82 ±4.70	39.30 ±12.23	78.32 ±40.37	54.99 ±21.71
UFM-sn = 3237	28.04 ±1.83	32.17 ±2.29	32.51 ±2.17	26.77 ±5.02	25.03 ±13.20	92.43 ±21.69	24.46 ±27.54
UFMn = 22	26.39 ±3.84	29.90 ±4.85	30.36 ±6.00	21.34 ±9.00	94.82 ±20.67	74.60 ±37.71	57.77 ±36.34
AFMn = 8	16.42 ±3.12	14.17 ±2.44	17.79 ±3.53	12.09 ±5.11	78.03 ±27.65	24.05 ±19.32	97.41 ±5.63
DFM-ln = 44	10.83 ±1.22	7.87 ±1.37	11.62 ±1.24	5.04 ±3.52	337.19 ±30.21	7.68 ±12.91	90.90 ±20.37
DFMn = 700	12.12 ±2.85	7.36 ±2.55	12.80 ±3.10	6.55 ±2.78	149.23 ±58.83	7.58 ±15.38	90.76 ±19.03
QCFn = 157	10.74 ±3.59	9.81 ±3.70	12.04 ±3.98	7.88 ±4.54	126.83 ±76.83	16.95 ±22.78	82.96 ±29.63
NBn = 26	9.70 ±2.23	7.11 ±1.66	11.22 ±2.43	4.79 ±2.16	155.97 ±77.44	15.83 ±22.75	73.09 ±30.42
QCF-NBn = 98	11.66 ±1.54	7.82 ±1.89	12.87 ±1.74	5.41 ±2.54	199.62 ±70.10	9.11 ±13.94	86.42 ±21.22

Spectro-temporal features of nine call types are shown. Frequency of fundamental frequency (F0) at which the call began (Initial freq.), frequency of F0 at which the call ended (Terminal freq.), maximum frequency attained by F0 (Max. freq.), minimum frequency attained by F0 (Min. freq.), total duration (Duration), maximum frequency location (percentage of total duration; Max. freq. location), and minimum frequency location (percentage of total duration; Min. freq. location) are shown. Mean ±S.D. are indicated.

We compared the number of calls per minute during the tests between the four tests. The median numbers of high-frequency calls per minute ranged from: 20.9 [12.5–24.5] in VM-VM, 4.0 [3.5–13.2] in EM-VM, 8.0 [6.1–10.5] in VF-VF, and 0.1 [0.0–3.1] in MF-VF (Fig 5A). The median numbers of low-frequency calls per minute ranged from: 0.0 [0.0–0.1] in VM-VM, 0.0 [0.0–0.0] in EM-VM, 0.1 [0.0–0.2] in VF-VF, and 13.6 [7.4–25.4] in MF-VF (Fig 5B). Vocalizations of MF-VF pairs exhibited a dramatically different distribution compared to those of the other groups. The rate of low-frequency calls was significantly higher in the MF-VF interaction than in the other pairs (Fig 5B; MF-VF vs. VM-VM, p < 0.05; MF-VF vs. EM-VM, p < 0.001; MF-VF vs. VF-VF, p < 0.05; Kruskal-Wallis test followed by Dunn test). For high-frequency vocalizations, VM-VM pairs emitted calls more frequently than MF-VF pairs (Fig 5A; MF-VF vs. VM-VM, p = 0.04; Kruskal-Wallis test followed by Dunn test). These results suggest that sexual dimorphism and sexual experience affected not only behavior but also vocalizations associated with aggressiveness. In addition, there was a negative correlation between the numbers of high- and low-frequency calls (Fig 5C; rs = -0.413, p = 0.036; n = 26). There was no statistically significant correlation within each pair-tests.

Fig 5

Numbers of two call types by during resident-intruder pairs.

(A) Numbers of high-frequency calls per minute by pair-tests. VM-VM pairs emitted high-frequency calls much more than MF-VF pairs. (B) Numbers of low-frequency calls per minute by pair-tests. MF-VF pairs emitted low-frequency calls much more than the other pairs. (C) There is a negative correlation between the numbers of high-frequency calls and that of low-frequency calls (rs = -0.413; p = 0.036). Each symbol represents calls from a single test (●, virgin male; ▲, experienced male; ○, virgin female; △, multiparous female). * represents p < 0.05. *** represents p < 0.001.

Relationship between each vocalization and aggressive interactions during the same-sex resident-intruder tests

First, we created the ethograms for 26 tests (examples shown in Fig 6A) and categorized the behaviors into two categories, non-agonistic or agonistic (indicated by blue or red in Fig 6A). Then, we quantified the type of behavior at the timing of high- or low-frequency calls emitted (Fig 6B). As shown in Fig 6B, high-frequency calls mostly coincided with no conflicts between the resident and intruder (96.2%), while aggressive interactions were associated much more frequently with low-frequency calls than with high-frequency calls (low, 45.0%; high 3.8%; Pearson’s χ2 = 1138.5, p < 0.001). These results suggest that high-frequency calls accompany non-agonistic interactions, whereas low-frequency calls are related to agonistic interactions. Next, we analyzed the correlation between aggressive behaviors and numbers of the two types of vocalizations. Normalized durations of aggressive behavior correlated negatively with the number of high-frequency vocalizations (Fig 7A and 7C; latency, rs = 0.165, p = 0.421; duration, rs = -0.479, p = 0.013; n = 26). In contrast, the latency of aggressive behavior and duration of aggressive behavior correlated negatively and positively, respectively, with the number of low-frequency vocalizations (Fig 7B and 7D; latency, rs = -0.419, p = 0.033; duration, rs = 0.562, p = 0.003; n = 26). There was no statistically significant correlation within each pair-tests. These results also support the idea that high-frequency calls accompany non-agonistic interactions while low-frequency calls are related to agonistic interactions. Finally, we compared five parameters (Initial freq., Terminal freq., Max. freq., Min. freq., and Duration; refer to Table 2) of high- and low-frequency vocalizations before and after the first agonistic behavior in the test. After the first agonis tic event, Terminal freq. and Max. freq. of high-frequency vocalizations shifted to slightly high frequency (Terminal freq., form 32.28 ± 2.13 kHz to 32.63 ± 2.27 kHz, p < 0.001; Max. freq., form 32.43 ± 2.08 kHz to 32.85 ± 2.07 kHz, p < 0.001, n (before) = 1167, n (after) = 2162). Initial freq., Max. freq., and Min. freq. of low-frequency vocalizations also shifted around 1 kHz higher frequency (Initial freq., form 10.35 ± 2.47 kHz to 11.96 ± 2.91 kHz, p = 0.004; Max. freq., form 10.82 ± 3.67 kHz to 12.37 ± 3.22 kHz, p = 0.013; Min. freq., form 6.44 ± 3.32 kHz to 7.9 ± 2.72 kHz, p = 0.006, n (before) = 28, n (after) = 1005). The duration of low-frequency vocalizations shifted significantly longer after the first agonistic event (Duration, form 105.39 ± 59.7 ms to 162.82 ± 75.9 ms, p < 0.001).

Fig 6

Relation between two types of calls and each behavior.

(A) Representative ethograms of each pair-test during resident-intruder tests. Ethograms for VM-VM, EM-VM, VF-VF, and MF-VF pairs were aligned from top to bottom. Bluish colors represent non-agonistic behaviors and reddish colors represent agonistic behaviors. Blue circles and red circles indicate >24.6 kHz calls and 24.6≤ kHz calls respectively. Red arrows indicate the timing when the resident initiated the first aggressive behavior. MF-VF pairs exhibited different patterns of vocalizations and behaviors. (B) The ratio of behaviors at the timing each call emitted. At the timing of high-frequency calls, agonistic interactions were rarely observed, contrary, at the timing of low-frequency calls, agonistic interactions were dominant. *** represents p < 0.001.

Fig 7

Relationship between two types of calls and aggressive behaviors.

(A) Distributions of latency of aggressive behavior and rate of high-frequency vocalizations. (B) Distributions of latency of aggressive behavior and rate of low-frequency vocalizations. (C) Distributions of duration of aggressive behavior and rate of high-frequency vocalizations. (D) Distributions of duration of aggressive behavior and rate of low-frequency vocalizations. Each symbol represents vocalizations from a single test (●, virgin male; ▲, experienced male; ○, virgin female; △, multiparous female). * represents p < 0.05. ** represents p < 0.01.

Discussion

We showed that both sexes of gerbils, especially MFs, exhibited aggressive responses toward intruders. High-frequency calls were related with non-agonistic interactions, while low-frequency calls, which might represent aggression, were more related to agonistic interactions than non-agonistic interactions. Low-frequency calls were mostly observed in MF–VF pairs, the pair-test that showed the most aggressive interaction. This is the first report that quantified the relationship between each call and aggressive behaviors and demonstrated the modulations of vocalization during agonistic encounters by sexual experiences in Mongolian gerbils.

We confirmed that male gerbils showed aggressive behaviors, and neither the latency of aggressive behavior nor the duration of aggressive behavior was significantly different between VM and EM (Fig 2). Many previous reports have studied aggressive interactions in gerbils. A study [27] showed that gerbils attack unfamiliar conspecifics and other rodent species. Moreover, EMs tend to be more aggressive toward VFs [13]. VMs exhibit aggressive behavior toward same-sex intruders [16]. On the other hand, another study reported that aggressive behaviors were not observed in VM-VM interactions [28]. In addition, another study showed that males were more aggressive during parental care of pups compared with the period of cohabitation with females [29]. Divergence among reports may be attributed to differences in the testing environment or breeding conditions. Mongolian gerbils exhibit territorial aggression in a considerably large territory [11,12]. Therefore, the apparatus size for the resident-intruder test might have effects on aggressiveness. In larger areas (e.g., 3m × 3m field), resident gerbils might exhibit a different repertoire of aggressive behaviors toward the intruder. This factor should be carefully examined in future studies. Furthermore, we defined males who experienced both mating and parenting as sexually experienced. If parenting is not a part of the criteria in other studies, this might cause differences between the previous and present findings.

We also observed aggressive behaviors in females, especially in MFs. Our data indicate that aggression in female gerbils was increased by parenting experience. Indeed, the normalized duration of aggressive behavior of the MFs was longer than that of the VFs (Fig 1). Aggressive behaviors in both virgin and multiparous female gerbils have been reported in previous studies. VF Mongolian gerbils exhibit aggressive behavior during same-sex encounters [15]. MF gerbils attack VMs more frequently [13] and VFs [17] than VFs do. Our results are consistent with these previous findings. It is well known that females of laboratory rodents, such as rats and mice, that are not lactating or pregnant are less aggressive than males [3,30-32]. From this point of view, MF gerbils could be a suitable model for studying female aggression.

Several studies in rodents have debated the relationship between aggressive behavior and the estrus cycle in some rodents. Ovariectomy reduces aggression in bank voles (Myodes glareolus) [33] and Syrian hamsters (Mesocricetus auratus) [34]. On the other hand, a study showed that female Wistar rats attacked an intruder female, independent of their estrous cycle [35]. Similarly, aggression was not affected by the estrous stage in female Siberian hamsters (Phodopus sungorus) [36]. A recent study showed that diestrus female Mongolian gerbils exhibited aggressive behaviors, but the study did not compare diestrus females with estrus females [15]. In the present study, the estrus cycle was not determined in any of the groups of female gerbils. Further studies are needed to investigate the relationship between aggressive behavior and the estrus cycle in Mongolian gerbils.

We found that gerbils emitted nine types of vocalizations during the resident-intruder test. The spectral shapes of these calls were basically consistent with those reported in previous studies [17,18]. Alert calls described in the same study [17] were not observed in our resident-intruder tests. The nine call types are categorized into two groups; high-frequency vocalization and low-frequency vocalization. It has been reported that Mongolian gerbils emit high-frequency vocalizations during amicable interactions, while low-frequency vocalizations seem to relate to aggression [17,18]. Mongolian gerbils have the good ability to perceive low-frequency sounds than the other small rodents, such as rats and mice [37]. Low-frequency sounds could conduct for a long distance. Hence, in the natural environment, Mongolian gerbils might use the low-frequency vocalizations to communicate with distant partners rather than the high-frequency vocalizations. Several studies in other rodents also have investigated the differences in vocalization frequency in different contexts. Some reports confirmed that rats emit 50 kHz high-frequency vocalizations when they approach, investigate their partners [4,38], and even in the aggressive situation as well as play [39]. Similarly, the number of ultrasonic vocalizations (USVs) among male mice increases with an increase in interaction time [40], and resident female mice emit USVs when interacting with a novel female intruder [41]. Broadband calls, which are composed with multiple harmonics at the range of 1–20 kHz, of Siberian hamsters were produced during aggressive behavior [42]. The bark calls, which have a fundamental frequency around 20 kHz with multiple harmonics, were positively correlated with an increase in aggressive behavior in California mice (Peromyscus acalifornicus) [43]. Some previous studies reported that the 22 kHz vocalizations emitted by an intruder inhibited biting attacks from the resident rat [4,44,45], while another report did not confirm this effect [46]. Morton has claimed an idea that mammals and birds use the low-frequency sounds when hostile and use higher frequency when frightened, appeasing, or approaching in a friendly manner, by reviewing a number of literatures [47]. Collectively, previous reports suggested that higher frequency calls accompanied friendly or non-agonistic interactions, while lower frequency calls related to agonistic conflicts in many rodents.

Consistent with this idea, our results showed that high-frequency vocalizations were accompanied by a reduction of the duration of aggressive behavior. This implies that vocalizations might have some influence on behavioral output including aggressiveness. However, it remains unknown whether this high-frequency call develops an amicable relationship with the intruder or merely represents lower aggression of both the residents and intruders. Here, we observed that high-frequency vocalization in gerbils was strongly related to lower conflicts between the two individuals. Some reports studied the influence of 50 kHz USV on inter-individual behaviors in rats. Devocalization caused the reduction of play behaviors, while the aggressive behaviors were increased [39,48]. Also, the playback experiment of 50 kHz USV promoted social approach behaviors [49]. These previous findings suggest that high-frequency vocalizations affect on behavioral interactions and help developing an amicable relationship at least in rat.

As for low-frequency vocalizations, the numbers of low-frequency calls had a positive correlation with agonistic interactions during the resident-intruder test and many low-frequency calls were observed during not only agonistic interactions but also during the entire period of the resident-intruder test. Behavioral repertories during agonistic interactions consist of multiple subsets, including physical attacks, vocalizations, and changing posture. Indeed, orchestrated multiple responses, such as combinations of biting, arching, and hissing, are known to be triggered by electrical stimulation of the hypothalamus in cats [50,51]. These responses (except for physical attack) are less apparent and have not been studied as much in other laboratory rodents. Low-frequency calls emitted in MF-VF gerbil interactions would be a good characteristic behavior to study multiple aspects of the behavioral response during agonistic events in rodents. In addition, there is a possibility that low-frequency vocalizations could be linked to not only agonistic interactions but also important characteristics of a vocalizer, such as sex, reproductive state, or social experiences. Further studies are required to examine these possibilities.

Finally, we showed that MF-VF pairs emitted low-frequency vocalizations more, while high-frequency vocalizations less, compared with that in the other three groups (Fig 4B). Even though our experiment could not determine whether the resident or intruder emitted a specific call, probably MFs (resident) would emit low-frequency calls during the resident-intruder test since low-frequency calls were rarely observed in VF-VF interactions. Of course, as reported by Ter-Mikaelian et al. [17], a submissive subject could also emit these low-frequency calls. We assume that this actually happened to some extent. However, given that low-frequency calls were recorded even during the period before the resident started attacking the intruder in MF-VF pairs (this was not observed in the other groups), it could be possible that some of low-frequency calls were emitted by MF residents.

These results indicated that vocalizations induced in a specific context could also be modulated by past experiences concerning reproduction. A previous study showed that male mice emitted more USVs after the sociosexual experience [52]. Low-frequency vocalizations increased during same-sex encounters in female pair-bonded California mice [43]. Collectively, these reports and our findings indicate that vocalizations could be modulated by mating experience. However, it remains unknown why MF-VF pair emitted more low-frequency calls compared with those in the other three pairs. We hypothesize that vocalizations are influenced by hormonal changes induced by parenting experiences in female Mongolian gerbils. To elucidate how hormonal changes trigger different patterns of vocalization and the main role of this change, we will carry out refined manipulations of emotion- and vocalization-related neural systems in the future.

(XLSX)

Click here for additional data file.

22 Feb 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Ludek Bartos

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: I Don't Know

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This MS is focused on relationship between agonistic behaviour and the number of high-frequency (above 25 kHz) and low-frequency (below 25 kHz) vocalizations during same-sex dyad encounters of sexually experienced and unexperienced (virgin) male and female Mongolian gerbils. The call rate of high-frequency vocalizations was negatively correlated with the duration of aggressive behavior, while the call rate of low-frequency vocalizations was negatively correlated with the latency of aggressive behavior and positively correlated with the duration of aggressive behavior.

I found the following main problems with MS.

The aim of this study not well formulated and should be re-written.

Introduction is focused on the Mongolian gerbils and does not provide the understanding the place of this particular study within the investigated problem, although many similar studies have been conducted on rats, mice and other rodents, including the gerbils.

In Methods, the authors do not indicate, how it was determined who is the caller during aggression, the resident (attacker) or intruder (defender). These limitations should be clearly explained. The authors should indicate, on what they based their separation to the aggressive/agonistic and non-agonistic behaviors. The authors should indicate how they classified the calls to the high- and low-frequency. Was it done based on visual inspection of spectrograms or measuring fundamental frequency?

The Results often represent mix of results and discussion; this should be clearly separated in the revised version.

Discussion is looking as some draft of literature review rather than discussing own results, and should be re-written concisely and logically, in relation to the obtained results. Discussion also contains and discusses the conclusions which are not supported by the results of this study.

The MS is difficult to read, although English is mostly appropriate. In the revised version, the authors should provide line numbers throughout the text, because numeration of each paper separately complicates enormously the work of Reviewer, Editor and Authors themselves.

Abstract

L. 9-10 We also confirmed two types of vocalisations: high-frequency (>25 kHz) and low-frequency (<25 kHz)

This is strange subdivision. Commonly, classifying calls in the Mongolian gerbil is to ultrasonic calls over 20 kHz and human-audible calls below 20 kHz.

L. 10 with multiple harmonics

This has not sense. Harmonics are integer multiples of fundamental frequency (the lowest frequency band). If the fundamental frequency is low, they are spaced more densely, if the fundamental frequency is high, they are spaces more rarely. So, you just do not see the harmonics of high-frequency calls.

L. 10-14

The call rate of high-frequency vocalizations was negatively correlated with the duration of aggressive behavior, while the call rate of low-frequency vocalizations was negatively correlated with the latency of aggressive behavior and positively correlated with the duration of aggressive behavior.

Who vocalized during the aggressive encounters: the victim, the victor, both fighters?

Introduction

L. 14-15 As it is well known, laboratory rodents such as mice or rats rarely exhibit this extent of aggression.

This statement needs in references, please provide them.

L. 1-6. Although several studies cited in the previous paragraph have reported aggression-specific vocalizations, it remains unclear how sex and sexual experience affect the call rate of various vocalizations in Mongolian gerbils. Thus, we analyzed vocalizations between four different pairs (virgin male–virgin male; experienced male– virgin male; virgin female–virgin female; multiparous female–virgin female) of Mongolian gerbils through the same-sex resident-intruder test.

The aim of the study is poorly formulated and does not reflect the content of the paper. Did you analysed vocalizations, call rates and behaviour latencies? For what? Please re-write the aim accordingly to you real results.

Methods

Same-sex resident-intruder test

L 16 We conducted resident-intruder tests

Добавьте 26 перед resident-intruder tests

L 29 - 04 The observed behaviors were classified as follows: alert posture, ano-genital sniff, approach, attack, clinch, chase, dig, explore, fight, flee, jump, move away, nasal sniff, push, self-groom, stop moving, and watch. ‘attack’, ‘chase’, ‘clinch’, ‘flee’, and ‘push’ were scored as aggressive/agonistic behaviors. The rest was scored as non-agonistic interaction.

It is necessary to provide the references to papers (they are numerous) and/or short descriptions of the forms of aggressive and non-aggressive behaviour of rodents, to make clear for the reader, on what you based your separation to the aggressive/agonistic and non-agonistic behaviors. Why fight is not assigned to the aggressive/agonistic behaviors?

Recording and analysis of vocalizations

L. 24 Sonograms

Thereafter you use more correct term “spectrogram”. Use it here too.

L 1 Delete total duration after F0

L 16-17 Thus, we identified nine calls in this paper; U-shape, UFM-s, UFM, AFM, DFM-l, DFM, QCF, NB, and QCF-NB

Please provide here as an illustration Fig.3A as Fig.

Please make it clear, whether these nine call types could be only low-frequency or only high-frequency or both low- and high-frequency?

The authors should indicate how they classified the calls to the high- and low-frequency (for example, if F0max=30 kHz and F0min=20 kHz). Was it done based on visual inspection of spectrograms or measuring fundamental frequency?

Statistical analysis

L 8-9 number of high-frequency vocalizations per minute; latency of aggressive behavior and the number of low-frequency vocalizations per minute;

Please explain in the previous section, how did you measure the number of high-frequency vocalizations per minute and number of low-frequency vocalizations per minute. From the entire test duration or only from the duration after the first aggression?

Results

Same-sex resident-intruder test

L 3-4 We conducted resident-intruder tests to assess the extent of territorial aggression in gerbils. All intruders were virgin and of the same sex as the resident.

Delete, these are not the results

L 4-7 Many of the residents expressed aggressive behaviors toward the intruder for several minutes and frequently emitted calls. The vast majority of aggressive behaviors were initiated by residents, thus basically we focused on the behaviors of residents in this study.

Please decode what are Many of the residents and The vast majority. In the Results, you should provide the digits, not judgements.

L 4-5 Many of the residents expressed aggressive behaviors toward the intruder for several minutes and frequently emitted calls

How the authors determined who is calling, the resident or the intruder? In gerbils, both can call. The low-frequency calls produce the fleeing animals during aggressive contacts (Ter-Mikaelian et al. 2012, cited in MS). This is a key question, which should be addressed in the Methods.

Analysis of vocalizations of gerbils during the same-sex resident-intruder tests

L. 14-15. Analysis of vocalizations of gerbils during the same-sex resident-intruder tests

Delete “Analysis of”. Here, you describe the results of analysis rather than analysis itself.

L 16 Gerbils frequently emitted calls during the resident-intruder test

Delete this sentence

L. 18. We detected nine calls of vocalizations

This sounds senseless. Do you mean “we detected nine call types”?

L 20-23 The spectral shapes of calls (Fig. 3A; except the U-shape call, which was newly defined in this study) in all pairs during the resident-intruder test were similar to those reported in previous studies [17, 18]. Alert calls described in the same study [17] were not observed in our resident-intruder tests.

This belongs to Discussion. Data from previous studies should not be discussed in the Results section.

L. 20 spectral shapes

This sounds unclear and imprecise. Replace with “contour shapes”.

L. 17-18 It has been reported that the frequency of vocalizations bears semantic significance with a dividing line at 25 kHz [17, 18]. Vocalizations with a maximum fundamental frequency above 25 kHz were observed during non-agonistic interactions, whereas those below 25 kHz were observed during agonistic interactions. The separation at 25 kHz is consistent with our observations (Fig. 3B).

This part of text represents a mix of Methods, own Results and Discussion. It should be separated accordingly to these sections. Please substantiate the separation to the low-frequency and high-frequency calls in Introduction. Please provide the criteria for separation to these categories in Methods. Otherwise, Statistical analysis remains perfectly unclear.

L. 3-5 high-frequency with short duration (U-shape, UFM-s, and UFM; 32.7 ± 2.0 kHz, 25.9 ± 14.7 ms), and low-frequency with long duration (AFM, DFM-l, DFM, QCF, NB, and QCF-NB; 12.1 ± 2.9 kHz, 160.4 ± 73.3 ms)

Please indicate that the fundamental frequency (maximal?) and the duration are provided for all high-frequency and low-frequency криков. Please indicate the numbers of the high-frequency and low-frequency calls.

Fig. 3 legend

Replace “submitted to” with “in”

Fig. 3 legend MF-VF pairs seemed to have a different composition of syllables from the other pairs.

Transfer this from the figure legend to main text.

Table 1. Spectro-temporal features of nine calls.

Replace “nine calls” “nine call types” here and in the sentence under the Table 1. Please add to the legend designations/ decoding of all acoustic variables presented in the column headings.

Table 1. Spectro-temporal features of nine calls.

Please indicate n (call number) for each call type

Table 1. UFM

Mean Min. freq. for UFM is 21.34±9.00 kHz. This means that most UFM calls have the mean fundamental frequency lower 25 kHz, what is the criterion for separation between the low-frequency and high-frequency calls. Please substantiate, why you assign all calls of this type to the high-frequency calls. It seems that, in this study you conduct the border between the low-frequency and high-frequency calls at 20 kHz.

L 3 between the four pairs.

Please replace “four pairs” with four tests.

L 16-17 In addition, there was a negative correlation between the numbers of high- and low-frequency calls

Please add for the duration of a test after calls

Fig 4. Numbers of two types of calls by pairs during resident-intruder tests.

Please consider re-writing to “Numbers of two call types by resident-intruder pairs.

Relationship between vocalizations and aggressive interactions during the same-sex resident-intruder tests

L. 2-3. Relationship between vocalizations and aggressive interactions during the same-sex resident-intruder tests

Please re-write this heading to make it clear for the reader. What relationship you look, during (along) the test or between tests?

L. 4-5. We then examined the relationship between vocalizations and behaviors, with an emphasis on aggressive behaviors.

Delete, these are not the Results

L. 5 First, we created the ethograms for each pair

You use “pair” for designations of particular tests (one of 26), and also for designations of one of the four test categories (MF-VF etc.). This makes understanding the text impossible. Please correct terminology throughout the MS.

L. 5-7 First, we created the ethograms for each pair (examples shown in Fig. 5A) and categorized the behaviors into two categories, non- agonistic or agonistic (indicated by blue or red in Fig. 5A). Then, we quantified the type of behavior at the timing of high- or low-frequency calls emitted

This is repetition (first sentence) and amendments to Methods. If the authors provide in the Results something that was not indicated in the Methods, this should be transferred to Methods. How the authors confronted type of behavior at the timing of high- or low-frequency calls emitted? Behaviour was recorded by video, whereas the calls by UltraSoundGate in the form of ultrasonic files. How it was synchronized? Please described this in detain in the Methods.

L. 8 Replace Fig. 5B with Fig. 5A.

Discussion

P.21 L. 24-26 Furthermore, we defined males who experienced both mating and parenting as sexually experienced. If parenting is not a part of the criteria in other studies, this might cause differences between the previous and present findings.

Was it really part of your results? At least, it is lacking in the aim of MS.

L 9-12 bank voles [33] and Syrian hamsters .... Siberian hamsters [36].

Add Latin names

P. 22 L. 20-21 These nine calls are categorized into two types; high-frequency vocalization and low-frequency vocalization with harmonics.

This belongs to Methods, delete this.

L 3 California mice [41].

Add Latin name

L 19-21 These correlations strongly suggest that low-frequency calls would represent aggression or serve as a threat and warning consistent with the idea implied in previous reports [17, 18].

This conclusion is not supported with results of this study. Indeed, you do not know who emits the low-frequency calls, aggressor or defender. Delete this.

L 26-01 Low-frequency calls emitted in MF-VF gerbil interactions would be a good characteristic behavior to study multiple aspects of the expression of aggression in rodents. In addition, there is a possibility that low-frequency vocalizations could be linked to not only aggression but also important characteristics of a vocalizer, such as sex, reproductive state, or social experiences.

Again, your data do not show which animal is vocalizer, the winner or defender. During encounters of rodents, as a rule, the defender produces the human-audible (below 20 kHz) calls. The defender attacks silently.

In rats, human-audible (below 20 kHz) calls are emitted by the defending individuals during agonistic interactions (squeal, Watts, 1980), during tail-clamp (Chen et al. 2017) and in response to electrical nociceptive stimuli (Jourdan et al., 1995).

Watts, 1980. Vocalizations of nine species of rat (Rattus; Muridae). J. Zool., 191:531-555.

Chen et al., 2017. Call divergence in three sympatric Rattus species. J. Acoust. Soc. Am. 142:29-34.

Jourdan et al., 1995. Audible and ultrasonic vocalization elicited by single electrical nociceptive stimuli to the tail in the rat. Pain, 63:237-249.

L 6-8 MFs (resident) most likely emitted low-frequency calls during the resident-intruder test since low-frequency calls were rarely observed in VF-VF interactions

Please indicate in Methods in detail, how did you determine, who emitted the low-frequency calls during the resident-intruder test. This is important thing, as your conclusions that low-frequency calls belong to aggressor rather than defender contradict to data by Ter-Mikaelian et al. 2012 (cited in MS) on Mongolian gerbils, as well as with data on other species of gerbils (see below).

Volodin et al., 1994. Situational changes in vocalization of Great gerbils (Rhombomys opimus Licht.) during defensive behavior. Doklady Biological Sciences, 334:65-68.

Volodin I.A., Goltsman M.E., 2000. Acoustic activity displayed in the agonistic behavior of Great and Light gerbils. Doklady Biological Sciences, 371:176-178.

L 10-12 However, given that low-frequency calls were recorded even during the period before the resident started attacking the intruder, it would be reasonable to consider that the majority of low-frequency calls were emitted by MF residents.

This is unsupported claim. The animal introduced on the territory of resident afraid of it and start calling.

Reviewer #2: Comments to the Author

Behaviors and vocalizations associated with aggression are essential for animals

to survive, reproduce, and maintain their community. This study reported that high- and low-frequency vocalizations relate to non-agonistic and agonistic interactions during encounters in Mongolian gerbils, respectively, which related to the sexual experience. These findings provide new insights into the modulatory effects of sex and sexual experience on vocalizations during agonistic encounters.

I have some major comments:

1) Why do intruders only use virgin gerbils ？

2) Whether or not there are difference between the structure of vocalizations before the attack and the after the attack? It might be possible that fighting has been expressed in voice before the beginning of physical conflict, which is a characteristic of vocalizations in the aggressive strategy.

3) It should be added to some explanations on the ecological function of the high- and low-frequency vocalizations related to non-agonistic and agonistic interactions in the discussion.

The minor comments:

1) Is it more appropriate to change the title to“ Vocalization during agonistic encounter in Mongolian gerbils：impact of sexual experience.”

2) Result Lines 3-7 should belong to the method behavior observation part，that is part of“same-sex resident-intruder test” , and it is recommended to adjust.

3) Page 23 line 13 “These results indicate that vocalizations….” should revised” These results indicated that vocalizations….”

Reviewer #3: This is a fantastic publication exploring not only vocalization production in Mongolian gerbils but also how sex and sexual experience effects aggression and the expression of these calls. I have a few minor points and questions to be addressed, some areas that require clarification or a better explanation, my comments are below.

Abstract

Line 3 – maintain their community, does aggression do this? Or would it be more appropriate to say maintain their social hierarchy?

Line 12 – perhaps spilt these sentences

Introduction

Line 5 – define clinching, and make sure the references cover this behaviour (1,2 don’t cover clinching, I couldn’t find clinching in 3, but of it is defined in reference 3 then ignore this comment)

Methods

Page 1 – line 29 what software was used to score the videos?

Page 2 – line 1-4 a table describing these behaviours would be helpful. Additionally, are these behaviours arbitrarily selected or have they been previously used/described?

Page 3 – If the calls were selected using a MATLAB script was this program 100% accurate at distinguishing calls? Also was this checked by an experimenter? Further, were the calls classified by an experimenter or was this also performed by the MATLAB script?

Page 4 – chi square symbol isn’t showing up

Page 4 – please explain why a non-parametric test was used instead of an ANOVA

Results

Page 1 - line 3 – in the gerbils

Page 2 – line 18 – nine calls of vocalizations should be nine types of vocalizations

Page 3 – line 1 – why were the calls grouped in those parameters?

Discussion

Page 1 – line 20 – territorial sentence could be re-worded so it flows better

Page 2 – line 24 – rats actually make 50-khz USV when they are in aggressive situations as well as play (but 22kHz calls are fairly unique to aggression, just like what you saw!) see Burke et al., 2017 (Avoiding escalation from play to aggression in adult male rats: The role of ultrasonic calls)

Page 3 – I think that these are great points, again the rat literature really mimics your findings so potentially a comparison to this literature would really make your point a bit stronger (recent publications from Pellis/Burke; Wohr/Schwarting/Kisko) This is just a suggestion, and is not at all necessary for the publication.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Comments to the Author-PONE-D-21-38093.docx

Click here for additional data file.

18 Mar 2022

Editor

As detailed below, we have addressed all of the reviewer’s reservations.

Here are responses to comments which are not written in the reviewer’s comments.

>It seems to me you used the subjects repeatedly on the tests. If not, you should clearly say it.

In the original text, this is clearly stated that “Each subject (resident and intruder) underwent the resident-intruder test for a single time.”.

>Your analysis did not show the interrelationships between the factors, such as the age of the subjects, their sexual experience, the number of animals kept in one plexiglass cage, the latency and duration of aggressive behaviours, etc.

The age of animals might have some effect on the behaviors, however, we did not conduct enough experiments to examine the effect. We raised the gerbils under the guideline which is internationally well accepted. We think that usually this factor “the number of animals kept in one plexiglass cage” is not considered in this kind of experiment. For the rest, we compared by using no-parametric tests. We used no-parametric tests, since the acquired data did not show normal distributions.

>The uncategorized correlation in Fig. 2 is misleading. You present a negative correlation. However, it seems the case for MF-VF, not that much for EM-VM, and it is unlikely for the rest of the categories. (Similarly, in Fig. 6 B, C, and D.)

We tested the correlation for each subgroup and there was no statistical significance. We removed the sentence emphasizing the correlation from the abstract and discussion.

Reviewer #1

Thank you for your kind suggestions to improve this manuscript.

As detailed below, we have addressed all of the reviewer’s reservations.

The changes are highlighted in the marked text and “LXX” corresponds to lines in the marked text.

Abstract

L. 9-10 We also confirmed two types of vocalisations: high-frequency (>25 kHz) and low-frequency (<25 kHz)

This is strange subdivision. Commonly, classifying calls in the Mongolian gerbil is to ultrasonic calls over 20 kHz and human-audible calls below 20 kHz.

Author’s response

We agree that the separation at 25 kHz seems arbitral and not common. However, we think that the separation human-audible or not is also an arbitral separation. Then, we calculated the fitting functions of two normal distributions for high- and low- vocalization distributions. The intersection of the two functions is used as the objective frequency border between the high-frequency and the low-frequency vocalizations. The number was “24.6 kHz”. This is now explained in the method section (L176-182).

L. 10 with multiple harmonics

This has not sense. Harmonics are integer multiples of fundamental frequency (the lowest frequency band). If the fundamental frequency is low, they are spaced more densely, if the fundamental frequency is high, they are spaces more rarely. So, you just do not see the harmonics of high-frequency calls.

Author’s response

We think that this explanation will help to grasp the character of the vocalization, for the readers not familiar with sound physics.

L. 10-14

The call rate of high-frequency vocalizations was negatively correlated with the duration of aggressive behavior, while the call rate of low-frequency vocalizations was negatively correlated with the latency of aggressive behavior and positively correlated with the duration of aggressive behavior.

Who vocalized during the aggressive encounters: the victim, the victor, both fighters?

Author’s response

We did not specify the caller, because of technical limitations. Probably both emitted the calls.

We changed the sentence to “At the timing of high-frequency vocalizations observed during the tests, the vast majority (96.2%) of the behavioral interactions were non-agonistic. While at the timing of low-frequency vocalizations observed during the tests, around half (45%) of the behavioral interactions were agonistic.” (L31-35).

Introduction

L. 14-15 As it is well known, laboratory rodents such as mice or rats rarely exhibit this extent of aggression.

This statement needs in references, please provide them.

Author’s response

We added references (L56-57).

L. 1-6. Although several studies cited in the previous paragraph have reported aggression-specific vocalizations, it remains unclear how sex and sexual experience affect the call rate of various vocalizations in Mongolian gerbils. Thus, we analyzed vocalizations between four different pairs (virgin male–virgin male; experienced male– virgin male; virgin female–virgin female; multiparous female–virgin female) of Mongolian gerbils through the same-sex resident-intruder test.

The aim of the study is poorly formulated and does not reflect the content of the paper. Did you analysed vocalizations, call rates and behaviour latencies? For what? Please re-write the aim accordingly to you real results.

Author’s response

We modified the paragraph (L71-78).

Methods

Same-sex resident-intruder test

L 16 We conducted resident-intruder tests

Добавьте 26 перед resident-intruder tests

Author’s response

We added the number (L95).

L 29 - 04 The observed behaviors were classified as follows: alert posture, ano-genital sniff, approach, attack, clinch, chase, dig, explore, fight, flee, jump, move away, nasal sniff, push, self-groom, stop moving, and watch. ‘attack’, ‘chase’, ‘clinch’, ‘flee’, and ‘push’ were scored as aggressive/agonistic behaviors. The rest was scored as non-agonistic interaction.

It is necessary to provide the references to papers (they are numerous) and/or short descriptions of the forms of aggressive and non-aggressive behaviour of rodents, to make clear for the reader, on what you based your separation to the aggressive/agonistic and non-agonistic behaviors. Why fight is not assigned to the aggressive/agonistic behaviors?

Author’s response

Thank you for pointing it out. We added a reference with which we made the classification and made a table to explain each behavior (L129). Also, we fixed the term lists for the aggressive behaviors (L111-116).

Recording and analysis of vocalizations

L. 24 Sonograms

Thereafter you use more correct term “spectrogram”. Use it here too.

Author’s response

We replaced the word ‘sonograms’ with ‘spectrogram’ (L142).

L 1 Delete total duration after F0

Author’s response

‘total duration’ was deleted (L147).

L 16-17 Thus, we identified nine calls in this paper; U-shape, UFM-s, UFM, AFM, DFM-l, DFM, QCF, NB, and QCF-NB

Please provide here as an illustration Fig.3A as Fig.

Please make it clear, whether these nine call types could be only low-frequency or only high-frequency or both low- and high-frequency?

The authors should indicate how they classified the calls to the high- and low-frequency (for example, if F0max=30 kHz and F0min=20 kHz). Was it done based on visual inspection of spectrograms or measuring fundamental frequency?

Author’s response

We provided Fig.3A (now Fig.1) here (L184).

We categorized calls using their maximum fundamental frequency. Calls F0max ≥24.6 kHz were assigned to high-frequency calls and calls F0max <24.6 kHz were assigned to low-frequency calls (the calculation to determine this border was described above and now in the method section). No call was assigned into both high & low calls. (L176-182)

Statistical analysis

L 8-9 number of high-frequency vocalizations per minute; latency of aggressive behavior and the number of low-frequency vocalizations per minute;

Please explain in the previous section, how did you measure the number of high-frequency vocalizations per minute and number of low-frequency vocalizations per minute. From the entire test duration or only from the duration after the first aggression?

Author’s response

Vocalization per minute was calculated from the entire test session.

We added explanations to make this clear (L192-195).

Results

Same-sex resident-intruder test

L 3-4 We conducted resident-intruder tests to assess the extent of territorial aggression in gerbils. All intruders were virgin and of the same sex as the resident.

Delete, these are not the results

Author’s response

We deleted these (L206-210).

L 4-7 Many of the residents expressed aggressive behaviors toward the intruder for several minutes and frequently emitted calls. The vast majority of aggressive behaviors were initiated by residents, thus basically we focused on the behaviors of residents in this study.

Please decode what are Many of the residents and The vast majority. In the Results, you should provide the digits, not judgements.

Author’s response

We provided the actual numbers, and this paragraph was moved to the method section (L104-107).

L 4-5 Many of the residents expressed aggressive behaviors toward the intruder for several minutes and frequently emitted calls

How the authors determined who is calling, the resident or the intruder? In gerbils, both can call. The low-frequency calls produce the fleeing animals during aggressive contacts (Ter-Mikaelian et al. 2012, cited in MS). This is a key question, which should be addressed in the Methods.

Author’s response

We did not specify the caller. We deleted this sentence about calls (L104-107). Sorry for the confusion.

Analysis of vocalizations of gerbils during the same-sex resident-intruder tests

L. 14-15. Analysis of vocalizations of gerbils during the same-sex resident-intruder tests

Delete “Analysis of”. Here, you describe the results of analysis rather than analysis itself.

Author’s response

We deleted this (L248).

L 16 Gerbils frequently emitted calls during the resident-intruder test

Delete this sentence

Author’s response

We deleted this (L250).

L. 18. We detected nine calls of vocalizations

This sounds senseless. Do you mean “we detected nine call types”?

Author’s response

We rewrite this as “We detected nine call types of vocalizations”.

L 20-23 The spectral shapes of calls (Fig. 3A; except the U-shape call, which was newly defined in this study) in all pairs during the resident-intruder test were similar to those reported in previous studies [17, 18]. Alert calls described in the same study [17] were not observed in our resident-intruder tests.

This belongs to Discussion. Data from previous studies should not be discussed in the Results section.

Author’s response

We moved this to the discussion with modification (L257, L418-422).

L. 20 spectral shapes

This sounds unclear and imprecise. Replace with “contour shapes”.

Author’s response

We replaced this (L254).

L. 17-18 It has been reported that the frequency of vocalizations bears semantic significance with a dividing line at 25 kHz [17, 18]. Vocalizations with a maximum fundamental frequency above 25 kHz were observed during non-agonistic interactions, whereas those below 25 kHz were observed during agonistic interactions. The separation at 25 kHz is consistent with our observations (Fig. 3B).

This part of text represents a mix of Methods, own Results and Discussion. It should be separated accordingly to these sections. Please substantiate the separation to the low-frequency and high-frequency calls in Introduction. Please provide the criteria for separation to these categories in Methods. Otherwise, Statistical analysis remains perfectly unclear.

Author’s response

We added the criteria to separate high- and low-frequency calls in the method section (L176-182).

Also, part of this section was moved to methods and discussions.

L. 3-5 high-frequency with short duration (U-shape, UFM-s, and UFM; 32.7 ± 2.0 kHz, 25.9 ± 14.7 ms), and low-frequency with long duration (AFM, DFM-l, DFM, QCF, NB, and QCF-NB; 12.1 ± 2.9 kHz, 160.4 ± 73.3 ms)

Please indicate that the fundamental frequency (maximal?) and the duration are provided for all high-frequency and low-frequency криков. Please indicate the numbers of the high-frequency and low-frequency calls.

Author’s response

F0 for each call is shown in table 2. The numbers of the high-frequency (3329) and low-frequency calls (1033) are now indicated (L268-270).

Fig. 3 legend

Replace “submitted to” with “in”

Author’s response

We replaced “submitted to” with “during” (L272).

Fig. 3 legend MF-VF pairs seemed to have a different composition of syllables from the other pairs.

Transfer this from the figure legend to main text.

Author’s response

We deleted this sentence (L277).

Table 1. Spectro-temporal features of nine calls.

Replace “nine calls” “nine call types” here and in the sentence under the Table 1. Please add to the legend designations/ decoding of all acoustic variables presented in the column headings.

Author’s response

We followed your suggestions (L279).

Table 1. Spectro-temporal features of nine calls.

Please indicate n (call number) for each call type

Author’s response

Now those are indicated in table 2 (L280-285).

Table 1. UFM

Mean Min. freq. for UFM is 21.34±9.00 kHz. This means that most UFM calls have the mean fundamental frequency lower 25 kHz, what is the criterion for separation between the low-frequency and high-frequency calls. Please substantiate, why you assign all calls of this type to the high-frequency calls. It seems that, in this study you conduct the border between the low-frequency and high-frequency calls at 20 kHz.

Author’s response

We categorized calls using their maximum fundamental frequency. Calls F0max ≥24.6 kHz were assigned to high-frequency calls and calls F0max <24.6 kHz were assigned to low-frequency calls (the calculation to determine this new border was described above and now in the method section; L176-182).

L 3 between the four pairs.

Please replace “four pairs” with four tests.

Author’s response

We replaced “four pairs” with “four tests” (L288).

L 16-17 In addition, there was a negative correlation between the numbers of high- and low-frequency calls

Please add for the duration of a test after calls

Fig 4. Numbers of two types of calls by pairs during resident-intruder tests.

Please consider re-writing to “Numbers of two call types by resident-intruder pairs.

Author’s response

We rewrote as suggested (L305).

Relationship between vocalizations and aggressive interactions during the same-sex resident-intruder tests

L. 2-3. Relationship between vocalizations and aggressive interactions during the same-sex resident-intruder tests

Please re-write this heading to make it clear for the reader. What relationship you look, during (along) the test or between tests?

Author’s response

We rewrote the heading as “Relationship between each vocalization and aggressive interactions during the same-sex resident-intruder tests” (L314).

L. 4-5. We then examined the relationship between vocalizations and behaviors, with an emphasis on aggressive behaviors.

Delete, these are not the Results

Author’s response

The sentence was deleted (L316-317).

L. 5 First, we created the ethograms for each pair

You use “pair” for designations of particular tests (one of 26), and also for designations of one of the four test categories (MF-VF etc.). This makes understanding the text impossible. Please correct terminology throughout the MS.

Author’s response

We changed “pairs” to “pair-tests” in some text which might cause confusion. For this sentence, “each pair” was replaced with “26 tests” (L317).

L. 5-7 First, we created the ethograms for each pair (examples shown in Fig. 5A) and categorized the behaviors into two categories, non- agonistic or agonistic (indicated by blue or red in Fig. 5A). Then, we quantified the type of behavior at the timing of high- or low-frequency calls emitted

This is repetition (first sentence) and amendments to Methods. If the authors provide in the Results something that was not indicated in the Methods, this should be transferred to Methods. How the authors confronted type of behavior at the timing of high- or low-frequency calls emitted? Behaviour was recorded by video, whereas the calls by UltraSoundGate in the form of ultrasonic files. How it was synchronized? Please described this in detain in the Methods.

Author’s response

We think this explanation will help readers to understand the flow, even it is containing a kind of method. The timing was synchronized with the sound in the video and the audio file. We added this explanation in the method section (L181-182).

L. 8 Replace Fig. 5B with Fig. 5A.

Author’s response

We think the original is correct.

Discussion

P.21 L. 24-26 Furthermore, we defined males who experienced both mating and parenting as sexually experienced. If parenting is not a part of the criteria in other studies, this might cause differences between the previous and present findings.

Was it really part of your results? At least, it is lacking in the aim of MS.

Author’s response

This is the discussion to speculate the difference of results between the reports including ours.

L 9-12 bank voles [33] and Syrian hamsters .... Siberian hamsters [36].

Add Latin names

Author’s response

We added these Latin names “Myodes glareolus”, “Mesocricetus auratus”, and “Phodopus sungorus” (L407,410).

P. 22 L. 20-21 These nine calls are categorized into two types; high-frequency vocalization and low-frequency vocalization with harmonics.

This belongs to Methods, delete this.

Author’s response

We think that this sentence will help to understand the flow.

L 3 California mice [41].

Add Latin name

Author’s response

We added “Peromyscus acalifornicus” (L439).

L 19-21 These correlations strongly suggest that low-frequency calls would represent aggression or serve as a threat and warning consistent with the idea implied in previous reports [17, 18].

This conclusion is not supported with results of this study. Indeed, you do not know who emits the low-frequency calls, aggressor or defender. Delete this.

Author’s response

We deleted this.

L 26-01 Low-frequency calls emitted in MF-VF gerbil interactions would be a good characteristic behavior to study multiple aspects of the expression of aggression in rodents. In addition, there is a possibility that low-frequency vocalizations could be linked to not only aggression but also important characteristics of a vocalizer, such as sex, reproductive state, or social experiences.

Again, your data do not show which animal is vocalizer, the winner or defender. During encounters of rodents, as a rule, the defender produces the human-audible (below 20 kHz) calls. The defender attacks silently.

In rats, human-audible (below 20 kHz) calls are emitted by the defending individuals during agonistic interactions (squeal, Watts, 1980), during tail-clamp (Chen et al. 2017) and in response to electrical nociceptive stimuli (Jourdan et al., 1995).

Watts, 1980. Vocalizations of nine species of rat (Rattus; Muridae). J. Zool., 191:531-555.

Chen et al., 2017. Call divergence in three sympatric Rattus species. J. Acoust. Soc. Am. 142:29-34.

Jourdan et al., 1995. Audible and ultrasonic vocalization elicited by single electrical nociceptive stimuli to the tail in the rat. Pain, 63:237-249.

L 6-8 MFs (resident) most likely emitted low-frequency calls during the resident-intruder test since low-frequency calls were rarely observed in VF-VF interactions

Please indicate in Methods in detail, how did you determine, who emitted the low-frequency calls during the resident-intruder test. This is important thing, as your conclusions that low-frequency calls belong to aggressor rather than defender contradict to data by Ter-Mikaelian et al. 2012 (cited in MS) on Mongolian gerbils, as well as with data on other species of gerbils (see below).

Volodin et al., 1994. Situational changes in vocalization of Great gerbils (Rhombomys opimus Licht.) during defensive behavior. Doklady Biological Sciences, 334:65-68.

Volodin I.A., Goltsman M.E., 2000. Acoustic activity displayed in the agonistic behavior of Great and Light gerbils. Doklady Biological Sciences, 371:176-178.

L 10-12 However, given that low-frequency calls were recorded even during the period before the resident started attacking the intruder, it would be reasonable to consider that the majority of low-frequency calls were emitted by MF residents.

This is unsupported claim. The animal introduced on the territory of resident afraid of it and start calling.

Author’s response

First of all, we have not determined which gerbil emitted the calls.

These discussions are rational speculations, not definite conclusions. Here we have raised possibilities to be tested with more elaborate experiments in the future.

The reviewer claimed that “During encounters of rodents, as a rule, the defender produces the human-audible (below 20 kHz) calls”. However, this is not a general rule. Occasionally, offending rats emit 22 kHz calls during the encounter as reported in Burke et al. 2017. Also, Morton has claimed that many animals including rodents use low-frequency vocalizations when they are hostile. We agree that aversive stimuli trigger 22 kHz calls in rats and defeated rats tend to emit 22 kHz calls. Meanwhile, many reports also suggested that rats emit 22 kHz calls in alert situations, the proximity of threats or presentation of predator signals (“Handbook of Ultrasonic Vocalization” edited by Brudzynski S.M.; published from Academic Press). In our experiment, resident gerbils could be the aggressive ones and also the ones who perceive threats from the outside. The territorial aggression could be also a response to threat, not only a simple expression of aggression.

Furthermore, Ter-Mikaelian and colleagues did not conclude as the reviewer wrote. I quote the corresponding paragraph here.

“3.2.2. Vocalization characteristics

Aggression calls were emitted during the watch posture, nasal sniff, sidling, or when the submissive animal began to flee. Since the vocalizations were heard when there was no body contact, they probably did not signal pain. It was not possible to determine with certainty which animal emitted the vocalizations during aggressive encounters; the calls may signal submission by the weaker animal or aggression by the dominant animal. However, on certain occasions, it was noted that the fleeing animal emitted vocalizations.”

The authors suggest that this kind of low-frequency call relates to aggression and submission, and clearly state that they were not able to determine the caller.

The experiments conducted by Volodin and colleagues also did not determine the caller, and used the other species of gerbils.

Our results indicated that a considerably large number of low-frequency calls were emitted in the tests MF involved compared with the other 3 tests. It is reasonable to estimate MFs will have different characters than others. Also, Low-frequency calls were confirmed before the first agonistic event in the MF-VF tests. If the intruders emit low-frequency calls because they are afraid of residents, low-frequency calls should be recorded before the first agonistic event in all 4 tests, but this was not the case. Taken together, our estimation that the MFs might emit low-frequency calls must be rational enough.

Reviewer #2

Thank you for your kind suggestions to improve this manuscript.

As detailed below, we have addressed all of the reviewer’s reservations.

The changes are highlighted in the marked text and “LXX” corresponds to lines in the marked text.

The major comments:

Reviewer’s comment 1: Why do intruders only use virgin gerbils?

Author’s response: Since we intended to mainly observe the territorial aggression of residents, virgin gerbils were used as intruders. If the experienced subjects were used as intruders, some of them would exhibit aggressive behaviors actively towards residents. In this experiment, we tried avoiding this.

Reviewer’s comment 2: Whether or not there are difference between the structure of vocalizations before the attack and the after the attack? It might be possible that fighting has been expressed in voice before the beginning of physical conflict, which is a characteristic of vocalizations in the aggressive strategy.

Author’s response: Thank you for your suggestion. We tested this and the result is shown in the L335-347. In brief, there were some modulations both in high- and low-frequency calls.

Reviewer’s comment 3: It should be added to some explanations on the ecological function of the high- and low-frequency vocalizations related to non-agonistic and agonistic interactions in the discussion.

Author’s response: We add discussion sentences about the ecological function (L423-427).

The minor comments:

Reviewer’s comment 1: Is it more appropriate to change the title to“ Vocalization during agonistic encounter in Mongolian gerbils: impact of sexual experience.”

Author’s response: Thank you for your suggestion. We changed the title.

Reviewer’s comment 2: Result Lines 3-7 should belong to the method behavior observation part, that is part of“same-sex resident-intruder test” , and it is recommended to adjust.

Author’s response: We moved the sentences to the method section and adjusted (L104-107).

Reviewer’s comment 3: Page 23 line 13 “These results indicate that vocalizations….” should revised” These results indicated that vocalizations….”

Author’s response: We changed the tense as pointed (L486).

Reviewer #3

Thank you for your kind suggestions to improve this manuscript.

As detailed below, we have addressed all of the reviewer’s reservations.

The changes are highlighted in the marked text and “LXX” corresponds to lines in the marked text.

Reviewer’s comment 1

Abstract

Line 3 – maintain their community, does aggression do this? Or would it be more appropriate to say maintain their social hierarchy?

Line 12 – perhaps spilt these sentences

Author’s response: We modified abstract L3. Please take a look at L24.

The phrase “maintain their community” is replaced with “organize their social hierarchy”

For line 12, we followed your suggestion and modified the sentence (L32-35).

Reviewer’s comment 2

Introduction

Line 5 – define clinching, and make sure the references cover this behaviour (1,2 don’t cover clinching, I couldn’t find clinching in 3, but of it is defined in reference 3 then ignore this comment)

Author’s response: “clinching” was not an appropriate term. We replaced “clinching” with “boxing” in the manuscript. L46, L114

Reviewer’s comment 3

Methods

Page 1 – line 29 what software was used to score the videos?

Author’s response: We manually scored the behaviors by watching the videos. The method was modified to make this clear (L113).

Page 2 – line 1-4 a table describing these behaviours would be helpful. Additionally, are these behaviours arbitrarily selected or have they been previously used/described?

Author’s response: These behaviors are selected by following the previous report (Ter-Mikaelian et al. 2012) with a small modification. The reference was added in this sentence (L111-112). A table was added (L129).

Page 3 – If the calls were selected using a MATLAB script was this program 100% accurate at distinguishing calls? Also was this checked by an experimenter? Further, were the calls classified by an experimenter or was this also performed by the MATLAB script?

Author’s response: Each vocalization was detected and cut out from the audio files manually by experimenters. The classification was also manually conducted by experimenters following the characters described in the ref 17. These explanations were added in the method section (L138-139).

Page 4 – chi square symbol isn’t showing up

Author’s response: We fixed this (L187).

Page 4 – please explain why a non-parametric test was used instead of an ANOVA

Author’s response: Since the acquired data did not show the normal distribution, we used the non-parametric tests.

Reviewer’s comment 4

Results

Page 1 - line 3 – in the gerbils

Author’s response: We changed as pointed out. The sentence was moved in the method section (L211).

Page 2 – line 18 – nine calls of vocalizations should be nine types of vocalizations

Author’s response: We replaced the word “calls” with “call types” (L252).

Page 3 – line 1 – why were the calls grouped in those parameters?

Author’s response: Now, an objective parameter was calculated. We calculated the fitting functions of two normal distributions for high- and low- vocalization distributions. The intersection of the two functions is used as the objective frequency border between the high-frequency and the low-frequency vocalizations. The number was “24.6 kHz”. This is explained in the method section (L176-182).

Reviewer’s comment 5

Discussion

Page 1 – line 20 – territorial sentence could be re-worded so it flows better

Author’s response: We rewrite the sentence as, “Mongolian gerbils exhibit territorial aggression in a considerably large territory [11, 12].” (L387-388).

Page 2 – line 24 – rats actually make 50-khz USV when they are in aggressive situations as well as play (but 22kHz calls are fairly unique to aggression, just like what you saw!) see Burke et al., 2017 (Avoiding escalation from play to aggression in adult male rats: The role of ultrasonic calls)

Author’s response: We added that rats emit USV in an aggressive situation (L430).

Page 3 – I think that these are great points, again the rat literature really mimics your findings so potentially a comparison to this literature would really make your point a bit stronger (recent publications from Pellis/Burke; Wohr/Schwarting/Kisko) This is just a suggestion, and is not at all necessary for the publication.

Author’s response: Thank you for the constructive suggestion! The literature listed are quite encouraging us. We add a comparison in the discussion (L453-458).

Submitted filename: Responses to reviewers.docx

Click here for additional data file.

12 May 2022

Submitted filename: Comments to the Author-PONE-D-21-38093R1.docx

Click here for additional data file.

16 May 2022

Reviewer #1

Thank you for your kind suggestions to improve this manuscript.

As detailed below, we have addressed all of the reviewer’s reservations.

The changes are highlighted in the marked text and “LXX” corresponds to lines in the marked text.

Reviewer’s comment:

L 30. We also confirmed two types of vocalizations during the encounters

Replace "two types" with "two groups of vocalizations" (as in L 176). Otherwise, types of vocalizations confuse with nine call types (see also L 256 and L 405).

Author’s response: We corrected as the reviewer suggested (L30).

Reviewer’s comment:

L 31. high-frequency (>24.6 kHz) and low-frequency (<24.6 kHz) with multiple harmonics

Delete "with multiple harmonics" after "low-frequency". The high-frequency vocalizations also have multiple harmonics. Create the spectrum up to 300 kHz, increase the intensity, and you will see them clearly.

Author’s response: We deleted "with multiple harmonics" (L31).

Reviewer’s comment:

L 106 Aggressive behaviors were initiated by residents in the vast majority (20/24) of tests

Why 24, not 26? You have in total 26 tests.

Author’s response: We corrected the number. The correct number is 20/26 (L106). Thank you for pointing this out.

Reviewer’s comment:

L 163. we identified nine calls

Replace "nine calls" with "nine call types"

Author’s response: We replaced it (L163).

Reviewer’s comment:

L 224 gerbil groups submitted to resident-intruder tests.

Replace "submitted to" with "during".

Author’s response: We replaced it (L224).

Reviewer’s comment:

L 254-256 the nine call types could be categorized into two groups. From the fitting functions of two normal distributions, the separation at 24.6 kHz was determined. Thus, for further analysis, we categorized the quantified nine calls of vocalizations into two types

Nine call types (L 254) cannot be categorized to the two call types. Replace with "we categorized the nine types of vocalizations into two groups"

Author’s response: We replaced it (L256).

Reviewer’s comment:

L 262-263 High-frequency syllables are represented by bluish symbols and low-frequency syllables

Replace "syllables" with "groups" (as earlier in L 176). Avoid synonyms.

Author’s response: We replaced it (L262-263).

Reviewer’s comment:

L 402-406 We found that gerbils emitted nine types of vocalizations during the resident-intruder test. The spectral shapes of these calls were basically consistent with those reported in previous studies [17, 18]. Alert calls described in the same study [17] were not observed in our resident-intruder tests. The nine calls are categorized into two types; high-frequency vocalization and low-frequency vocalization with harmonics.

Nine types of vocalizations (L 402) cannot be categorized to the two call types. Using consistently own terms throughout the text is important for understanding the content by the readers. Replace with "The nine call types are categorized into two groups"

Author’s response: We replaced it (L405-406).

Reviewer’s comment:

L 406 low-frequency vocalization with harmonics.

Delete "with harmonics". Presence of visible harmonics is not exclusively attributive to low-frequency vocalizations, but depends on the size of the used for analysis spectral window.

Author’s response: We deleted "with harmonics" (L406).

Reviewer’s comment:

L 444 resident intruder

Replace with resident-intruder

Author’s response: We replaced it (L444).

Reviewer’s comment:

L 443-456 This paragraph do not provides useful information and is very far from the results of this study. Delete it.

Author’s response: We think that these discussions are useful to consider the multiple aspects of behaviors observed during agonistic interactions. We also rewrote the paragraph to make this more related to the present study.

Reviewer’s comment:

L 464-467 However, given that low-frequency calls were recorded even during the period before the resident started attacking the intruder in MF-VF pairs (this was not observed in the other groups), it would be reasonable to estimate that the majority of low-frequency calls were emitted by MF residents.

This not an argument. Avoid speculations. In rodents, a lot of information comes from olfactory channel. It is easy to propose that VF-intruders placed on territory of MF-residents, perceive the smell of adult females and start calling of fear before the first aggressive interaction. A few more opposite arguments can be advanced.

Author’s response: We made the statement moderate. We think that estimations/assumptions belong to the discussion and there is no digital separation between estimations (or rational speculations) and argument.

Reviewer’s comment:

L 473-476 However, it remains unknown why MF gerbils (or MF-VF pair) emitted more low-frequency calls compared with those in the other three pairs. We hypothesize that vocalizations are influenced by hormonal changes induced by parenting experiences in female Mongolian gerbils.

In the preceding paragraph, you write that it is impossible to establish, who is calling the low-frequency calls, MF or VF. However, here you attribute the calls to one of females of the pair. Please re-write to make the content consistent with the text above.

Author’s response: We rewrote that (L474).

Reviewer #2

Thank you for your kind suggestion to improve this manuscript.

As detailed below, we have addressed the reviewer’s reservation.

Reviewer’s comment: I think the descriptions of several behaviors that were scored should be more specific and detailed. We cannot define a target behavior with itself, for example, dig was defined by digging, and what is dig?.

Author’s response: We updated the descriptions in the table.

Submitted filename: Responses to reviewers.docx

Click here for additional data file.

20 Jul 2022

Vocalization during agonistic encounter in Mongolian gerbils: impact of sexual experience.

PONE-D-21-38093R2

Dear Dr. Yamamoto,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

George Vousden

Staff Editor

PLOS ONE

Additional Editor Comments (optional):

Please accept our apologies for the delay in processing this decision.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors addressed all my comments.

I suggest minor changes, consistent to previous corrections.

L. 259 Delete “and multi-harmonics”

L. 264 Replace “syllable” with “group”

L. 446 Please corrects the typo; replace “repertories” with “repertoires”

L. 481 Consider replacing "Acknowledgment" with "Acknowledgement"

Reviewer #2: no major recommendation, but Ethical Note need to be added the Ethical Inspection License No:XXX,if you have.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

25 Jul 2022

PONE-D-21-38093R2

Vocalization during agonistic encounter in Mongolian gerbils: impact of sexual experience.

Dear Dr. Yamamoto:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. George Vousden

Staff Editor

PLOS ONE