Flexible use of multimodal communicative strategies in adult chimpanzees (Pan troglodytes)

Flexible use of multimodal communicative strategies in adult chimpanzees (Pan troglodytes) | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Flexible use of multimodal communicative strategies in adult chimpanzees (Pan troglodytes) Angèle Lombrey, Adriana Luna Martinez, Nick Dannenmann, Katerina Harvati, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6901931/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 09 Sep, 2025 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract Human communication is remarkable for its flexibility, a trait largely reflected in its multimodal nature and shared to some extent with nonhuman primates. Although individual differences in social behaviour have known evolutionary implications, their role in shaping primate communication remains largely unexplored. This study adopts a multimodal framework to partition variation in chimpanzees’ use of multicomponent and multisensory communicative strategies into socio-environmental, between-individual, and within-individual sources. Results showed that research setting and signaller’s sex affected communicative expression. Importantly, we also detected consistent between-individual differences in both strategies, independent of age, sex, or setting. While only multicomponent signal use was predicted by behavioural context at the population level, only for the use of multisensory acts did individuals vary in how they adjusted to context. These findings reveal profound flexibility in chimpanzee communication, highlighting individual-specific patterns and supporting a gradual evolutionary pathway toward the complexity of human multimodal communication. Biological sciences/Evolution/Anthropology/Biological anthropology Biological sciences/Zoology/Animal behaviour behavioural reaction norms context individual variation multimodal communication multicomponent signalling multisensory signalling Pan troglodytes plasticity Figures Figure 1 Figure 2 Figure 3 Figure 4 1. Introduction Although the key features of human face-to-face communication have been studied for decades and the range of empirical approaches is becoming ever wider 1 , 2 , satisfactory answers concerning the evolutionary building blocks of language are yet to be found 3 – 5 . As our closest living relatives, great apes are a model of choice to address these questions with comparative research. A critical feature of human communication is its extraordinary plasticity, reflected in the flexible production and interpretation of signals based on the common ground shared by interaction partners 4 , 6 . This is partly reflected in linguistic politeness and other forms of audience design, through which interactants adjust their signalling to one another in order to reduce uncertainty, building and maintaining relationships, and adhering to social norms 7 , 8 . Although it may seem that this flexibility, such as adjustments made in vocabulary and grammatical structure, adds redundancy, leading to more cognitively demanding language production, redundancy actually allows the speakers to reinforce mutual understanding 9 . Great ape communication system is among the most flexible in the animal kingdom: similarly to humans, they are able to combine multiple signals at the same time and adjust their signal use to the social context, the attention state of the interlocutor and to past social interactions (gorillas ( Gorilla gorilla ) 10 ; orang-utans ( Pongo pygmaeus ) 11 , 12 ; chimpanzees ( Pan troglodytes ) 13 , 14 ; bonobos ( Pan paniscus ) 15 ). This is also reflected in their capacity to persist and elaborate their signalling when their social goal is not achieved 16 , 17 , and to flexibly dissociate signals and outcomes: a single (e.g. gestural) signal can be used to achieve several purposes, and multiple signals can lead to the same outcome 12 , 18 – 20 . The multimodal core of language serves as a key source of its flexibility 21 – 23 , hence, one major shortcoming of the comparative study of primate communication is that, for decades, it has been approached in a unimodal way 24 , 25 . For a long time, researchers have been opposing in two major camps: proponents of either the gesture-first or the vocal-first theory of language origin both argue that language evolved from one or the other of these two modalities of communication 26 , 27 . However, multimodality is essential in language expression and integration, as is evidenced by the McGurk effect: our understanding of speech comes from the simultaneous integration of the vocal (speech itself) and visual (in this case, facial) components, and different combinations (e.g. unmatched vocal and facial components) result in different speech understandings 28 – 30 . Considering the pervasiveness of multimodality in interactive language use (where speech is usually accompanied by facial expressions and gestures 31 ) and the multimodal production of signals by many non-human species 32 , a unimodal approach most likely results in the loss of important information on the question of language origins. While many non-human multimodal signals are fixed (vocalizations, including most speech sounds, cannot be expressed without their complementary facial expression), multimodality in signal combination (e.g. combinatory use of vocal and gestural signals) can be adapted to specific environments and enables highly flexible communication 20 , 33 . In addition, the unimodal approach of communication hinders reliable comparisons between the different modalities 24 . Thus, to gain a more holistic understanding of communicative interactions and enable better comparisons with human communication, a growing number of studies have focused on the combination of signal modalities 33 , 34 . Within this framework, one key goal of this study is to investigate the flexibility of multimodal communicative behaviour in chimpanzees. Although multimodal approaches have become more common, another critical problem with current comparative research is that variation in communicative behaviour is mostly studied on the population or species level, where individual variation is considered negligible and essentially treated as ‘noise’ which is assumed to disappear once sufficient data have been collected. However, between-individual variation in behavioural type, or individual differences in the average expression of behaviours (i.e. personality: consistent individual differences in social behaviour) has been shown to have biological and ecological importance, whether for population dynamics and survival or for social evolution 35 , 36 , including in great ape species 37 . Thus, variation in behavioural and communicative patterns found between populations may often be due to extreme behavioural expression in only a few individuals 34 . Moreover, natural selection acts directly on individual differences, potentially leading to adaptive evolution 38 . As a consequence, individual variation is highly biologically relevant and must be taken into account in studies of human behavioural evolution 39 . Importantly, individuals may not only differ consistently in their average expression of behaviour, but they may also modify their behaviour when the social or environmental conditions change. This flexibility in behavioural expression (often called behavioural plasticity in behavioural ecology) is particularly characteristic of the primate order 40 . For instance, maternal style was found to be altered when new individuals were added to the social group of vervet monkeys ( Cercopithecus aethiops ) 41 . In addition to this component of within-individual variation, individuals may differ in the strength of such responses to environmental changes, that is, in the expression of behavioural plasticity (i.e. individual plasticity, or between-individual variation in the adjustment to environmental conditions): for instance, some individuals may adjust their behaviour to each environment they encounter, while others may not 42 , 43 . Although they have largely been studied separately, behavioural type and individual plasticity tend to be linked, and flexibility can only be fully understood when the individual level of variation is explicitly addressed. Dingemanse et al. (2010) 42 argue that studying these two phenomena together is essential to accurately interpret individual variation and to gain the most comprehensive insight into variability. Such an approach has already been applied to social and spatial behaviour in several species (e.g. Hertel et al. (2020) 44 used Generalized Linear Mixed Models to assess reversible and intrinsic individual variation in the spatial movements of African elephants), but very little is known about between-individual variation in behavioural type and individual plasticity in primate communicative behaviour 34 . Behavioural reaction norm (BRN) frameworks allow assessing the interaction between behavioural types and behavioural plasticity by plotting them along an “environmental” scale (e.g., social context): the intercept mirrors the individual’s behavioural type and the slope its behavioural plasticity. If the slope equals zero, the individual does not express context-driven variation in its behavioural response 34 , 42 – 44 . When multiple measures of the same individuals across contexts are available, a random regression analysis can be used to quantify between-individual variation in behavioural type, behavioural plasticity (i.e. the adjustment to environmental conditions), as well as the correlation between the two across individuals, (i.e. between-individual variation in behavioural plasticity). A multi-level study (including individuals of two different species in two different research settings) using this method on the infant-directed communication of orangutan mothers recently provided the first evidence for individual plasticity in primate communication 34 . This proof-of-concept study showed significant differences in the modification of mothers’ behaviour across contexts, with some showing greater responsiveness to their infants than others, suggesting that the degree of behavioural plasticity varies among individuals. Here, we aim to build on this preliminary work by examining individual variation and plasticity in the communicative behaviour of chimpanzees in two different research settings. Chimpanzees, our closest living relatives along with bonobos, have been extensively studied for their communicative repertoire across modalities (e.g. 13 , 33 , 45 – 47 ), providing a solid basis for more in-depth research on the communicative strategies of great apes. Vocalizations, gestures and facial expressions all play major roles in chimpanzee communication, but they are often not used for the same social goals, and particular signals can be more or less tightly linked to specific outcomes. Ten different context-specific vocalizations are expressed in the wild 48 ; the gesture “leaf-clip” is used only to acquire sexual attention, while “grab” and “pull” are used toward two or more outcomes 47 . The multimodal use of facial expressions and vocalizations can also be expressed flexibly by chimpanzees, with facial expressions not necessarily being paired with their typically associated vocalisations, which can also lead to altered meaning 49 . Furthermore, chimpanzees adapt their communicative strategies to the attentional state of the recipient: silent gestures are preferably chosen when the signaller is in the recipient’s visual field, whereas physical contact is preferred when the signaller is out of sight. This directed, purposeful use of gestures reflects their intentional nature 14 . The principal goal of this study was to examine the sources of variation in the communicative behaviour of adult zoo- and sanctuary-housed chimpanzees living in multiple social groups through a multimodal approach. To evaluate both the production and perception of multimodal communication, this study distinguished two different facets of it, building on Fröhlich et al. (2021) 11 . Multicomponent (MC ) communicative acts involve multiple signals (of the same or different signal types, i.e. vocalizations, bodily gestures, manual gestures, facial expressions) produced simultaneously by the sender. Multisensory (MS) communicative acts are those that are received and processed through several sensory channels (e.g., auditory, visual, tactile). This study had four core objectives. The first aim was to explore the specific types of MC and MS acts used by the study subjects, given how little is known about the role of multimodality in the communicative system of chimpanzees (but see 33 , 50 , 51 ). Second, we examined the extent to which individual (e.g. sex, age) and environmental (e.g. research setting, social group) factors influence both MC and MS signal use. We hypothesized that socio-demographic parameters, especially sex and research setting, determine whether these two communicative strategies are expressed, and predicted that sanctuary-housed and female individuals likely combined signals and sensory channels more often than their zoo-housed and male counterparts. Indeed, the more restricted space available in the zoos compared to sanctuaries induces higher levels of proximity. As a consequence, zoo-housed chimpanzees experience reduced social uncertainty and have less need to resort to complex communicative strategies 52 . Similarly, females were expected to use more MC and MS acts than males, reflecting the differential sociality level and social challenges between the sexes 53 , 54 : chimpanzee males are known to form strong intrasexual social bonds, involving intense cooperation and affiliative behaviour 55 , 56 , which is thought to not be the case to the same extent in females 57 , who would then need to compensate for reduced familiarity levels through communicative strategies. Individuals from different social groups might also express variation in their communicative signal use, as differences in size and composition of groups (e.g. sex ratio, presence and availability of juveniles) might impose different social constraints on their members (e.g. affiliation rates 58 , feeding tolerance 59 ). Third, we investigated whether individual chimpanzees differed in the expression of these strategies on average, regardless of possible socio-demographic differences and other confounding factors (e.g. sex, social context), given that consistent and repeatable differences in social behaviour over time have already been identified in both captive and wild chimpanzees 37 , 60 . These individual differences in sociality are likely reflected in the chimpanzees’ communicative behaviour as well. Finally, we explored the possibility that chimpanzees might shift their communicative strategies between behavioural contexts, and also differ individually in how they do so, using the behavioural reaction norm framework introduced above. Since chimpanzees are known to dissociate signals and outcomes at the population level 17 , 61 , 62 , it can be assumed that individuals flexibly adjust their communicative strategy to the context in which the signaller was involved prior to the initiation of the interaction (i.e. social vs non-social contexts). Specifically, some individuals may combine signals more often if they are not already engaged in a social activity prior to signalling, e.g. due to different prior interactional experiences. Alternatively, it is also possible that we do not find any evidence for individual plasticity in the studied contexts, as shifts of communicative behaviour in the same direction may be highly adaptive across all individuals in a certain group or setting. In this case, we would expect socio-environmental (e.g. setting) or demographic (e.g. sex) effects to override individual-level effects. Our results showed that the signaller’s research setting and sex, but not social group, affected the multimodal use of the communicative repertoire by chimpanzees, but also that chimpanzees expressed consistent differences in behavioural type for both MC and MS communicative strategies. Moreover, plasticity in relation to the behavioural context was identified for both communicative strategies, but individuals differed in their expressed plasticity only in their use of MS acts. These findings demonstrate the importance of considering individual-level variation in the investigation of great ape communicative behaviour. 2. Materials and methods a) Data collection Data on zoo-housed chimpanzees were collected in two social groups at Zoo Leipzig, Germany, from January to March 2022 (at that time, group A comprised 21 individuals, including 13 adults; group B comprised 6 adults) and one social group at Leintalzoo from May to June and from September to November 2022 (group L, 33 individuals, including 30 adults). Data on sanctuary-housed chimpanzees were collected at Chimfunshi Wildlife Sanctuary, Zambia, between July and August 2022 from two social groups: group C2 with 56 individuals, including 29 adults; and group C4 with 13 individuals, including 9 adults. The two sanctuary groups were kept in fenced enclosures of 65 ha and 25 ha, respectively. Each enclosure contained native fruit groves, grasslands, and densely forested Miombo woodland, as well as an indoor area for feeding and medical check-ups 63 . Only individuals over the age of 15 years were included in this study to ensure the examination of mature communicative repertoires 37 . However, subjects under 15 who exhibited adult-like behaviours, such as actively participating in reciprocated grooming with adults and adolescent females with infants, were also included (one additional 13-year-old in group L, two 14-year-olds in group C2 and one 13-year-old in group C4; see electronic supplementary material, Tabs. S1 and S2 for detailed information on subjects and group composition). Dyadic social interactions were recorded using a high-definition camera (Panasonic HC-VXF 999) and an external directional microphone (Sennheiser MKE 600) following a 15-minute focal sampling scheme, which was complemented by ad libitum sampling at Chimfunshi, due to the subjects’ unpredictable availability for observation in the sanctuary enclosures. Only the social interactions (of all types) of the focal subject, both as signaller and recipient of communicative acts, and with all conspecifics (adults and juveniles) were recorded. Data collection was carried out during the zoo opening hours (from 9 am to 5 pm) for the zoo-housed groups, and mainly around feeding sessions (twice a day, 11:30 am to 1:30 pm and 1:30 pm to 3:30 pm) and availability for observation in the sanctuary groups. The enclosures at Chimfunshi were very large (see above) and chimpanzees roam freely in their dedicated area, which prevented systematic focal follows of the individuals. Therefore, data were collected on most days (five to seven days a week, from 8 am to 4 pm) when the sanctuary-housed chimpanzees were relatively close enough to the fences and the feeding area, resulting in good visibility conditions. The observation order of the individuals was randomized and ca. 900 hours (ca. 660 hours for the zoo setting and 200 hours for the sanctuary setting) of focal observations were conducted (see electronic supplementary material, Tab. S2, for detailed information on sample size per individual). Ethical approval for the study on captive chimpanzees was granted by the Chimfunshi Research Advisory Board (CRAB) as well as the research boards of the Zoo Leipzig and the Leintalzoo Schwaigern. b) Video coding procedure A total of 6925 (zoo: 3013, sanctuary: 3912) video-recorded intraspecific communicative acts were coded in BORIS v 7.13 64 using a 96-behaviour ethogram of communicative acts based on the repertoires reported in Hobaiter and Byrne (2011) 14 and Wilke et al. (2017) 33 , including vocalizations, manual and body gestures, and facial expressions. This study included communicative behaviours that were mechanically ineffective and emitted in a dyadic interaction by a signaller, presumably to elicit a behavioural response by the receiver 47 . This definition thus excluded mechanically effective acts involving physical force, for which no response is expected from the “recipient”. Since this study focused on dyadic interactions, signals that were presumably directed towards several recipients, such as most of the pant hoot vocalizations, or signals produced in ambiguous interactions (e.g. multi-party conflicts), were excluded from the dataset. Additionally, signals emitted to maintain a joint behaviour (e.g. during play) were not considered, as they were not expected to elicit a behavioural response from the recipient. Furthermore, even though the range of signallers was limited to mature individuals, no restriction was placed on the recipients of the signals which therefore included both other mature and immature individuals. In addition to the identities of the signaller and recipient, several descriptive modifiers were coded for each signal instance, based on a previous study on orangutan communication 11 : behavioural context of the signaller (i.e. whether the signaller was engaged in a social or a non-social activity, with the recipient of the signal or other individual(s), prior to the emission of the signal; see electronic supplementary material, Figure S1 for detailed information on the coded contexts), the sensory modalities through which the signal was likely perceived (i.e. auditory, visual, tactile or seismic), the combinatory use of the signal (whether or not it was simultaneously associated with another signal and if so of which type – namely bodily, manual, facial or vocal), the recipient’s attentional state, and the social goal of the signal. Eighteen social goals were identified, based on Hobaiter and Byrne (2014): (1) acquire object, (2) attend to specific location, (3) climb on me, (4) dominance, (5) affiliation/greeting, (6) move away, (7) mover closer, (8) sexual attention, (9) reposition body, (10) seek reassurance, (11) give reassurance, (12) start grooming, (13) start play, (14) stop action, (15) travel with me, (16) tolerance, (17) follow me – sex, (18) other 47 (see electronic supplementary material, Tab. S3 for details on the coding scheme and the modifiers). Inter-rater reliability was assessed using Cohen’s Kappa coefficient (K, 65 ) to ensure that coding was consistent between all three coders, which was the case (signal: K = 0.68; behavioural context: K = 0.80; signal combination: K = 0.87; social goal: K = 0.85; 50 videos – including 142 signal instances – were used for the zoo groups and 48 videos – including 130 signal instances – for the sanctuary groups). c) Statistics and reproducibility After excluding incomplete entries, the total dataset consisted of 6865 communicative acts available for descriptive analyses on multimodality and regression analyses on between- and within-individual differences in multimodal strategies. In line with previous work on signal complexity in great apes 11 , 52 , two communicative strategies were analysed, coded as binary variables: multicomponent acts (MC), which represent the instances where at least two signals co-occurred in the same communicative act, regardless of their behavioural type (gestural, facial, vocal; fixed combinations of signals of different communicative domains that necessarily co-occur, such as scream and scream face, were not included as a signal combination), and multisensory acts (MS), namely the perception of signals through at least two sensory channels (i.e. auditory, tactile, visual, seismic) within the same communicative act. Considering the data-intensive nature of the analyses 66 , those two categories were deliberately widely defined to ensure a sufficient amount of data points were included in each of them. The behavioural reaction norm framework used in behavioural ecology 42 , 43 was used to partition variation in communicative strategies into its between- and within-individual sources. This enabled to quantify (i) between-individual variation in behavioural type, namely in the average expression of the communicative behaviour, (ii) plasticity, as within-individual variation in communicative behaviour across contexts and (iii) between-individual variation in behavioural plasticity. First, the different types of MC and MS communicative acts were investigated in order to determine which specific signal types on the one hand and sensory channels on the other hand were most often combined. Specifically, the proportion of each combination was calculated at the individual level and compared between settings and sexes. Second, generalized linear mixed models were fitted with a binomial error structure and logit link function, using the R package lme4 (v. 1.1-35.1 67 ), to analyse between-individual variation in behavioural type, specifically in the expression of MC and MS communicative acts. These models were used to assess the effects of individual and socio-environmental parameters on the use of these communicative strategies. In all models were included the following fixed effects: signaller’s age (in years, covariate, range = 13–56, z-transformed), sex (2 levels: female, male), research setting (2 levels: zoo, sanctuary), the recipient’s attention state (2 levels: attending, not attending), and the behavioural context just before signal production (2 levels: social or non-social, based on whether the subject was in association with another individual prior to the emission of the signal). Signaller (82 levels), social goal (18 levels, see above) and recipient identity (128 levels) were included as random effects, allowing the mean behavioural expression (i.e. intercept) to vary among signallers, social goals and recipients. Social group was initially included as another random effect, but it was removed because it did not explain any variance. Repeatability was calculated for the two response variables (MC and MS) using the R package rptR (default settings, v. 0.9.22; 68 ) to estimate the amount of variation in the response variable explained by random effects (typically, signaller identity) or other grouping factors in the mixed models 69 . To ensure the datasets were representative of the chimpanzee’s communicative behaviour, only individuals that contributed more than two datapoints were included. Third, to test whether individuals differ in how they shift their communicative strategies across contexts (i.e. behavioural plasticity), models with random intercepts and random slopes were fitted, based on Hertel et al. (2020) 44 and Fröhlich et al. (2022) 34 . Specifically, in addition to the previously described model, another model was fitted, including the same fixed and random effects as described above and a random slope over an “environmental gradient” (here a behavioural context contrast: social vs non-social contexts). Two models were thus fitted for each communicative strategy. We compared these models to the models without a random slope using a likelihood ratio test (LRT) 70 . The strength of the models was insured by only including individuals that contributed more than two datapoints to each behavioural context. The absence of collinearity between predictor variables was confirmed using variance inflation factors (VIFs; 71 ) from models including only the fixed effects (max VIF = 1.19). To test whether signaller identity played a statistically significant role, we also compared the full models to a null model without the random intercept and slopes effects using an LRT. All statistical analyses were conducted using R v 4.3.1 72 . 3. Results MC acts comprised 17.3% (N = 1186) and MC acts comprised 42.7% (N = 2931) of the total number of 6865 coded communicative acts. On average, 15.6% (range 0–45.9%) of individuals’ communicative acts were multicomponent and 39.9% (range 9.4–68.0%) were multisensory. Most communicative acts were produced when the subjects were involved in non-social contexts (N = 4992; 72.7% of the dataset), and in the sanctuary groups (N = 3852; 56.1% of the dataset). Table S4 provides an overview of how MC and MS acts were broken down by social group and behavioural context. a) Types of expressed signal and sensory combinations In a first step, the specific types of signal and sensory combinations were explored. Due to the number of categories and the widely contrasting amount of data in each of them, advanced statistical tests could not be run, but a descriptive overview of the types of MC and MS acts observed is presented in Figs. 1 and 2 . Males and females did not seem to differ systematically in the specific types of MC (e.g. manual-vocal) acts used, although the different types seemed to be more evenly expressed by females than by males (Fig. 1 a). Similarly, there did not seem to be obvious differences between research settings in the frequency of each combination type (Fig. 1 b). Regardless of the sex/setting class, the proportion of combined signal use varied dramatically between individuals, ranging from 0 to up to 100% for some combinations. Within a single MC communicative act, individuals mostly combined only two signals, while the combination of more than two signals was rarer, with the exception of the relatively common Facial + Manual + Vocal combination (11.1% of the MC acts; on average 12.2% of the individual repertoires). The most frequent combinations were Manual + Vocal (24.7% of the MC acts; on average 17.4% of the individual samples) and Body + Vocal (14.5% of the MC acts; on average 10.9% of the individual samples). With regard to MS acts, we found that Tactile and Visual sensory channels were most commonly associated (Fig. 2 ; 97% of the MS acts; on average 95.6% of the individual samples), followed by Auditory and Visual sensory channels (1.3% of the MS acts; on average 1.7% of the individual samples). There appeared to be no major differences between the sexes (Fig. 2 a) or the setting (Fig. 2 b). b) Effects of individual and environmental variables on the expression of multimodal strategies Generalized linear mixed effects models (GLMMs) were used to assess the effects of individual and environmental parameters on the two response variables (proportion of MC and MS communicative acts). Age, sex, recipient’s attention state and behavioural context were included as fixed effects and signaller identity, recipient identity and social goal as random effects. For both response variables, setting had a significant effect: both the average expression of MC and of MS communicative acts were significantly higher in sanctuary-housed individuals compared to zoo-housed individuals (MC: estimate ± s.e.m. = 0.497 ± 0.176, p = 0.005; MS: 0.294 ± 0.132, p = 0.027). Additionally, the behavioural context had a significant effect on the average expression of MC, with individuals using more MC acts in social compared to non-social contexts (0.346 ± 0.124, p = 0.005), but only showed a positive trend on the expression of MS communicative acts (0.189 ± 0.097, p = 0.052). Males expressed less MC acts than females (-0.337 ± 0.162, p = 0.037), and more MS acts were expressed when the recipient was visually attentive (2.815 ± 0.132, p < 0.001). Age did not have an effect on any communicative strategy (MC: 0.066 ± 0.073, p = 0.369; MS: 0.077 ± 0.057, p = 0.179; see electronic supplementary material, Tab. S5, for detailed information on the model outputs). c) Individual variation in behavioural type for the multimodal strategies To investigate whether individual chimpanzees differed in their communicative strategies on average, the previously introduced GLMMs with random intercepts only were fitted. The full model explained behavioural variation significantly better than the respective null model (which excluded signaller identity as a random effect) for both response variables (MC: χ 2 = 64.551, p < 0.001, df = 1; MS: χ 2 = 72.808, p < 0.001, df = 1). This means that considering variation between individuals in the models improved their explanatory power. For both MC and MS, significant repeatability was found for signaller identity (MC: R = 0.025, s.e. = 0.008, 95% confidence interval [CI] = 0.008–0.037, p < 0.001; MS: R = 0.024, s.e. = 0.006, 95% CI = 0.012–0.035, p < 0.001), recipient identity (MC: R = 0.023, s.e. = 0.007, 95% CI = 0.006–0.036, p < 0.001; MS: R = 0.014, s.e. = 0.005, 95% CI = 0.004–0.023, p < 0.001) and to a larger extent for the social goal (MC: R = 0.156, s.e. = 0.058, 95% CI = 0.038–0.257, p < 0.001; MS: R = 0.176, s.e. = 0.051, 95% CI = 0.073–0.27, p < 0.001). This means that around 2.5% of the remaining variance in MC and MS acts after controlling for confounding effects of age, sex, research setting, and behavioural context could be attributed to differences between individuals. Specifically, some individuals consistently expressed more MS (Fig. 3 a) or MC (Fig. 3 b) acts compared to others, and these differences were not primarily caused by predictable differences between socio-environmental (research setting or behavioural context) or individual (age, sex) parameters. d) Individual variation in behavioural plasticity for two multimodal strategies To examine whether chimpanzees differ in how they shift communicative behaviour across social conditions, two models for each response variable (i.e. multicomponent and multisensory acts) were compared: one with a random intercept for signaller identity (see above), and one with a random intercept for signaller identity and a random slope for behavioural context within signaller identity. As previously mentioned, MC communicative acts were used more often when the individuals were involved in social than in non-social contexts (Table S5), but the model including the random slope did not add any explanatory power when compared to the simpler model (χ 2 = 1.295, p = 0.730, df = 3; Fig. 4 a,b). Thus, although individuals showed plasticity in their use of MC acts according to the behavioural context, they all did so in the same manner (Fig. 4 b). The comparison of the two models showed a better fit of the more complex model (including the random slope for behavioural context within signaller identity) only for the use of MS acts (χ 2 = 44.5055, p < 0.001, df = 3; Fig. 4 c,d). This indicates that not only do individuals adjust their use of MS communicative acts to the context of the interaction, they also do so in different ways. Most of the individuals expressed a higher proportion of MS acts when already engaged in social contexts (Table S5), others showed no plastic response, and some even expressed the opposite pattern (Fig. 4 b; see electronic supplementary material, Tabs. S5 and S6, for detailed information on the models estimates). 4. Discussion Multimodal approaches in studying primate communication are still severely underrepresented 73 , and the proximate factors impacting complex communicative strategies are still poorly understood. To address this research gap, the goal of this study was to partition variation in multimodal communicative strategies of zoo- and sanctuary-housed chimpanzees into its social, between- and within-individual sources. To do so, the effect of individual and environmental parameters on the expression on two behaviours, multicomponent and multisensory communicative acts, were examined. Individual socio-demographic parameters were expected to impact the use of MC and MS, particularly the sex and research setting. Nevertheless, individuals were expected to differ in their communicative strategies regardless of individual variables, namely in their behavioural types. Within-individual variation (plasticity) was expected for both communicative strategies, with individuals adjusting their use to the behavioural context. Finally, between- and within-individual variation were expected to interact, leading to variation in plasticity between individuals. Looking at the role of socio-demographic parameters in the multimodal signalling of chimpanzees, a significant effect of the research setting was found on the expression of both MC and MS communicative acts: sanctuary-housed individuals combined signals and sensory channels more frequently than the zoo-housed individuals, in line with the prediction. Although there are similarities between zoo and sanctuary environments (e.g. limited space, food provided at specific times of the day), there are also major differences. Sanctuary-housed chimpanzees live in far bigger enclosures than zoo-housed ones, allowing them to engage in more wild-typical behaviour, such as fission-fusion dynamics 74 . In contrast, zoo individuals are forced to stay in relatively close proximity to one another. This can have far-reaching consequences for the social dynamics within the groups. Social interactions come with their load of uncertainty, and thus aggression risk for the interaction partners. As a consequence, individuals may use a variety of strategies to limit ambiguity in order to regulate social relationships with group members 75 , and avoid misunderstandings or communicative failure. The MC and MS use of communicative signals may be highly beneficial as it allows for better detection and interpretation of messages. It might thus be particularly important to invest in such strategies, especially when interactional contexts can change very quickly: wild chimpanzees have been shown to express greater signal complexity in contexts of high social uncertainty 52 , and orangutans to rely more on multicomponent communicative acts in less predictable interaction outcomes and on multisensory communicative acts to improve effectiveness 11 . Because zoo-housed individuals spend more time in close proximity to all individuals in the group, they are more familiar with each other and probably need to rely less on complex signalling than sanctuary-housed individuals 11 , 52 . This is consistent with previous research showing that repetitions of vocal greeting signals are rarely used among strongly bonded individuals, but are frequently used among distant dyads 76 . Another explanation for this might be the differing physical and ecological environment between the two settings: the bigger enclosures of the sanctuary potentially increase the distance between the interaction partners and the richer vegetation in this setting causes the visual noise. These two features might lead sanctuary-housed chimpanzees to increase their use of MC and MS acts as a compensation strategy, to improve the chances of the communicative act to properly reach the recipient, as has been identified in frogs (e.g. Staurois parvus 77 ) . They might thus combine vocal and auditory signals with gestural and visual signals more often than zoo-housed chimpanzees. Females were found to use significantly more MC acts than males, but not more MS acts, which is partly in line with the predictions. Similarly to setting differences, the strong and cooperative male-male social bonds 53 , 55 , 56 might require less complex communicative strategies than female-female communication to successfully negotiate interactions, although the traditionally ascribed asocial nature of female chimpanzees has been heavily debated (e.g. 78 ). Moreover, males and females might signal to conspecifics of the opposite sex with widely different social goals. For example, females may express more submissive, reassurance-seeking signals (e.g. “greetings”) towards males than males do towards females 79 . These types of interactions pose a particular risk for the signallers, who are typically lower-ranking and have less fighting ability than their recipients. The use of MC acts might thus be more important for females than for their male counterparts. Although this was not investigated in detail, the social goal of the interaction appears to play an important role in the use of both MC and MS communicative strategies. It would be interesting to explore this effect further, especially its interactions with other parameters, such as signaller and recipient sexes, as previous research has shown that these factors influence gestural signalling in chimpanzees 80 . These findings may have intriguing implications for the study of evolutionary origins of politeness in human communication. Linguistic politeness in humans is thought of as a means to reduce social friction and soften threatening acts 8 , 81 . Namely, when interaction partners share a stronger social bond, meaning that they interact often and know each other well, they need to put less effort into their communicative interactions 6 . This can be understood as a reduced need to rely on complex communicative strategies such as signal and sensory combinations. This phenomenon is mirrored by our findings, including the differentiated use of both these strategies between zoo- and sanctuary-housed individuals. Similar processes thus seem to occur in the communication of human and non-human primates, including the potentially similar reliance on social norm that is characteristic of human social behaviour 82 . The behavioural context, namely the activity of the signaller prior to signalling, had a significant impact on the use of MC acts, but only a marginal one on that of MS acts. Surprisingly, individuals relied more on MC and MS acts when they were already engaged in a social activity with a conspecific (e.g. grooming or playing) than when they were not (e.g. resting or feeding). This seems counterintuitive, since one might expect that interactions without prior social engagement would involve less uncertainty. One possible explanation may be that, although the signaller was already socially engaged, their social partner might not necessarily be the intended recipient of the communicative act. Thus, switching interaction partners might require even more ambiguity-reducing strategies than initiating a social interaction from a non-social context. This might be particularly true here, considering that signals expressed with the affiliation/greeting social goal, and are thus likely to be directed at individuals outside of the current social activity, represent 30% of the datasets. While this factor was not considered in the present analyses, it presents an interesting avenue for future research. Nonetheless, the findings demonstrate that chimpanzees exhibit plasticity in their communicative behaviour, adjusting their expressions in response to changes in behavioural context. When examining the distribution of the types of signal and sensory channel combinations, considerable variation between individuals of the same research setting was observed. This was somewhat supported by the repeatability results, which showed that 2.5% of the variance was explained by consistent individual differences, regardless of the social or contextual environment. These values are relatively small compared to other repeatability scores found in chimpanzees and other species 37 , 83 , and even to a similar study in orangutans 34 . Most of the variation seemed to be explained by the social goal of the interaction (accounting for 16 to 18% of the remaining variance), but other factors, such as context or recipient-related variables (as well as interacting factors, as mentioned above; e.g. sex/social goal), could also contribute to these low values. This would not be surprising given the immense flexibility chimpanzees exhibit in their communicative behaviour, and the social goal of the interaction is a major source of flexibility (means-end dissociation; 12 , 18 – 20 ). Nevertheless, these results demonstrate that there are consistent individual differences in communicative behavioural types. To assess not only whether there was a setting-level contextual shift in communicative strategy, namely plasticity, but also between-individual variation in plasticity, a behavioural reaction norm framework was applied to the dataset. Contrary to our predictions, individuals varied in plasticity in their expression of MS acts but not MC acts, despite the apparent similarity between the two strategies. In particular, individuals seemed to adjust their use of MC acts to the behavioural context they were in prior to the emission of the signal (population-level behavioural plasticity, see above), but all of them did so in a similar way. Fröhlich et al. (2022) found variation in plasticity in orangutan mothers in two very different variables, which seems to suggest that such variation is more broadly present in orangutan communicative behaviour 34 . However, a reason for this could be that this study focused on mother-infant interactions and not on all available interaction types, as was done here. Their results might thus specifically reflect variation in maternal styles, which commonly occurs in primates (e.g. 41 , 84 ). Kin relationship and rank could not be controlled for in the present study due to the highly unbalanced data, but these, as well as an affiliation index, could have important effects on the strategies used by individuals in their communicative attempts. For example, lower-ranking chimpanzees tend to become more agitated in the presence of a higher-ranking group member, leading them to produce more vocalizations 85 , and modality combinations have been shown to be affected by dominance relationships and familiarity between individuals 11 , 86 . Another explanation could be that the need to appropriately adjust one’s use of MC acts across behavioural contexts is so critical that it prevents the expression of differences in plasticity. Indeed, the results presented here show that individuals do express differing behavioural types in both social and non-social contexts; there is simply no variation between individuals in the strength and direction of their behavioural adjustment when these contexts change. This probably means that there is a required amount of MC communicative acts that individuals must express when involved in a social context for the interaction to proceed smoothly. Some social interactions are risky for the signaller, since they might be followed by directed aggression from the recipient if the signaller does not assess the situation properly. As mentioned above, individuals might thus rely on strategies to limit ambiguity, but they might also have to follow strict social norms to improve their chances of success 82 , 87 , as has been found with the greeting behaviour of this species: vocal sequences answer to specific rules based on the hierarchical and social relationship between social partners 76 . In this case, individuals seem to rely more on MC acts in social contexts, but they probably also need to do so in a specific way to minimize the uncertainty of the interaction. Plasticity necessarily comes with a trade-off between the benefits an individual might gain from adjusting its behaviour to changing environments (e.g. fitness benefit to express the context-appropriate behaviour) and the costs of doing so (e.g. investing energy in an additional communicative effort). As a consequence, variation in plasticity among individuals may arise because of two phenomena: state- and frequency-dependency of the payoffs of behavioural plasticity 43 , 88 – 90 . The phenomenon at play here seems to be at a crossroads between these, in the sense that the benefits of plasticity depend on external factors, not in terms of frequency but rather in terms of behavioural context (~ state of the social environment ). Namely, individuals might regularly find themselves in situations (e.g. behavioural contexts) where the risks (e.g. aggression) are reduced by a specific behavioural change (e.g. an increase in MC act expression). In other words, if increasing one’s use of MC acts is necessary to avoid communicative failure, individuals who decrease their MC acts expression might be faced with serious consequences (e.g. aggression, exclusion from food resources). In this case, variation in behavioural plasticity among individuals is regulated by the intrinsic characteristics of the environment of the interaction, namely, the strength of the risk associated with such interaction: the greater the risk, the more standardized the behaviour, and the weaker the differences in individual plasticity. In contrast, chimpanzees were found to vary in how they individually shift their use of MS acts between contexts (individual-level behavioural plasticity). This suggests that, contrary to the use of MC acts, the use of MS communicative acts might not be an aspect of chimpanzees’ communicative behaviour that is particularly important for the success of communicative interactions, which results in less constraint on how individuals use them: MS acts might not need to be standardized. If the costs associated with the misuse of MS acts are low, individuals are not under strong selective pressure in their expression of MS acts, leaving room for individual differences in behavioural plasticity. The finding that chimpanzees did not adjust their behaviour to the behavioural context at the population-level supports this: MS acts were used equally in social and non-social contexts. This is also reflected in the frequent use of communicative acts that are both tactile and visual to the recipient, and are likely to be prevalent in interactions between familiar individuals, as tactile gestures require physical proximity and social tolerance. Because these interactions involve less risk, there is less pressure to always communicate in a standardized, unambiguous way. Although they seem closely related, both communicative strategies were not similarly impacted by context changes. Mathot et al. (2011) showed that the payoffs of plasticity are not affected in the same way in all behavioural traits by the frequency-dependency phenomena: some payoffs are negative-frequency dependent (e.g. the benefits of being plastic decrease when more individuals are plastic) and others are positive-frequency dependent (e.g. the benefits of being plastic increase with increasing number of plastic individuals) 90 . Similarly, different communicative strategies seem to be affected to a varying degree by social risks. This result is in agreement with the previous finding that multicomponent and multisensory communicative acts have different functions in orangutans 11 and shows that these analyses are highly variable-sensitive: plasticity and variation in plasticity might not be found in, or be adaptive for all communicative strategies. Conclusions One constitutive feature of language is the great flexibility that it requires in order to be properly adjusted to the situation in which the interactions take place, and this flexibility is partly achieved through multimodality. However, and despite the importance of the comparative approach in studies on language evolution, individual variation, and especially plasticity, have been neglected in studies on non-human communication systems. This study shows that individual chimpanzees vary in their use of multicomponent and multisensory communicative acts (i.e. individual variation in behavioural type), which are largely shaped by the setting they live in. Chimpanzees also showed some level of behavioural plasticity between behavioural contexts, but individual variation in plasticity was identified in only one of the two communicative strategies. Multimodality in chimpanzee communicative flexibility appears to play a similar role as it is in human language: communicative complexity is associated both with the familiarity between and the adjustments of the communicative behaviour (i.e. plasticity) to the interaction partner. This underscores the relevance of comparative communication studies in understanding the evolutionary origins of language. Declarations Acknowledgements We thank Barbara Struve (Leintalzoo Schwaigern), Hanna Petschauer and Daniel Hanus(Zoo Leipzig, MPI-EVA), Edwin van Leeuwen, Katja Liebal and Thalita Simões Calvi (Chimfunshi Wildlife Orphanage Trust) as well as the research staff and zoo keepers for their helpful collaboration during this study. We also thank the entire “Pathways to Language” group for support and help in making this study possible, and in particular Wytse Wilhelm for helpful advice on statistical analysis. Funding This research was supported by the Volkswagen Foundation (Freigeist Fellowship 9B138 awarded to MF) and the German Research Foundation (DFG FOR 2237: Words, Bones, Genes, Tools – Tracking Linguistic, Cultural and Biological Trajectories of the Human Past; PIs: Katerina Harvati and Gerhard Jäger). Data accessibility The datasets and R script that support the findings of this study have been deposited on GitHub: https://github.com/angelelombrey/ALombrey-multi-plasti-Data-and-code.git. Author contribution MF and AL designed the study. AL and ALM collected and coded the data. ND coded the data. AL analysed the data with critical help from UK and MF. AL and MF wrote the manuscript with inputs from ALM, UK and KH. Competing interest The authors declare no competing interests. References Schwarz, N. Cognition and Communication: Judgmental Biases, Research Methods, and the Logic of Conversation (Psychology, 2014). 10.4324/9781315805887 Heesen, R. & Fröhlich, M. Revisiting the human ‘interaction engine’: comparative approaches to social action coordination. Philos. Trans. R Soc. B Biol. Sci. 377 , 20210092 (2022). Christiansen, M. H. & Kirby, S. Language evolution: consensus and controversies. Trends Cogn. Sci. 7 , 300–307 (2003). Tomasello, M. Origins of human communication. MIT Press. 37 , 393 (2008). Fitch, W. T. Empirical approaches to the study of language evolution. Psychon Bull. Rev. 24 , 3–33 (2017). Clark, H. H. & Brennan, S. E. Grounding in communication. in Perspectives on socially shared cognition. (eds Resnick, L. B., Levine, J. M. & Teasley, S. D.) 127–149 (American Psychological Association, Washington, doi: 10.1037/10096-006 . (1991). Brown, L., Kim, S. U. & Kim, H. The embodied enactment of politeness metapragmatics. J. Politeness Res. 19 , 149–183 (2023). Brown, P. & Levinson, S. C. Politeness: Some Universals in Language Usage (Cambridge University Press, 1987). Ferreira, V. S. Is It Better to Give Than to Donate? Syntactic Flexibility in Language Production. J. Mem. Lang. 35 , 724–755 (1996). Poss, S. R., Kuhar, C., Stoinski, T. S. & Hopkins, W. D. Differential use of attentional and visual communicative signaling by orangutans (Pongo pygmaeus) and gorillas (Gorilla gorilla) in response to the attentional status of a human. Am. J. Primatol. 68 , 978–992 (2006). Fröhlich, M. et al. Multicomponent and multisensory communicative acts in orang-utans may serve different functions. Commun. Biol. 4 , 1–13 (2021). Liebal, K., Pika, S. & Tomasello, M. Gestural communication of orangutans ( Pongo pygmaeus ). Gesture 6 , 1–38 (2006). Leavens, D. A., Russell, J. L. & Hopkins, W. D. Multimodal communication by captive chimpanzees (Pan troglodytes). Anim. Cogn. 13 , 33–40 (2010). Hobaiter, C. & Byrne, R. W. The gestural repertoire of the wild chimpanzee. Anim. Cogn. 14 , 745–767 (2011). Genty, E., Clay, Z., Hobaiter, C. & Zuberbühler, K. Multi-Modal Use of a Socially Directed Call in Bonobos. PLoS ONE . 9 , e84738 (2014). Cartmill, E. A. & Byrne, R. W. Orangutans Modify Their Gestural Signaling According to Their Audience’s Comprehension. Curr. Biol. 17 , 1345–1348 (2007). Roberts, A. I., Vick, S. J. & Buchanan-Smith, H. M. Communicative intentions in wild chimpanzees: persistence and elaboration in gestural signalling. Anim. Cogn. 16 , 187–196 (2013). Tomasello, M., Call, J., Nagell, K., Olguin, R. & Carpenter, M. The learning and use of gestural signals by young chimpanzees: A trans-generational study. Primates 35 , 137–154 (1994). Pika, S., Liebal, K. & Tomasello, M. Gestural communication in young gorillas (Gorilla gorilla): Gestural repertoire, learning, and use. Am. J. Primatol. 60 , 95–111 (2003). Fröhlich, M. & Hobaiter, C. The development of gestural communication in great apes. Behav. Ecol. Sociobiol. 72 , 194 (2018). Kendon, A. Gesture: Visible Action as Utterance (Cambridge University Press, 2004). Perniss, P. Why We Should Study Multimodal Language. Front Psychol 9 , (2018). Holler, J. & Levinson, S. C. Multimodal Language Processing in Human Communication. Trends Cogn. Sci. 23 , 639–652 (2019). Slocombe, K. E., Waller, B. M. & Liebal, K. The language void: the need for multimodality in primate communication research. Anim. Behav. 81 , 919–924 (2011). Fröhlich, M., Sievers, C., Townsend, S. W., Gruber, T. & van Schaik, C. P. Multimodal communication and language origins: integrating gestures and vocalizations. Biol. Rev. 94 , 1809–1829 (2019). Fitch, W. T. The Evolution of Language (Cambridge University Press, 2010). Waller, B. M., Liebal, K., Burrows, A. M. & Slocombe Katie. E. How Can a Multimodal Approach to Primate Communication Help Us Understand the Evolution of Communication? Evol. Psychol. 11 , 538–549 (2013). Mcgurk, H. & Macdonald, J. Hearing lips and seeing voices. Nature 264 , 746–748 (1976). Macdonald, J. & McGurk, H. Visual influences on speech perception processes. Percept. Psychophys . 24 , 253–257 (1978). Tiippana, K. What is the McGurk effect? Front Psychol 5 , (2014). Ekman, P. & Friesen, W. V. The Repertoire of Nonverbal Behavior: Categories, Origins, Usage, and Coding. Semiotica 1 , 49–98 (1969). Hebets, E. A. & Papaj, D. R. Complex signal function: developing a framework of testable hypotheses. Behav. Ecol. Sociobiol. 57 , 197–214 (2005). Wilke, C. et al. Production of and responses to unimodal and multimodal signals in wild chimpanzees, Pan troglodytes schweinfurthii. Anim. Behav. 123 , 305–316 (2017). Fröhlich, M., van Schaik, C. P., van Noordwijk, M. A. & Knief, U. Individual variation and plasticity in the infant-directed communication of orang-utan mothers. Proc. R. Soc. B Biol. Sci. 289, 20220200 (2022). Wolf, M. & Weissing, F. J. Animal personalities: consequences for ecology and evolution. Trends Ecol. Evol. 27 , 452–461 (2012). Dingemanse, N. J. & Araya-Ajoy, Y. G. Interacting personalities: behavioural ecology meets quantitative genetics. Trends Ecol. Evol. 30 , 88–97 (2015). Tkaczynski, P. J. et al. Long-term repeatability in social behaviour suggests stable social phenotypes in wild chimpanzees. R Soc. Open. Sci. 7 , 200454 (2020). Dingemanse, N. J. & Dochtermann, N. A. Quantifying individual variation in behaviour: mixed-effect modelling approaches. J. Anim. Ecol. 82 , 39–54 (2013). Lombrey, A. & Fröhlich, M. Individual variation in animal communication: from species averages to unique voices. Biol Rev (in press). van Schaik, C. P. The costs and benefits of flexibility as an expression of behavioural plasticity: a primate perspective. Philos. Trans. R Soc. B Biol. Sci. 368 , 20120339 (2013). Fairbanks, L. A. Individual Differences in Maternal Style: Causes and Consequences for Mothers and offspring. in Advances in the Study of Behavior (eds Rosenblatt, J. S. & Snowdon, C. T.) vol. 25 579–611 (Academic, (1996). Dingemanse, N. J., Kazem, A. J. N., Réale, D. & Wright, J. Behavioural reaction norms: animal personality meets individual plasticity. Trends Ecol. Evol. 25 , 81–89 (2010). Dingemanse, N. J. & Wolf, M. Between-individual differences in behavioural plasticity within populations: causes and consequences. Anim. Behav. 85 , 1031–1039 (2013). Hertel, A. G., Niemelä, P. T., Dingemanse, N. J. & Mueller, T. A guide for studying among-individual behavioral variation from movement data in the wild. Mov. Ecol. 8 , 30 (2020). Roberts, A. I., Roberts, S. G. B. & Vick, S. J. The repertoire and intentionality of gestural communication in wild chimpanzees. Anim. Cogn. 17 , 317–336 (2014). Taglialatela, J. P. et al. Multimodal communication in chimpanzees. Am. J. Primatol. 77 , 1143–1148 (2015). Hobaiter, C. & Byrne, R. W. The Meanings of Chimpanzee Gestures. Curr. Biol. 24 , 1596–1600 (2014). Crockford, C. Why does the chimpanzee vocal repertoire remain poorly understood and what can be done about it? in The Chimpanzees of the Taï Forest (eds Boesch, C. et al.) 394–409 (Cambridge University Press, doi: 10.1017/9781108674218.025 . (2019). Davila-Ross, M., Jesus, G., Osborne, J. & Bard, K. A. Chimpanzees (Pan troglodytes) Produce the Same Types of ‘Laugh Faces’ when They Emit Laughter and when They Are Silent. PLOS ONE . 10 , e0127337 (2015). Hobaiter, C., Byrne, R. W. & Zuberbühler, K. Wild chimpanzees’ use of single and combined vocal and gestural signals. Behav. Ecol. Sociobiol. 71 , 96 (2017). Doherty, E., Davila-Ross, M. & Clay, Z. Multimodal communication development in semiwild chimpanzees. Anim. Behav. 201 , 175–190 (2023). Grampp, M. et al. Social uncertainty promotes signal complexity during approaches in wild chimpanzees (Pan troglodytes verus) and mangabeys (Cercocebus atys atys). R Soc. Open. Sci. 10 , 231073 (2023). de Waal, F. B. M. Sex differences in the formation of coalitions among chimpanzees. Ethol. Sociobiol. 5 , 239–255 (1984). Lehmann, J. & Boesch, C. Sexual Differences in Chimpanzee Sociality. Int. J. Primatol. 29 , 65–81 (2008). Morin, P. A. et al. Kin Selection, Social Structure, Gene Flow, and the Evolution of Chimpanzees. Science 265 , 1193–1201 (1994). Nishida, T. & Hosaka, K. Coalition strategies among adult male chimpanzees of the Mahale Mountains, Tanzania. in Great Ape Societies (eds McGrew, W., Marchant, L. & Nishida, T.) 114–134 (Cambridge University Press, Cambridge, (1996). Gilby, I. C. & Wrangham, R. W. Association patterns among wild chimpanzees (Pan troglodytes schweinfurthii) reflect sex differences in cooperation. Behav. Ecol. Sociobiol. 62 , 1831–1842 (2008). Neal Webb, S. J., Hau, J. & Schapiro, S. J. Does group size matter? Captive chimpanzee (Pan troglodytes) behavior as a function of group size and composition. Am. J. Primatol. 81 , e22947 (2019). DeTroy, S. E., Ross, C. T., Cronin, K. A., van Leeuwen, E. J. C. & Haun, D. B. M. Cofeeding tolerance in chimpanzees depends on group composition: A longitudinal study across four communities. iScience 24, (2021). Koski, S. E. Social personality traits in chimpanzees: temporal stability and structure of behaviourally assessed personality traits in three captive populations. Behav. Ecol. Sociobiol. 65 , 2161–2174 (2011). Liebal, K., Call, J. & Tomasello, M. Use of gesture sequences in chimpanzees. Am. J. Primatol. 64 , 377–396 (2004). Call, J. How Apes Use Gestures: The Issue of Flexibility. in Evolution of Communicative Flexibility: Complexity, Creativity, and Adaptability in Human and Animal Communication (eds. Oller, D. K. & Griebel, U.) 235–252MIT Press, Cambridge, (2008). van Leeuwen, E. J. C., Bruinstroop, B. M. C. & Haun, D. B. M. Early Trauma Leaves No Social Signature in Sanctuary-Housed Chimpanzees (Pan troglodytes). Animals 13 , 49 (2022). Friard, O. & Gamba, M. BORIS: a free, versatile open-source event‐logging software for video/audio coding and live observations. Methods Ecol. Evol. 7 , 1325–1330 (2016). Bakeman, R. & Quera, V. Sequential Analysis and Observational Methods for the Behavioral Sciences (Cambridge University Press, 2011). van de Pol, M. Quantifying individual variation in reaction norms: how study design affects the accuracy, precision and power of random regression models. Methods Ecol. Evol. 3 , 268–280 (2012). Bates, D. et al. lme4: Linear Mixed-Effects Models using ‘Eigen’ and S4. (2025). Stoffel, M. A., Nakagawa, S. & Schielzeth, H. rptR: repeatability estimation and variance decomposition by generalized linear mixed-effects models. Methods Ecol. Evol. 8 , 1639–1644 (2017). Nakagawa, S. & Schielzeth, H. Repeatability for Gaussian and non-Gaussian data: a practical guide for biologists. Biol. Rev. 85 , 935–956 (2010). Dobson, A. J. An Introduction to Generalized Linear Models (Chapman & Hall / CRC, 2002). Quinn, G. P. & Keough, M. J. Experimental Design and Data Analysis for Biologists (Cambridge University Press, 2002). R Core Team. R: A Language and Environment for Statistical Computing. Liebal, K., Slocombe, K. E. & Waller, B. M. The language void 10 years on: multimodal primate communication research is still uncommon. Ethol. Ecol. Evol. 34 , 274–287 (2022). van Leeuwen, E. J. C. Temporal stability of chimpanzee social culture. Biol. Lett. 17 , 20210031 (2021). Dal Pesco, F. & Fischer, J. On the evolution of baboon greeting rituals. Philos. Trans. R Soc. B Biol. Sci. 375 , 20190420 (2020). Luef, E. M. & Pika, S. Social relationships and greetings in wild chimpanzees (Pan troglodytes): use of signal combinations. Primates 60 , 507–515 (2019). Grafe, T. U. et al. Multimodal Communication in a Noisy Environment: A Case Study of the Bornean Rock Frog Staurois parvus. PLOS ONE . 7 , e37965 (2012). Lehmann, J. & Boesch, C. Sociality of the dispersing sex: the nature of social bonds in West African female chimpanzees, Pan troglodytes . Anim. Behav. 77 , 377–387 (2009). Bernstein, I. S. & Gordon, T. P. The Function of Aggression in Primate Societies: Uncontrolled aggression may threaten human survival, but aggression may be vital to the establishment and regulation of primate societies and sociality. Am. Sci. 62 , 304–311 (1974). Scott, N. M. Gesture Use by Chimpanzees (an troglodytes): Differences Between Sexes in Inter- and Intra-Sexual Interactions. Am. J. Primatol. 75 , 555–567 (2013). Lakoff, R. Linguistic Theory and the Real World. Lang. Learn. 25 , 309–338 (1975). von Rudolf, C. & Burkart, J. M. Schaik, C. P. Evolutionary precursors of social norms in chimpanzees: a new approach. Biol. Philos. 26 , 1–30 (2011). van. Bell, A. M., Hankison, S. J. & Laskowski, K. L. The repeatability of behaviour: a meta-analysis. Anim. Behav. 77 , 771–783 (2009). Verderane, M. P. & Izar, P. Maternal care styles in primates: considering a New World species. Psicol. USP . 30 , e190055 (2019). Goodall, J. The chimpanzees of Gombe: patterns of behaviour. No Title (1986). Fedurek, P. et al. The function of chimpanzee greeting calls is modulated by their acoustic variation. Anim. Behav. 174 , 279–289 (2021). Flack, J. C., Jeannotte, L. A. & de Waal, F. B. M. Play Signaling and the Perception of Social Rules by Juvenile Chimpanzees (Pan troglodytes). J. Comp. Psychol. 118 , 149–159 (2004). Wolf, M., van Doorn, G. S. & Weissing, F. J. Evolutionary emergence of responsive and unresponsive personalities. Proc. Natl. Acad. Sci. 105, 15825–15830 (2008). Van Gils, J. A. State-dependent Bayesian foraging on spatially autocorrelated food distributions. Oikos 119 , 237–244 (2010). Mathot, K. J. et al. Disentangling the roles of frequency-vs. state-dependence in generating individual differences in behavioural plasticity. Ecol. Lett. 14 , 1254–1262 (2011). Additional Declarations No competing interests reported. Supplementary Files ALmultiplasti230525ESM.docx Cite Share Download PDF Status: Published Journal Publication published 09 Sep, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 09 Jul, 2025 Reviews received at journal 28 Jun, 2025 Reviews received at journal 26 Jun, 2025 Reviewers agreed at journal 23 Jun, 2025 Reviewers agreed at journal 19 Jun, 2025 Reviewers invited by journal 19 Jun, 2025 Editor assigned by journal 19 Jun, 2025 Editor invited by journal 17 Jun, 2025 Submission checks completed at journal 16 Jun, 2025 First submitted to journal 16 Jun, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6901931","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":482770580,"identity":"dd47c0d1-174f-414c-9f5c-01f73a264a02","order_by":0,"name":"Angèle Lombrey","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABSklEQVRIie3RsWqDQBgH8E8EXS64lStN01ewBFIKkr6KIuhiuxQkQwZBuC55gEq69QWylW6C4GSVbgdxsBScOpilZGjTnsYh1XboVqj/4bjz7sfndwfQpctfTQbQB0DlVAHgq28+2u4J3xO1Ol+dMUrCOb8hQTlUBH4i0vzqMVMngKS9h7BYkUS7E0F8RpO0fyLdhhnYSpPgNLqU1QjQ/vxC9zyy1O5d4FwU5ej0OjdliI1WGWoZWCOA5KU15HuMLAJp5fZIgGQajTBHgqY4opa51jY1eScxI6xKb1OS5JWRjyaRqRmC5tSEI35NHEaSmcCI3yTH1OKxGuKqF24W68OSeDdhzqoII6zGepMMqPlUFFNlIB2cB7C2x4eLxIfiZZqeyUmQ48Iet9pnv1Re3HbOfXkFrFZP1o6Y7SzednekVhtdunTp8j/zCZTKdJG8StBhAAAAAElFTkSuQmCC","orcid":"","institution":"University of Tübingen","correspondingAuthor":true,"prefix":"","firstName":"Angèle","middleName":"","lastName":"Lombrey","suffix":""},{"id":482770581,"identity":"f0569861-514c-4fa3-b8c9-067fdbefe5d7","order_by":1,"name":"Adriana Luna Martinez","email":"","orcid":"","institution":"University of Tübingen","correspondingAuthor":false,"prefix":"","firstName":"Adriana","middleName":"Luna","lastName":"Martinez","suffix":""},{"id":482770582,"identity":"5212c036-4449-4d7d-a532-22392274e16f","order_by":2,"name":"Nick Dannenmann","email":"","orcid":"","institution":"University of Tübingen","correspondingAuthor":false,"prefix":"","firstName":"Nick","middleName":"","lastName":"Dannenmann","suffix":""},{"id":482770583,"identity":"b167ef15-5681-4192-8e74-1788d1ebd236","order_by":3,"name":"Katerina Harvati","email":"","orcid":"","institution":"University of Tübingen","correspondingAuthor":false,"prefix":"","firstName":"Katerina","middleName":"","lastName":"Harvati","suffix":""},{"id":482770584,"identity":"1b180421-2fb9-4a5c-8592-6ca0b8ab9545","order_by":4,"name":"Ulrich Knief","email":"","orcid":"","institution":"University of Freiburg","correspondingAuthor":false,"prefix":"","firstName":"Ulrich","middleName":"","lastName":"Knief","suffix":""},{"id":482770585,"identity":"8f257995-1f05-48a0-827e-37256499dfa4","order_by":5,"name":"Marlen Fröhlich","email":"","orcid":"","institution":"University of Tübingen","correspondingAuthor":false,"prefix":"","firstName":"Marlen","middleName":"","lastName":"Fröhlich","suffix":""}],"badges":[],"createdAt":"2025-06-16 05:53:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6901931/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6901931/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-14835-x","type":"published","date":"2025-09-09T15:56:57+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":87032353,"identity":"5ab43a4b-f2f3-4d6d-8c4d-8434270f7c6a","added_by":"auto","created_at":"2025-07-18 12:57:30","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":202410,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eMean proportions of multicomponent (MC) use for each signal combination. Panel (a) shows MC use across sexes and panel (b) across research settings. Indicated are individual means per MC type (circles) and population means (filled circles) with standard error (vertical lines).\u003c/em\u003e \u003cem\u003eColours represent the types of signal combination and circle size represents the sampling effort per individual (i.e. total number of MC acts recorded per individual).\u003c/em\u003e\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-6901931/v1/859d463e3ae547ba8dcc7c2c.png"},{"id":87030149,"identity":"4c5b234c-7de2-49d7-8b3e-d94a5903bcf1","added_by":"auto","created_at":"2025-07-18 12:41:30","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":148140,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eIndividual proportion of multisensory (MS) use for each sensory combination. Panel (a) represents MS use across sexes and panel (b) across settings. Indicated are individual means per MC type (circles) and population means (horizontal lines) with standard error (vertical lines). Colours represent the types of sensory channel combination and circle size represents the sampling effort per individual (namely the total number of MC acts recorded for each individual).\u003c/em\u003e\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-6901931/v1/ead9620c35986e11c0bcff98.png"},{"id":87030143,"identity":"d736fa21-4dff-4ddf-895c-2ac7b580fb21","added_by":"auto","created_at":"2025-07-18 12:41:30","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":29510,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eBetween-individual variation in multimodal behavioural types in adult chimpanzees. Panels show individual behavioural types for multicomponent (a) and multisensory (b) communicative acts. Plotted are individual random effect intercepts (best linear unbiased predictors, BLUPs) from models examining variation in the expression of multicomponent and multisensory communicative acts. Colours represent the different sex/setting categories.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-6901931/v1/6a54f47055e9ddd7b3cf4369.png"},{"id":87033388,"identity":"9b3db861-a79a-4cb5-84d8-d518dd87e604","added_by":"auto","created_at":"2025-07-18 13:05:30","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":43532,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cem\u003eIndividual shifts in chimpanzees’ expression of multimodal communication in social versus non-social contexts. Panels show shifts in the use of multicomponent (a and b) and multisensory (c and d) communicative acts. (a) and (c) show prediction lines assuming individual chimpanzees adjust their behaviour in the same way between behavioural contexts (random intercept). Because predictors were back-transformed, the lines are not strictly parallel as they are on the logit-scale (i.e. only the intercepts vary). (b) and (d) show prediction lines assuming individual chimpanzees differ in the extent to which they change their behaviour between behavioural contexts (random intercept and slope). The lack of differences between plots (a) and (b) shows that the random slope does not add significant explanatory power to the model. Colours represent different sex/setting categories.\u003c/em\u003e\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-6901931/v1/6b00d285e92beb42c69c9db0.png"},{"id":91359296,"identity":"1df60609-a976-448c-84ea-fcddbf1f6a50","added_by":"auto","created_at":"2025-09-15 16:05:39","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1093584,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6901931/v1/b1e5b7cf-8330-4edf-8d58-5115a7386b05.pdf"},{"id":87031178,"identity":"d85d0b0a-7426-4627-9069-11661d067d87","added_by":"auto","created_at":"2025-07-18 12:49:30","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":68767,"visible":true,"origin":"","legend":"","description":"","filename":"ALmultiplasti230525ESM.docx","url":"https://assets-eu.researchsquare.com/files/rs-6901931/v1/89a5d6037051b39f688eefaa.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Flexible use of multimodal communicative strategies in adult chimpanzees (Pan troglodytes)","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAlthough the key features of human face-to-face communication have been studied for decades and the range of empirical approaches is becoming ever wider \u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e, satisfactory answers concerning the evolutionary building blocks of language are yet to be found \u003csup\u003e\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. As our closest living relatives, great apes are a model of choice to address these questions with comparative research. A critical feature of human communication is its extraordinary plasticity, reflected in the flexible production and interpretation of signals based on the common ground shared by interaction partners \u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. This is partly reflected in linguistic politeness and other forms of audience design, through which interactants adjust their signalling to one another in order to reduce uncertainty, building and maintaining relationships, and adhering to social norms\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Although it may seem that this flexibility, such as adjustments made in vocabulary and grammatical structure, adds redundancy, leading to more cognitively demanding language production, redundancy actually allows the speakers to reinforce mutual understanding \u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e. Great ape communication system is among the most flexible in the animal kingdom: similarly to humans, they are able to combine multiple signals at the same time and adjust their signal use to the social context, the attention state of the interlocutor and to past social interactions (gorillas (\u003cem\u003eGorilla gorilla\u003c/em\u003e)\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e; orang-utans (\u003cem\u003ePongo pygmaeus\u003c/em\u003e)\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e; chimpanzees (\u003cem\u003ePan troglodytes\u003c/em\u003e)\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e; bonobos (\u003cem\u003ePan paniscus\u003c/em\u003e)\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e). This is also reflected in their capacity to persist and elaborate their signalling when their social goal is not achieved \u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, and to flexibly dissociate signals and outcomes: a single (e.g. gestural) signal can be used to achieve several purposes, and multiple signals can lead to the same outcome \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan additionalcitationids=\"CR19\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe multimodal core of language serves as a key source of its flexibility \u003csup\u003e\u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e, hence, one major shortcoming of the comparative study of primate communication is that, for decades, it has been approached in a unimodal way \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e. For a long time, researchers have been opposing in two major camps: proponents of either the gesture-first or the vocal-first theory of language origin both argue that language evolved from one or the other of these two modalities of communication \u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. However, multimodality is essential in language expression and integration, as is evidenced by the McGurk effect: our understanding of speech comes from the simultaneous integration of the vocal (speech itself) and visual (in this case, facial) components, and different combinations (e.g. unmatched vocal and facial components) result in different speech understandings \u003csup\u003e\u003cspan additionalcitationids=\"CR29\" citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e. Considering the pervasiveness of multimodality in interactive language use (where speech is usually accompanied by facial expressions and gestures\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e) and the multimodal production of signals by many non-human species \u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e, a unimodal approach most likely results in the loss of important information on the question of language origins. While many non-human multimodal signals are fixed (vocalizations, including most speech sounds, cannot be expressed without their complementary facial expression), multimodality in signal combination (e.g. combinatory use of vocal and gestural signals) can be adapted to specific environments and enables highly flexible communication \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. In addition, the unimodal approach of communication hinders reliable comparisons between the different modalities \u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u003c/sup\u003e. Thus, to gain a more holistic understanding of communicative interactions and enable better comparisons with human communication, a growing number of studies have focused on the combination of signal modalities \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Within this framework, one key goal of this study is to investigate the flexibility of multimodal communicative behaviour in chimpanzees.\u003c/p\u003e\u003cp\u003eAlthough multimodal approaches have become more common, another critical problem with current comparative research is that variation in communicative behaviour is mostly studied on the population or species level, where individual variation is considered negligible and essentially treated as \u0026lsquo;noise\u0026rsquo; which is assumed to disappear once sufficient data have been collected. However, between-individual variation in behavioural type, or individual differences in the average expression of behaviours (i.e. personality: \u003cem\u003econsistent individual differences in social behaviour)\u003c/em\u003e has been shown to have biological and ecological importance, whether for population dynamics and survival or for social evolution \u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e, including in great ape species \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. Thus, variation in behavioural and communicative patterns found between populations may often be due to extreme behavioural expression in only a few individuals \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Moreover, natural selection acts directly on individual differences, potentially leading to adaptive evolution \u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. As a consequence, individual variation is highly biologically relevant and must be taken into account in studies of human behavioural evolution\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eImportantly, individuals may not only differ consistently in their average expression of behaviour, but they may also modify their behaviour when the social or environmental conditions change. This flexibility in behavioural expression (often called behavioural plasticity in behavioural ecology) is particularly characteristic of the primate order \u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. For instance, maternal style was found to be altered when new individuals were added to the social group of vervet monkeys (\u003cem\u003eCercopithecus aethiops\u003c/em\u003e) \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e. In addition to this component of within-individual variation, individuals may differ in the strength of such responses to environmental changes, that is, in the expression of behavioural plasticity (i.e. individual plasticity, or between-individual variation in the adjustment to environmental conditions): for instance, some individuals may adjust their behaviour to each environment they encounter, while others may not \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Although they have largely been studied separately, behavioural type and individual plasticity tend to be linked, and flexibility can only be fully understood when the individual level of variation is explicitly addressed. Dingemanse et al. (2010) \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e argue that studying these two phenomena together is essential to accurately interpret individual variation and to gain the most comprehensive insight into variability. Such an approach has already been applied to social and spatial behaviour in several species (e.g. Hertel et al. (2020) \u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e used Generalized Linear Mixed Models to assess reversible and intrinsic individual variation in the spatial movements of African elephants), but very little is known about between-individual variation in behavioural type and individual plasticity in primate communicative behaviour \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eBehavioural reaction norm (BRN) frameworks allow assessing the interaction between behavioural types and behavioural plasticity by plotting them along an \u0026ldquo;environmental\u0026rdquo; scale (e.g., social context): the intercept mirrors the individual\u0026rsquo;s behavioural type and the slope its behavioural plasticity. If the slope equals zero, the individual does not express context-driven variation in its behavioural response \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e,\u003cspan additionalcitationids=\"CR43\" citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. When multiple measures of the same individuals across contexts are available, a random regression analysis can be used to quantify between-individual variation in behavioural type, behavioural plasticity (i.e. the adjustment to environmental conditions), as well as the correlation between the two across individuals, (i.e. between-individual variation in behavioural plasticity). A multi-level study (including individuals of two different species in two different research settings) using this method on the infant-directed communication of orangutan mothers recently provided the first evidence for individual plasticity in primate communication\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. This proof-of-concept study showed significant differences in the modification of mothers\u0026rsquo; behaviour across contexts, with some showing greater responsiveness to their infants than others, suggesting that the degree of behavioural plasticity varies among individuals.\u003c/p\u003e\u003cp\u003eHere, we aim to build on this preliminary work by examining individual variation and plasticity in the communicative behaviour of chimpanzees in two different research settings. Chimpanzees, our closest living relatives along with bonobos, have been extensively studied for their communicative repertoire across modalities (e.g. \u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan additionalcitationids=\"CR46\" citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e), providing a solid basis for more in-depth research on the communicative strategies of great apes. Vocalizations, gestures and facial expressions all play major roles in chimpanzee communication, but they are often not used for the same social goals, and particular signals can be more or less tightly linked to specific outcomes. Ten different context-specific vocalizations are expressed in the wild \u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e; the gesture \u0026ldquo;leaf-clip\u0026rdquo; is used only to acquire sexual attention, while \u0026ldquo;grab\u0026rdquo; and \u0026ldquo;pull\u0026rdquo; are used toward two or more outcomes\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. The multimodal use of facial expressions and vocalizations can also be expressed flexibly by chimpanzees, with facial expressions not necessarily being paired with their typically associated vocalisations, which can also lead to altered meaning\u003csup\u003e\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. Furthermore, chimpanzees adapt their communicative strategies to the attentional state of the recipient: silent gestures are preferably chosen when the signaller is in the recipient\u0026rsquo;s visual field, whereas physical contact is preferred when the signaller is out of sight. This directed, purposeful use of gestures reflects their intentional nature \u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe principal goal of this study was to examine the sources of variation in the communicative behaviour of adult zoo- and sanctuary-housed chimpanzees living in multiple social groups through a multimodal approach. To evaluate both the production and perception of multimodal communication, this study distinguished two different facets of it, building on Fr\u0026ouml;hlich et al. (2021)\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. \u003cem\u003eMulticomponent (MC\u003c/em\u003e) communicative acts involve multiple signals (of the same or different signal types, i.e. vocalizations, bodily gestures, manual gestures, facial expressions) produced simultaneously by the sender. \u003cem\u003eMultisensory (MS)\u003c/em\u003e communicative acts are those that are received and processed through several sensory channels (e.g., auditory, visual, tactile).\u003c/p\u003e\u003cp\u003eThis study had four core objectives. The first aim was to explore the specific types of MC and MS acts used by the study subjects, given how little is known about the role of multimodality in the communicative system of chimpanzees (but see \u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e,\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u003c/sup\u003e). Second, we examined the extent to which individual (e.g. sex, age) and environmental (e.g. research setting, social group) factors influence both MC and MS signal use. We hypothesized that socio-demographic parameters, especially sex and research setting, determine whether these two communicative strategies are expressed, and predicted that sanctuary-housed and female individuals likely combined signals and sensory channels more often than their zoo-housed and male counterparts. Indeed, the more restricted space available in the zoos compared to sanctuaries induces higher levels of proximity. As a consequence, zoo-housed chimpanzees experience reduced social uncertainty and have less need to resort to complex communicative strategies \u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. Similarly, females were expected to use more MC and MS acts than males, reflecting the differential sociality level and social challenges between the sexes \u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e,\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e: chimpanzee males are known to form strong intrasexual social bonds, involving intense cooperation and affiliative behaviour \u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e, which is thought to not be the case to the same extent in females \u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e, who would then need to compensate for reduced familiarity levels through communicative strategies. Individuals from different social groups might also express variation in their communicative signal use, as differences in size and composition of groups (e.g. sex ratio, presence and availability of juveniles) might impose different social constraints on their members (e.g. affiliation rates\u003csup\u003e\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e\u003c/sup\u003e, feeding tolerance\u003csup\u003e\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e\u003c/sup\u003e). Third, we investigated whether individual chimpanzees differed in the expression of these strategies on average, regardless of possible socio-demographic differences and other confounding factors (e.g. sex, social context), given that consistent and repeatable differences in social behaviour over time have already been identified in both captive and wild chimpanzees \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e\u003c/sup\u003e. These individual differences in sociality are likely reflected in the chimpanzees\u0026rsquo; communicative behaviour as well. Finally, we explored the possibility that chimpanzees might shift their communicative strategies between behavioural contexts, and also differ individually in how they do so, using the behavioural reaction norm framework introduced above. Since chimpanzees are known to dissociate signals and outcomes at the \u003cem\u003epopulation level\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e,\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e, it can be assumed that \u003cem\u003eindividuals\u003c/em\u003e flexibly adjust their communicative strategy to the context in which the signaller was involved prior to the initiation of the interaction (i.e. social vs non-social contexts). Specifically, some individuals may combine signals more often if they are not already engaged in a social activity prior to signalling, e.g. due to different prior interactional experiences. Alternatively, it is also possible that we do not find any evidence for individual plasticity in the studied contexts, as shifts of communicative behaviour in the same direction may be highly adaptive across all individuals in a certain group or setting. In this case, we would expect socio-environmental (e.g. setting) or demographic (e.g. sex) effects to override individual-level effects.\u003c/p\u003e\u003cp\u003eOur results showed that the signaller\u0026rsquo;s research setting and sex, but not social group, affected the multimodal use of the communicative repertoire by chimpanzees, but also that chimpanzees expressed consistent differences in behavioural type for both MC and MS communicative strategies. Moreover, plasticity in relation to the behavioural context was identified for both communicative strategies, but individuals differed in their expressed plasticity only in their use of MS acts. These findings demonstrate the importance of considering individual-level variation in the investigation of great ape communicative behaviour.\u003c/p\u003e"},{"header":"2. Materials and methods","content":"\u003cp\u003ea) \u003cem\u003eData collection\u003c/em\u003e\u003c/p\u003e\u003cp\u003eData on zoo-housed chimpanzees were collected in two social groups at Zoo Leipzig, Germany, from January to March 2022 (at that time, group A comprised 21 individuals, including 13 adults; group B comprised 6 adults) and one social group at Leintalzoo from May to June and from September to November 2022 (group L, 33 individuals, including 30 adults). Data on sanctuary-housed chimpanzees were collected at Chimfunshi Wildlife Sanctuary, Zambia, between July and August 2022 from two social groups: group C2 with 56 individuals, including 29 adults; and group C4 with 13 individuals, including 9 adults. The two sanctuary groups were kept in fenced enclosures of 65 ha and 25 ha, respectively. Each enclosure contained native fruit groves, grasslands, and densely forested Miombo woodland, as well as an indoor area for feeding and medical check-ups \u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e. Only individuals over the age of 15 years were included in this study to ensure the examination of mature communicative repertoires \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. However, subjects under 15 who exhibited adult-like behaviours, such as actively participating in reciprocated grooming with adults and adolescent females with infants, were also included (one additional 13-year-old in group L, two 14-year-olds in group C2 and one 13-year-old in group C4; see electronic supplementary material, Tabs. S1 and S2 for detailed information on subjects and group composition).\u003c/p\u003e\u003cp\u003eDyadic social interactions were recorded using a high-definition camera (Panasonic HC-VXF 999) and an external directional microphone (Sennheiser MKE 600) following a 15-minute focal sampling scheme, which was complemented by \u003cem\u003ead libitum\u003c/em\u003e sampling at Chimfunshi, due to the subjects\u0026rsquo; unpredictable availability for observation in the sanctuary enclosures. Only the social interactions (of all types) of the focal subject, both as signaller and recipient of communicative acts, and with all conspecifics (adults and juveniles) were recorded. Data collection was carried out during the zoo opening hours (from 9 am to 5 pm) for the zoo-housed groups, and mainly around feeding sessions (twice a day, 11:30 am to 1:30 pm and 1:30 pm to 3:30 pm) and availability for observation in the sanctuary groups. The enclosures at Chimfunshi were very large (see above) and chimpanzees roam freely in their dedicated area, which prevented systematic focal follows of the individuals. Therefore, data were collected on most days (five to seven days a week, from 8 am to 4 pm) when the sanctuary-housed chimpanzees were relatively close enough to the fences and the feeding area, resulting in good visibility conditions. The observation order of the individuals was randomized and ca. 900 hours (ca. 660 hours for the zoo setting and 200 hours for the sanctuary setting) of focal observations were conducted (see electronic supplementary material, Tab. S2, for detailed information on sample size per individual).\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eEthical approval\u003c/strong\u003e\u003cp\u003e for the study on captive chimpanzees was granted by the Chimfunshi Research Advisory Board (CRAB) as well as the research boards of the Zoo Leipzig and the Leintalzoo Schwaigern.\u003c/p\u003e\u003c/p\u003e\u003cp\u003eb) \u003cem\u003eVideo coding procedure\u003c/em\u003e\u003c/p\u003e\u003cp\u003eA total of 6925 (zoo: 3013, sanctuary: 3912) video-recorded intraspecific communicative acts were coded in BORIS v 7.13 \u003csup\u003e64\u003c/sup\u003e using a 96-behaviour ethogram of communicative acts based on the repertoires reported in Hobaiter and Byrne (2011)\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e and Wilke et al. (2017)\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e, including vocalizations, manual and body gestures, and facial expressions. This study included communicative behaviours that were mechanically ineffective and emitted in a dyadic interaction by a signaller, presumably to elicit a behavioural response by the receiver \u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e. This definition thus excluded mechanically effective acts involving physical force, for which no response is expected from the \u0026ldquo;recipient\u0026rdquo;. Since this study focused on dyadic interactions, signals that were presumably directed towards several recipients, such as most of the pant hoot vocalizations, or signals produced in ambiguous interactions (e.g. multi-party conflicts), were excluded from the dataset. Additionally, signals emitted to maintain a joint behaviour (e.g. during play) were not considered, as they were not expected to elicit a behavioural response from the recipient. Furthermore, even though the range of signallers was limited to mature individuals, no restriction was placed on the recipients of the signals which therefore included both other mature and immature individuals.\u003c/p\u003e\u003cp\u003eIn addition to the identities of the signaller and recipient, several descriptive modifiers were coded for each signal instance, based on a previous study on orangutan communication \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e: behavioural context of the signaller (i.e. whether the signaller was engaged in a social or a non-social activity, with the recipient of the signal or other individual(s), prior to the emission of the signal; see electronic supplementary material, Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e for detailed information on the coded contexts), the sensory modalities through which the signal was likely perceived (i.e. auditory, visual, tactile or seismic), the combinatory use of the signal (whether or not it was simultaneously associated with another signal and if so of which type \u0026ndash; namely bodily, manual, facial or vocal), the recipient\u0026rsquo;s attentional state, and the social goal of the signal. Eighteen social goals were identified, based on Hobaiter and Byrne (2014): (1) acquire object, (2) attend to specific location, (3) climb on me, (4) dominance, (5) affiliation/greeting, (6) move away, (7) mover closer, (8) sexual attention, (9) reposition body, (10) seek reassurance, (11) give reassurance, (12) start grooming, (13) start play, (14) stop action, (15) travel with me, (16) tolerance, (17) follow me \u0026ndash; sex, (18) other\u003csup\u003e\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e (see electronic supplementary material, Tab. S3 for details on the coding scheme and the modifiers).\u003c/p\u003e\u003cp\u003eInter-rater reliability was assessed using Cohen\u0026rsquo;s Kappa coefficient (K, \u003csup\u003e\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e) to ensure that coding was consistent between all three coders, which was the case (signal: K\u0026thinsp;=\u0026thinsp;0.68; behavioural context: K\u0026thinsp;=\u0026thinsp;0.80; signal combination: K\u0026thinsp;=\u0026thinsp;0.87; social goal: K\u0026thinsp;=\u0026thinsp;0.85; 50 videos \u0026ndash; including 142 signal instances \u0026ndash; were used for the zoo groups and 48 videos \u0026ndash; including 130 signal instances \u0026ndash; for the sanctuary groups).\u003c/p\u003e\u003cp\u003ec) \u003cem\u003eStatistics and reproducibility\u003c/em\u003e\u003c/p\u003e\u003cp\u003eAfter excluding incomplete entries, the total dataset consisted of 6865 communicative acts available for descriptive analyses on multimodality and regression analyses on between- and within-individual differences in multimodal strategies.\u003c/p\u003e\u003cp\u003eIn line with previous work on signal complexity in great apes \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e, two communicative strategies were analysed, coded as binary variables: multicomponent acts (MC), which represent the instances where at least two signals co-occurred in the same communicative act, regardless of their behavioural type (gestural, facial, vocal; fixed combinations of signals of different communicative domains that necessarily co-occur, such as scream and scream face, were not included as a signal combination), and multisensory acts (MS), namely the perception of signals through at least two sensory channels (i.e. auditory, tactile, visual, seismic) within the same communicative act. Considering the data-intensive nature of the analyses \u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e, those two categories were deliberately widely defined to ensure a sufficient amount of data points were included in each of them.\u003c/p\u003e\u003cp\u003eThe behavioural reaction norm framework used in behavioural ecology \u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e was used to partition variation in communicative strategies into its between- and within-individual sources. This enabled to quantify (i) between-individual variation in behavioural type, namely in the average expression of the communicative behaviour, (ii) plasticity, as within-individual variation in communicative behaviour across contexts and (iii) between-individual variation in behavioural plasticity.\u003c/p\u003e\u003cp\u003eFirst, the different types of MC and MS communicative acts were investigated in order to determine which specific signal types on the one hand and sensory channels on the other hand were most often combined. Specifically, the proportion of each combination was calculated at the individual level and compared between settings and sexes.\u003c/p\u003e\u003cp\u003eSecond, generalized linear mixed models were fitted with a binomial error structure and logit link function, using the R package lme4 (v. 1.1-35.1\u003csup\u003e67\u003c/sup\u003e), to analyse between-individual variation in behavioural type, specifically in the expression of MC and MS communicative acts. These models were used to assess the effects of individual and socio-environmental parameters on the use of these communicative strategies. In all models were included the following fixed effects: signaller\u0026rsquo;s age (in years, covariate, range\u0026thinsp;=\u0026thinsp;13\u0026ndash;56, z-transformed), sex (2 levels: female, male), research setting (2 levels: zoo, sanctuary), the recipient\u0026rsquo;s attention state (2 levels: attending, not attending), and the behavioural context just before signal production (2 levels: social or non-social, based on whether the subject was in association with another individual prior to the emission of the signal). Signaller (82 levels), social goal (18 levels, see above) and recipient identity (128 levels) were included as random effects, allowing the mean behavioural expression (i.e. intercept) to vary among signallers, social goals and recipients. Social group was initially included as another random effect, but it was removed because it did not explain any variance. Repeatability was calculated for the two response variables (MC and MS) using the R package rptR (default settings, v. 0.9.22; \u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e) to estimate the amount of variation in the response variable explained by random effects (typically, signaller identity) or other grouping factors in the mixed models \u003csup\u003e\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e\u003c/sup\u003e. To ensure the datasets were representative of the chimpanzee\u0026rsquo;s communicative behaviour, only individuals that contributed more than two datapoints were included.\u003c/p\u003e\u003cp\u003eThird, to test whether individuals differ in how they shift their communicative strategies across contexts (i.e. behavioural plasticity), models with random intercepts and random slopes were fitted, based on Hertel et al. (2020)\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e and Fr\u0026ouml;hlich et al. (2022)\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Specifically, in addition to the previously described model, another model was fitted, including the same fixed and random effects as described above and a random slope over an \u0026ldquo;environmental gradient\u0026rdquo; (here a behavioural context contrast: social vs non-social contexts). Two models were thus fitted for each communicative strategy. We compared these models to the models without a random slope using a likelihood ratio test (LRT)\u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e. The strength of the models was insured by only including individuals that contributed more than two datapoints to each behavioural context.\u003c/p\u003e\u003cp\u003eThe absence of collinearity between predictor variables was confirmed using variance inflation factors (VIFs; \u003csup\u003e\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e\u003c/sup\u003e) from models including only the fixed effects (max VIF\u0026thinsp;=\u0026thinsp;1.19). To test whether signaller identity played a statistically significant role, we also compared the full models to a null model without the random intercept and slopes effects using an LRT. All statistical analyses were conducted using R v 4.3.1\u003csup\u003e72\u003c/sup\u003e.\u003c/p\u003e"},{"header":"3. Results","content":"\u003cp\u003eMC acts comprised 17.3% (N\u0026thinsp;=\u0026thinsp;1186) and MC acts comprised 42.7% (N\u0026thinsp;=\u0026thinsp;2931) of the total number of 6865 coded communicative acts. On average, 15.6% (range 0\u0026ndash;45.9%) of individuals\u0026rsquo; communicative acts were multicomponent and 39.9% (range 9.4\u0026ndash;68.0%) were multisensory. Most communicative acts were produced when the subjects were involved in non-social contexts (N\u0026thinsp;=\u0026thinsp;4992; 72.7% of the dataset), and in the sanctuary groups (N\u0026thinsp;=\u0026thinsp;3852; 56.1% of the dataset). Table S4 provides an overview of how MC and MS acts were broken down by social group and behavioural context.\u003c/p\u003e\u003cp\u003ea) \u003cem\u003eTypes of expressed signal and sensory combinations\u003c/em\u003e\u003c/p\u003e\u003cp\u003eIn a first step, the specific types of signal and sensory combinations were explored. Due to the number of categories and the widely contrasting amount of data in each of them, advanced statistical tests could not be run, but a descriptive overview of the types of MC and MS acts observed is presented in Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Males and females did not seem to differ systematically in the specific types of MC (e.g. manual-vocal) acts used, although the different types seemed to be more evenly expressed by females than by males (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). Similarly, there did not seem to be obvious differences between research settings in the frequency of each combination type (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). Regardless of the sex/setting class, the proportion of combined signal use varied dramatically between individuals, ranging from 0 to up to 100% for some combinations. Within a single MC communicative act, individuals mostly combined only two signals, while the combination of more than two signals was rarer, with the exception of the relatively common Facial\u0026thinsp;+\u0026thinsp;Manual\u0026thinsp;+\u0026thinsp;Vocal combination (11.1% of the MC acts; on average 12.2% of the individual repertoires). The most frequent combinations were Manual\u0026thinsp;+\u0026thinsp;Vocal (24.7% of the MC acts; on average 17.4% of the individual samples) and Body\u0026thinsp;+\u0026thinsp;Vocal (14.5% of the MC acts; on average 10.9% of the individual samples).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eWith regard to MS acts, we found that Tactile and Visual sensory channels were most commonly associated (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; 97% of the MS acts; on average 95.6% of the individual samples), followed by Auditory and Visual sensory channels (1.3% of the MS acts; on average 1.7% of the individual samples). There appeared to be no major differences between the sexes (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea) or the setting (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eb) \u003cem\u003eEffects of individual and environmental variables on the expression of multimodal strategies\u003c/em\u003e\u003c/p\u003e\u003cp\u003eGeneralized linear mixed effects models (GLMMs) were used to assess the effects of individual and environmental parameters on the two response variables (proportion of MC and MS communicative acts). Age, sex, recipient\u0026rsquo;s attention state and behavioural context were included as fixed effects and signaller identity, recipient identity and social goal as random effects. For both response variables, setting had a significant effect: both the average expression of MC and of MS communicative acts were significantly higher in sanctuary-housed individuals compared to zoo-housed individuals (MC: estimate\u0026thinsp;\u0026plusmn;\u0026thinsp;s.e.m. = 0.497\u0026thinsp;\u0026plusmn;\u0026thinsp;0.176, p\u0026thinsp;=\u0026thinsp;0.005; MS: 0.294\u0026thinsp;\u0026plusmn;\u0026thinsp;0.132, p\u0026thinsp;=\u0026thinsp;0.027). Additionally, the behavioural context had a significant effect on the average expression of MC, with individuals using more MC acts in social compared to non-social contexts (0.346\u0026thinsp;\u0026plusmn;\u0026thinsp;0.124, p\u0026thinsp;=\u0026thinsp;0.005), but only showed a positive trend on the expression of MS communicative acts (0.189\u0026thinsp;\u0026plusmn;\u0026thinsp;0.097, p\u0026thinsp;=\u0026thinsp;0.052). Males expressed less MC acts than females (-0.337\u0026thinsp;\u0026plusmn;\u0026thinsp;0.162, p\u0026thinsp;=\u0026thinsp;0.037), and more MS acts were expressed when the recipient was visually attentive (2.815\u0026thinsp;\u0026plusmn;\u0026thinsp;0.132, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Age did not have an effect on any communicative strategy (MC: 0.066\u0026thinsp;\u0026plusmn;\u0026thinsp;0.073, p\u0026thinsp;=\u0026thinsp;0.369; MS: 0.077\u0026thinsp;\u0026plusmn;\u0026thinsp;0.057, p\u0026thinsp;=\u0026thinsp;0.179; see electronic supplementary material, Tab. S5, for detailed information on the model outputs).\u003c/p\u003e\u003cp\u003ec) \u003cem\u003eIndividual variation in behavioural type for the multimodal strategies\u003c/em\u003e\u003c/p\u003e\u003cp\u003eTo investigate whether individual chimpanzees differed in their communicative strategies on average, the previously introduced GLMMs with random intercepts only were fitted. The full model explained behavioural variation significantly better than the respective null model (which excluded signaller identity as a random effect) for both response variables (MC: χ\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;64.551, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, \u003cem\u003edf\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1; MS: χ\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;72.808, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, \u003cem\u003edf\u003c/em\u003e\u0026thinsp;=\u0026thinsp;1). This means that considering variation between individuals in the models improved their explanatory power.\u003c/p\u003e\u003cp\u003eFor both MC and MS, significant repeatability was found for signaller identity (MC: R\u0026thinsp;=\u0026thinsp;0.025, s.e. = 0.008, 95% confidence interval [CI]\u0026thinsp;=\u0026thinsp;0.008\u0026ndash;0.037, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; MS: R\u0026thinsp;=\u0026thinsp;0.024, s.e. = 0.006, 95% CI\u0026thinsp;=\u0026thinsp;0.012\u0026ndash;0.035, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), recipient identity (MC: R\u0026thinsp;=\u0026thinsp;0.023, s.e. = 0.007, 95% CI\u0026thinsp;=\u0026thinsp;0.006\u0026ndash;0.036, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; MS: R\u0026thinsp;=\u0026thinsp;0.014, s.e. = 0.005, 95% CI\u0026thinsp;=\u0026thinsp;0.004\u0026ndash;0.023, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) and to a larger extent for the social goal (MC: R\u0026thinsp;=\u0026thinsp;0.156, s.e. = 0.058, 95% CI\u0026thinsp;=\u0026thinsp;0.038\u0026ndash;0.257, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001; MS: R\u0026thinsp;=\u0026thinsp;0.176, s.e. = 0.051, 95% CI\u0026thinsp;=\u0026thinsp;0.073\u0026ndash;0.27, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). This means that around 2.5% of the remaining variance in MC and MS acts after controlling for confounding effects of age, sex, research setting, and behavioural context could be attributed to differences between individuals. Specifically, some individuals consistently expressed more MS (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea) or MC (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb) acts compared to others, and these differences were not primarily caused by predictable differences between socio-environmental (research setting or behavioural context) or individual (age, sex) parameters.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003ed) \u003cem\u003eIndividual variation in behavioural plasticity for two multimodal strategies\u003c/em\u003e\u003c/p\u003e\u003cp\u003eTo examine whether chimpanzees differ in how they shift communicative behaviour across social conditions, two models for each response variable (i.e. multicomponent and multisensory acts) were compared: one with a random intercept for signaller identity (see above), and one with a random intercept for signaller identity and a random slope for behavioural context within signaller identity.\u003c/p\u003e\u003cp\u003eAs previously mentioned, MC communicative acts were used more often when the individuals were involved in social than in non-social contexts (Table S5), but the model including the random slope did not add any explanatory power when compared to the simpler model (χ\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;1.295, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.730, \u003cem\u003edf\u0026thinsp;=\u0026thinsp;3;\u003c/em\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea,b). Thus, although individuals showed plasticity in their use of MC acts according to the behavioural context, they all did so in the same manner (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). The comparison of the two models showed a better fit of the more complex model (including the random slope for behavioural context within signaller identity) only for the use of MS acts (χ\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;44.5055, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001, \u003cem\u003edf\u0026thinsp;=\u0026thinsp;3;\u003c/em\u003e Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec,d). This indicates that not only do individuals adjust their use of MS communicative acts to the context of the interaction, they also do so in different ways. Most of the individuals expressed a higher proportion of MS acts when already engaged in social contexts (Table S5), others showed no plastic response, and some even expressed the opposite pattern (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb; see electronic supplementary material, Tabs. S5 and S6, for detailed information on the models estimates).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"4. Discussion","content":"\u003cp\u003eMultimodal approaches in studying primate communication are still severely underrepresented \u003csup\u003e\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e\u003c/sup\u003e, and the proximate factors impacting complex communicative strategies are still poorly understood. To address this research gap, the goal of this study was to partition variation in multimodal communicative strategies of zoo- and sanctuary-housed chimpanzees into its social, between- and within-individual sources. To do so, the effect of individual and environmental parameters on the expression on two behaviours, multicomponent and multisensory communicative acts, were examined. Individual socio-demographic parameters were expected to impact the use of MC and MS, particularly the sex and research setting. Nevertheless, individuals were expected to differ in their communicative strategies regardless of individual variables, namely in their behavioural types. Within-individual variation (plasticity) was expected for both communicative strategies, with individuals adjusting their use to the behavioural context. Finally, between- and within-individual variation were expected to interact, leading to variation in plasticity between individuals.\u003c/p\u003e\u003cp\u003eLooking at the role of socio-demographic parameters in the multimodal signalling of chimpanzees, a significant effect of the research setting was found on the expression of both MC and MS communicative acts: sanctuary-housed individuals combined signals and sensory channels more frequently than the zoo-housed individuals, in line with the prediction. Although there are similarities between zoo and sanctuary environments (e.g. limited space, food provided at specific times of the day), there are also major differences. Sanctuary-housed chimpanzees live in far bigger enclosures than zoo-housed ones, allowing them to engage in more wild-typical behaviour, such as fission-fusion dynamics \u003csup\u003e\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e\u003c/sup\u003e. In contrast, zoo individuals are forced to stay in relatively close proximity to one another. This can have far-reaching consequences for the social dynamics within the groups. Social interactions come with their load of uncertainty, and thus aggression risk for the interaction partners. As a consequence, individuals may use a variety of strategies to limit ambiguity in order to regulate social relationships with group members \u003csup\u003e\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e\u003c/sup\u003e, and avoid misunderstandings or communicative failure. The MC and MS use of communicative signals may be highly beneficial as it allows for better detection and interpretation of messages. It might thus be particularly important to invest in such strategies, especially when interactional contexts can change very quickly: wild chimpanzees have been shown to express greater signal complexity in contexts of high social uncertainty \u003csup\u003e\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e, and orangutans to rely more on multicomponent communicative acts in less predictable interaction outcomes and on multisensory communicative acts to improve effectiveness \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e. Because zoo-housed individuals spend more time in close proximity to all individuals in the group, they are more familiar with each other and probably need to rely less on complex signalling than sanctuary-housed individuals \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e. This is consistent with previous research showing that repetitions of vocal greeting signals are rarely used among strongly bonded individuals, but are frequently used among distant dyads \u003csup\u003e\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e\u003c/sup\u003e. Another explanation for this might be the differing physical and ecological environment between the two settings: the bigger enclosures of the sanctuary potentially increase the distance between the interaction partners and the richer vegetation in this setting causes the visual noise. These two features might lead sanctuary-housed chimpanzees to increase their use of MC and MS acts as a compensation strategy, to improve the chances of the communicative act to properly reach the recipient, as has been identified in frogs (e.g. \u003cem\u003eStaurois parvus\u003c/em\u003e \u003csup\u003e\u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e77\u003c/span\u003e\u003c/sup\u003e\u003cem\u003e)\u003c/em\u003e. They might thus combine vocal and auditory signals with gestural and visual signals more often than zoo-housed chimpanzees.\u003c/p\u003e\u003cp\u003eFemales were found to use significantly more MC acts than males, but not more MS acts, which is partly in line with the predictions. Similarly to setting differences, the strong and cooperative male-male social bonds \u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e,\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e might require less complex communicative strategies than female-female communication to successfully negotiate interactions, although the traditionally ascribed asocial nature of female chimpanzees has been heavily debated (e.g. \u003csup\u003e\u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e78\u003c/span\u003e\u003c/sup\u003e). Moreover, males and females might signal to conspecifics of the opposite sex with widely different social goals. For example, females may express more submissive, reassurance-seeking signals (e.g. \u0026ldquo;greetings\u0026rdquo;) towards males than males do towards females \u003csup\u003e\u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e79\u003c/span\u003e\u003c/sup\u003e. These types of interactions pose a particular risk for the signallers, who are typically lower-ranking and have less fighting ability than their recipients. The use of MC acts might thus be more important for females than for their male counterparts. Although this was not investigated in detail, the social goal of the interaction appears to play an important role in the use of both MC and MS communicative strategies. It would be interesting to explore this effect further, especially its interactions with other parameters, such as signaller and recipient sexes, as previous research has shown that these factors influence gestural signalling in chimpanzees \u003csup\u003e\u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e80\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThese findings may have intriguing implications for the study of evolutionary origins of politeness in human communication. Linguistic politeness in humans is thought of as a means to reduce social friction and soften threatening acts \u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e81\u003c/span\u003e\u003c/sup\u003e. Namely, when interaction partners share a stronger social bond, meaning that they interact often and know each other well, they need to put less effort into their communicative interactions \u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. This can be understood as a reduced need to rely on complex communicative strategies such as signal and sensory combinations. This phenomenon is mirrored by our findings, including the differentiated use of both these strategies between zoo- and sanctuary-housed individuals. Similar processes thus seem to occur in the communication of human and non-human primates, including the potentially similar reliance on social norm that is characteristic of human social behaviour \u003csup\u003e\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe behavioural context, namely the activity of the signaller prior to signalling, had a significant impact on the use of MC acts, but only a marginal one on that of MS acts. Surprisingly, individuals relied more on MC and MS acts when they were already engaged in a social activity with a conspecific (e.g. grooming or playing) than when they were not (e.g. resting or feeding). This seems counterintuitive, since one might expect that interactions without prior social engagement would involve less uncertainty. One possible explanation may be that, although the signaller was already socially engaged, their social partner might not necessarily be the intended recipient of the communicative act. Thus, switching interaction partners might require even more ambiguity-reducing strategies than initiating a social interaction from a non-social context. This might be particularly true here, considering that signals expressed with the affiliation/greeting social goal, and are thus likely to be directed at individuals outside of the current social activity, represent 30% of the datasets. While this factor was not considered in the present analyses, it presents an interesting avenue for future research. Nonetheless, the findings demonstrate that chimpanzees exhibit plasticity in their communicative behaviour, adjusting their expressions in response to changes in behavioural context.\u003c/p\u003e\u003cp\u003eWhen examining the distribution of the types of signal and sensory channel combinations, considerable variation between individuals of the same research setting was observed. This was somewhat supported by the repeatability results, which showed that 2.5% of the variance was explained by consistent individual differences, regardless of the social or contextual environment. These values are relatively small compared to other repeatability scores found in chimpanzees and other species \u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e,\u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e83\u003c/span\u003e\u003c/sup\u003e, and even to a similar study in orangutans \u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. Most of the variation seemed to be explained by the social goal of the interaction (accounting for 16 to 18% of the remaining variance), but other factors, such as context or recipient-related variables (as well as interacting factors, as mentioned above; e.g. sex/social goal), could also contribute to these low values. This would not be surprising given the immense flexibility chimpanzees exhibit in their communicative behaviour, and the social goal of the interaction is a major source of flexibility (means-end dissociation; \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan additionalcitationids=\"CR19\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e). Nevertheless, these results demonstrate that there are consistent individual differences in communicative behavioural types.\u003c/p\u003e\u003cp\u003eTo assess not only whether there was a setting-level contextual shift in communicative strategy, namely plasticity, but also between-individual variation in plasticity, a behavioural reaction norm framework was applied to the dataset. Contrary to our predictions, individuals varied in plasticity in their expression of MS acts but not MC acts, despite the apparent similarity between the two strategies. In particular, individuals seemed to adjust their use of MC acts to the behavioural context they were in prior to the emission of the signal (population-level behavioural plasticity, see above), but all of them did so in a similar way. Fr\u0026ouml;hlich et al. (2022) found variation in plasticity in orangutan mothers in two very different variables, which seems to suggest that such variation is more broadly present in orangutan communicative behaviour\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. However, a reason for this could be that this study focused on mother-infant interactions and not on all available interaction types, as was done here. Their results might thus specifically reflect variation in maternal styles, which commonly occurs in primates (e.g. \u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e,\u003cspan citationid=\"CR84\" class=\"CitationRef\"\u003e84\u003c/span\u003e\u003c/sup\u003e). Kin relationship and rank could not be controlled for in the present study due to the highly unbalanced data, but these, as well as an affiliation index, could have important effects on the strategies used by individuals in their communicative attempts. For example, lower-ranking chimpanzees tend to become more agitated in the presence of a higher-ranking group member, leading them to produce more vocalizations \u003csup\u003e\u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e85\u003c/span\u003e\u003c/sup\u003e, and modality combinations have been shown to be affected by dominance relationships and familiarity between individuals \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR86\" class=\"CitationRef\"\u003e86\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAnother explanation could be that the need to appropriately adjust one\u0026rsquo;s use of MC acts across behavioural contexts is so critical that it prevents the expression of differences in plasticity. Indeed, the results presented here show that individuals do express differing behavioural types in both social and non-social contexts; there is simply no variation between individuals in the strength and direction of their behavioural adjustment when these contexts change. This probably means that there is a required amount of MC communicative acts that individuals must express when involved in a social context for the interaction to proceed smoothly. Some social interactions are risky for the signaller, since they might be followed by directed aggression from the recipient if the signaller does not assess the situation properly. As mentioned above, individuals might thus rely on strategies to limit ambiguity, but they might also have to follow strict social norms to improve their chances of success \u003csup\u003e\u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e82\u003c/span\u003e,\u003cspan citationid=\"CR87\" class=\"CitationRef\"\u003e87\u003c/span\u003e\u003c/sup\u003e, as has been found with the greeting behaviour of this species: vocal sequences answer to specific rules based on the hierarchical and social relationship between social partners \u003csup\u003e\u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e76\u003c/span\u003e\u003c/sup\u003e. In this case, individuals seem to rely more on MC acts in social contexts, but they probably also \u003cem\u003eneed\u003c/em\u003e to do so in a specific way to minimize the uncertainty of the interaction.\u003c/p\u003e\u003cp\u003ePlasticity necessarily comes with a trade-off between the benefits an individual might gain from adjusting its behaviour to changing environments (e.g. fitness benefit to express the context-appropriate behaviour) and the costs of doing so (e.g. investing energy in an additional communicative effort). As a consequence, variation in plasticity among individuals may arise because of two phenomena: state- and frequency-dependency of the payoffs of behavioural plasticity \u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e,\u003cspan additionalcitationids=\"CR89\" citationid=\"CR88\" class=\"CitationRef\"\u003e88\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e90\u003c/span\u003e\u003c/sup\u003e. The phenomenon at play here seems to be at a crossroads between these, in the sense that the benefits of plasticity depend on external factors, not in terms of frequency but rather in terms of behavioural context \u003cem\u003e(~\u0026thinsp;state of the social environment\u003c/em\u003e). Namely, individuals might regularly find themselves in situations (e.g. behavioural contexts) where the risks (e.g. aggression) are reduced by a specific behavioural change (e.g. an increase in MC act expression). In other words, if increasing one\u0026rsquo;s use of MC acts is necessary to avoid communicative failure, individuals who decrease their MC acts expression might be faced with serious consequences (e.g. aggression, exclusion from food resources). In this case, variation in behavioural plasticity among individuals is regulated by the intrinsic characteristics of the environment of the interaction, namely, the strength of the risk associated with such interaction: the greater the risk, the more standardized the behaviour, and the weaker the differences in individual plasticity.\u003c/p\u003e\u003cp\u003eIn contrast, chimpanzees were found to vary in how they individually shift their use of MS acts between contexts (individual-level behavioural plasticity). This suggests that, contrary to the use of MC acts, the use of MS communicative acts might not be an aspect of chimpanzees\u0026rsquo; communicative behaviour that is particularly important for the success of communicative interactions, which results in less constraint on how individuals use them: MS acts might not need to be standardized. If the costs associated with the misuse of MS acts are low, individuals are not under strong selective pressure in their expression of MS acts, leaving room for individual differences in behavioural plasticity. The finding that chimpanzees did not adjust their behaviour to the behavioural context at the population-level supports this: MS acts were used equally in social and non-social contexts. This is also reflected in the frequent use of communicative acts that are both tactile and visual to the recipient, and are likely to be prevalent in interactions between familiar individuals, as tactile gestures require physical proximity and social tolerance. Because these interactions involve less risk, there is less pressure to always communicate in a standardized, unambiguous way.\u003c/p\u003e\u003cp\u003eAlthough they seem closely related, both communicative strategies were not similarly impacted by context changes. Mathot et al. (2011) showed that the payoffs of plasticity are not affected in the same way in all behavioural traits by the frequency-dependency phenomena: some payoffs are negative-frequency dependent (e.g. the benefits of being plastic decrease when more individuals are plastic) and others are positive-frequency dependent (e.g. the benefits of being plastic increase with increasing number of plastic individuals)\u003csup\u003e\u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e90\u003c/span\u003e\u003c/sup\u003e. Similarly, different communicative strategies seem to be affected to a varying degree by social risks. This result is in agreement with the previous finding that multicomponent and multisensory communicative acts have different functions in orangutans \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u003c/sup\u003e and shows that these analyses are highly variable-sensitive: plasticity and variation in plasticity might not be found in, or be adaptive for all communicative strategies.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eOne constitutive feature of language is the great flexibility that it requires in order to be properly adjusted to the situation in which the interactions take place, and this flexibility is partly achieved through multimodality. However, and despite the importance of the comparative approach in studies on language evolution, individual variation, and especially plasticity, have been neglected in studies on non-human communication systems. This study shows that individual chimpanzees vary in their use of multicomponent and multisensory communicative acts (i.e. individual variation in behavioural type), which are largely shaped by the setting they live in. Chimpanzees also showed some level of behavioural plasticity between behavioural contexts, but individual variation in plasticity was identified in only one of the two communicative strategies.\u003c/p\u003e\u003cp\u003eMultimodality in chimpanzee communicative flexibility appears to play a similar role as it is in human language: communicative complexity is associated both with the familiarity between and the adjustments of the communicative behaviour (i.e. plasticity) to the interaction partner. This underscores the relevance of comparative communication studies in understanding the evolutionary origins of language.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Barbara Struve (Leintalzoo Schwaigern), Hanna Petschauer and Daniel Hanus(Zoo Leipzig, MPI-EVA), Edwin van Leeuwen, Katja Liebal and Thalita Sim\u0026otilde;es Calvi (Chimfunshi Wildlife Orphanage Trust) as well as the research staff and zoo keepers for their helpful collaboration during this study. We also thank the entire \u0026ldquo;Pathways to Language\u0026rdquo; group for support and help in making this study possible, and in particular Wytse Wilhelm for helpful advice on statistical analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by the Volkswagen Foundation (Freigeist Fellowship 9B138 awarded to MF) and the German Research Foundation (DFG FOR 2237: Words, Bones, Genes, Tools \u0026ndash; Tracking Linguistic, Cultural and Biological Trajectories of the Human Past; PIs: Katerina Harvati and Gerhard J\u0026auml;ger).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData accessibility\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets and R script that support the findings of this study have been deposited on GitHub: https://github.com/angelelombrey/ALombrey-multi-plasti-Data-and-code.git.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contribution\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eMF and AL designed the study. AL and ALM collected and coded the data. ND coded the data. AL analysed the data with critical help from UK and MF. AL and MF wrote the manuscript with inputs from ALM, UK and KH.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eSchwarz, N. \u003cem\u003eCognition and Communication: Judgmental Biases, Research Methods, and the Logic of Conversation\u003c/em\u003e (Psychology, 2014). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.4324/9781315805887\u003c/span\u003e\u003cspan address=\"10.4324/9781315805887\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHeesen, R. \u0026amp; Fr\u0026ouml;hlich, M. Revisiting the human \u0026lsquo;interaction engine\u0026rsquo;: comparative approaches to social action coordination. \u003cem\u003ePhilos. Trans. R Soc. B Biol. Sci.\u003c/em\u003e \u003cb\u003e377\u003c/b\u003e, 20210092 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChristiansen, M. H. \u0026amp; Kirby, S. Language evolution: consensus and controversies. \u003cem\u003eTrends Cogn. Sci.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 300\u0026ndash;307 (2003).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTomasello, M. Origins of human communication. \u003cem\u003eMIT Press.\u003c/em\u003e \u003cb\u003e37\u003c/b\u003e, 393 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFitch, W. T. Empirical approaches to the study of language evolution. \u003cem\u003ePsychon Bull. Rev.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e, 3\u0026ndash;33 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eClark, H. H. \u0026amp; Brennan, S. E. Grounding in communication. in Perspectives on socially shared cognition. (eds Resnick, L. B., Levine, J. M. \u0026amp; Teasley, S. D.) 127\u0026ndash;149 (American Psychological Association, Washington, doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1037/10096-006\u003c/span\u003e\u003cspan address=\"10.1037/10096-006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. (1991).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrown, L., Kim, S. U. \u0026amp; Kim, H. The embodied enactment of politeness metapragmatics. \u003cem\u003eJ. Politeness Res.\u003c/em\u003e \u003cb\u003e19\u003c/b\u003e, 149\u0026ndash;183 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBrown, P. \u0026amp; Levinson, S. C. \u003cem\u003ePoliteness: Some Universals in Language Usage\u003c/em\u003e (Cambridge University Press, 1987).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFerreira, V. S. Is It Better to Give Than to Donate? Syntactic Flexibility in Language Production. \u003cem\u003eJ. Mem. Lang.\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 724\u0026ndash;755 (1996).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePoss, S. R., Kuhar, C., Stoinski, T. S. \u0026amp; Hopkins, W. D. Differential use of attentional and visual communicative signaling by orangutans (Pongo pygmaeus) and gorillas (Gorilla gorilla) in response to the attentional status of a human. \u003cem\u003eAm. J. Primatol.\u003c/em\u003e \u003cb\u003e68\u003c/b\u003e, 978\u0026ndash;992 (2006).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFr\u0026ouml;hlich, M. et al. Multicomponent and multisensory communicative acts in orang-utans may serve different functions. \u003cem\u003eCommun. Biol.\u003c/em\u003e \u003cb\u003e4\u003c/b\u003e, 1\u0026ndash;13 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiebal, K., Pika, S. \u0026amp; Tomasello, M. Gestural communication of orangutans (\u003cem\u003ePongo pygmaeus\u003c/em\u003e). \u003cem\u003eGesture\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 1\u0026ndash;38 (2006).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeavens, D. A., Russell, J. L. \u0026amp; Hopkins, W. D. Multimodal communication by captive chimpanzees (Pan troglodytes). \u003cem\u003eAnim. Cogn.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 33\u0026ndash;40 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHobaiter, C. \u0026amp; Byrne, R. W. The gestural repertoire of the wild chimpanzee. \u003cem\u003eAnim. Cogn.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 745\u0026ndash;767 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGenty, E., Clay, Z., Hobaiter, C. \u0026amp; Zuberb\u0026uuml;hler, K. Multi-Modal Use of a Socially Directed Call in Bonobos. \u003cem\u003ePLoS ONE\u003c/em\u003e. \u003cb\u003e9\u003c/b\u003e, e84738 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCartmill, E. A. \u0026amp; Byrne, R. W. Orangutans Modify Their Gestural Signaling According to Their Audience\u0026rsquo;s Comprehension. \u003cem\u003eCurr. Biol.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 1345\u0026ndash;1348 (2007).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRoberts, A. I., Vick, S. J. \u0026amp; Buchanan-Smith, H. M. Communicative intentions in wild chimpanzees: persistence and elaboration in gestural signalling. \u003cem\u003eAnim. Cogn.\u003c/em\u003e \u003cb\u003e16\u003c/b\u003e, 187\u0026ndash;196 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTomasello, M., Call, J., Nagell, K., Olguin, R. \u0026amp; Carpenter, M. The learning and use of gestural signals by young chimpanzees: A trans-generational study. \u003cem\u003ePrimates\u003c/em\u003e \u003cb\u003e35\u003c/b\u003e, 137\u0026ndash;154 (1994).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePika, S., Liebal, K. \u0026amp; Tomasello, M. Gestural communication in young gorillas (Gorilla gorilla): Gestural repertoire, learning, and use. \u003cem\u003eAm. J. Primatol.\u003c/em\u003e \u003cb\u003e60\u003c/b\u003e, 95\u0026ndash;111 (2003).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFr\u0026ouml;hlich, M. \u0026amp; Hobaiter, C. The development of gestural communication in great apes. \u003cem\u003eBehav. Ecol. Sociobiol.\u003c/em\u003e \u003cb\u003e72\u003c/b\u003e, 194 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKendon, A. \u003cem\u003eGesture: Visible Action as Utterance\u003c/em\u003e (Cambridge University Press, 2004).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePerniss, P. Why We Should Study Multimodal Language. \u003cem\u003eFront Psychol\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHoller, J. \u0026amp; Levinson, S. C. Multimodal Language Processing in Human Communication. \u003cem\u003eTrends Cogn. Sci.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e, 639\u0026ndash;652 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSlocombe, K. E., Waller, B. M. \u0026amp; Liebal, K. The language void: the need for multimodality in primate communication research. \u003cem\u003eAnim. Behav.\u003c/em\u003e \u003cb\u003e81\u003c/b\u003e, 919\u0026ndash;924 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFr\u0026ouml;hlich, M., Sievers, C., Townsend, S. W., Gruber, T. \u0026amp; van Schaik, C. P. Multimodal communication and language origins: integrating gestures and vocalizations. \u003cem\u003eBiol. Rev.\u003c/em\u003e \u003cb\u003e94\u003c/b\u003e, 1809\u0026ndash;1829 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFitch, W. T. \u003cem\u003eThe Evolution of Language\u003c/em\u003e (Cambridge University Press, 2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWaller, B. M., Liebal, K., Burrows, A. M. \u0026amp; Slocombe Katie. E. How Can a Multimodal Approach to Primate Communication Help Us Understand the Evolution of Communication? \u003cem\u003eEvol. Psychol.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 538\u0026ndash;549 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMcgurk, H. \u0026amp; Macdonald, J. Hearing lips and seeing voices. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e264\u003c/b\u003e, 746\u0026ndash;748 (1976).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMacdonald, J. \u0026amp; McGurk, H. Visual influences on speech perception processes. \u003cem\u003ePercept. Psychophys\u003c/em\u003e. \u003cb\u003e24\u003c/b\u003e, 253\u0026ndash;257 (1978).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTiippana, K. What is the McGurk effect? \u003cem\u003eFront Psychol\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e, (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEkman, P. \u0026amp; Friesen, W. V. The Repertoire of Nonverbal Behavior: Categories, Origins, Usage, and Coding. \u003cem\u003eSemiotica\u003c/em\u003e \u003cb\u003e1\u003c/b\u003e, 49\u0026ndash;98 (1969).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHebets, E. A. \u0026amp; Papaj, D. R. Complex signal function: developing a framework of testable hypotheses. \u003cem\u003eBehav. Ecol. Sociobiol.\u003c/em\u003e \u003cb\u003e57\u003c/b\u003e, 197\u0026ndash;214 (2005).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWilke, C. et al. Production of and responses to unimodal and multimodal signals in wild chimpanzees, Pan troglodytes schweinfurthii. \u003cem\u003eAnim. Behav.\u003c/em\u003e \u003cb\u003e123\u003c/b\u003e, 305\u0026ndash;316 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFr\u0026ouml;hlich, M., van Schaik, C. P., van Noordwijk, M. A. \u0026amp; Knief, U. Individual variation and plasticity in the infant-directed communication of orang-utan mothers. \u003cem\u003eProc. R. Soc. B Biol. Sci.\u003c/em\u003e 289, 20220200 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWolf, M. \u0026amp; Weissing, F. J. Animal personalities: consequences for ecology and evolution. \u003cem\u003eTrends Ecol. Evol.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e, 452\u0026ndash;461 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDingemanse, N. J. \u0026amp; Araya-Ajoy, Y. G. Interacting personalities: behavioural ecology meets quantitative genetics. \u003cem\u003eTrends Ecol. Evol.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, 88\u0026ndash;97 (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTkaczynski, P. J. et al. Long-term repeatability in social behaviour suggests stable social phenotypes in wild chimpanzees. \u003cem\u003eR Soc. Open. Sci.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 200454 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDingemanse, N. J. \u0026amp; Dochtermann, N. A. Quantifying individual variation in behaviour: mixed-effect modelling approaches. \u003cem\u003eJ. Anim. Ecol.\u003c/em\u003e \u003cb\u003e82\u003c/b\u003e, 39\u0026ndash;54 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLombrey, A. \u0026amp; Fr\u0026ouml;hlich, M. Individual variation in animal communication: from species averages to unique voices. \u003cem\u003eBiol Rev\u003c/em\u003e (in press).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Schaik, C. P. The costs and benefits of flexibility as an expression of behavioural plasticity: a primate perspective. \u003cem\u003ePhilos. Trans. R Soc. B Biol. Sci.\u003c/em\u003e \u003cb\u003e368\u003c/b\u003e, 20120339 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFairbanks, L. A. Individual Differences in Maternal Style: Causes and Consequences for Mothers and offspring. in Advances in the Study of Behavior (eds Rosenblatt, J. S. \u0026amp; Snowdon, C. T.) vol. 25 579\u0026ndash;611 (Academic, (1996).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDingemanse, N. J., Kazem, A. J. N., R\u0026eacute;ale, D. \u0026amp; Wright, J. Behavioural reaction norms: animal personality meets individual plasticity. \u003cem\u003eTrends Ecol. Evol.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 81\u0026ndash;89 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDingemanse, N. J. \u0026amp; Wolf, M. Between-individual differences in behavioural plasticity within populations: causes and consequences. \u003cem\u003eAnim. Behav.\u003c/em\u003e \u003cb\u003e85\u003c/b\u003e, 1031\u0026ndash;1039 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHertel, A. G., Niemel\u0026auml;, P. T., Dingemanse, N. J. \u0026amp; Mueller, T. A guide for studying among-individual behavioral variation from movement data in the wild. \u003cem\u003eMov. Ecol.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 30 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRoberts, A. I., Roberts, S. G. B. \u0026amp; Vick, S. J. The repertoire and intentionality of gestural communication in wild chimpanzees. \u003cem\u003eAnim. Cogn.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 317\u0026ndash;336 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTaglialatela, J. P. et al. Multimodal communication in chimpanzees. \u003cem\u003eAm. J. Primatol.\u003c/em\u003e \u003cb\u003e77\u003c/b\u003e, 1143\u0026ndash;1148 (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHobaiter, C. \u0026amp; Byrne, R. W. The Meanings of Chimpanzee Gestures. \u003cem\u003eCurr. Biol.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e, 1596\u0026ndash;1600 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCrockford, C. Why does the chimpanzee vocal repertoire remain poorly understood and what can be done about it? in The Chimpanzees of the Ta\u0026iuml; Forest (eds Boesch, C. et al.) 394\u0026ndash;409 (Cambridge University Press, doi:\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1017/9781108674218.025\u003c/span\u003e\u003cspan address=\"10.1017/9781108674218.025\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDavila-Ross, M., Jesus, G., Osborne, J. \u0026amp; Bard, K. A. Chimpanzees (Pan troglodytes) Produce the Same Types of \u0026lsquo;Laugh Faces\u0026rsquo; when They Emit Laughter and when They Are Silent. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e10\u003c/b\u003e, e0127337 (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHobaiter, C., Byrne, R. W. \u0026amp; Zuberb\u0026uuml;hler, K. Wild chimpanzees\u0026rsquo; use of single and combined vocal and gestural signals. \u003cem\u003eBehav. Ecol. Sociobiol.\u003c/em\u003e \u003cb\u003e71\u003c/b\u003e, 96 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDoherty, E., Davila-Ross, M. \u0026amp; Clay, Z. Multimodal communication development in semiwild chimpanzees. \u003cem\u003eAnim. Behav.\u003c/em\u003e \u003cb\u003e201\u003c/b\u003e, 175\u0026ndash;190 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGrampp, M. et al. Social uncertainty promotes signal complexity during approaches in wild chimpanzees (Pan troglodytes verus) and mangabeys (Cercocebus atys atys). \u003cem\u003eR Soc. Open. Sci.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 231073 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ede Waal, F. B. M. Sex differences in the formation of coalitions among chimpanzees. \u003cem\u003eEthol. Sociobiol.\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e, 239\u0026ndash;255 (1984).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLehmann, J. \u0026amp; Boesch, C. Sexual Differences in Chimpanzee Sociality. \u003cem\u003eInt. J. Primatol.\u003c/em\u003e \u003cb\u003e29\u003c/b\u003e, 65\u0026ndash;81 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMorin, P. A. et al. Kin Selection, Social Structure, Gene Flow, and the Evolution of Chimpanzees. \u003cem\u003eScience\u003c/em\u003e \u003cb\u003e265\u003c/b\u003e, 1193\u0026ndash;1201 (1994).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNishida, T. \u0026amp; Hosaka, K. Coalition strategies among adult male chimpanzees of the Mahale Mountains, Tanzania. in Great Ape Societies (eds McGrew, W., Marchant, L. \u0026amp; Nishida, T.) 114\u0026ndash;134 (Cambridge University Press, Cambridge, (1996).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGilby, I. C. \u0026amp; Wrangham, R. W. Association patterns among wild chimpanzees (Pan troglodytes schweinfurthii) reflect sex differences in cooperation. \u003cem\u003eBehav. Ecol. Sociobiol.\u003c/em\u003e \u003cb\u003e62\u003c/b\u003e, 1831\u0026ndash;1842 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNeal Webb, S. J., Hau, J. \u0026amp; Schapiro, S. J. Does group size matter? Captive chimpanzee (Pan troglodytes) behavior as a function of group size and composition. \u003cem\u003eAm. J. Primatol.\u003c/em\u003e \u003cb\u003e81\u003c/b\u003e, e22947 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeTroy, S. E., Ross, C. T., Cronin, K. A., van Leeuwen, E. J. C. \u0026amp; Haun, D. B. M. Cofeeding tolerance in chimpanzees depends on group composition: A longitudinal study across four communities. \u003cem\u003eiScience\u003c/em\u003e 24, (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKoski, S. E. Social personality traits in chimpanzees: temporal stability and structure of behaviourally assessed personality traits in three captive populations. \u003cem\u003eBehav. Ecol. Sociobiol.\u003c/em\u003e \u003cb\u003e65\u003c/b\u003e, 2161\u0026ndash;2174 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiebal, K., Call, J. \u0026amp; Tomasello, M. Use of gesture sequences in chimpanzees. \u003cem\u003eAm. J. Primatol.\u003c/em\u003e \u003cb\u003e64\u003c/b\u003e, 377\u0026ndash;396 (2004).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCall, J. How Apes Use Gestures: The Issue of Flexibility. in \u003cem\u003eEvolution of Communicative Flexibility: Complexity, Creativity, and Adaptability in Human and Animal Communication\u003c/em\u003e (eds. Oller, D. K. \u0026amp; Griebel, U.) 235\u0026ndash;252MIT Press, Cambridge, (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Leeuwen, E. J. C., Bruinstroop, B. M. C. \u0026amp; Haun, D. B. M. Early Trauma Leaves No Social Signature in Sanctuary-Housed Chimpanzees (Pan troglodytes). \u003cem\u003eAnimals\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 49 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFriard, O. \u0026amp; Gamba, M. BORIS: a free, versatile open-source event‐logging software for video/audio coding and live observations. \u003cem\u003eMethods Ecol. Evol.\u003c/em\u003e \u003cb\u003e7\u003c/b\u003e, 1325\u0026ndash;1330 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBakeman, R. \u0026amp; Quera, V. \u003cem\u003eSequential Analysis and Observational Methods for the Behavioral Sciences\u003c/em\u003e (Cambridge University Press, 2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan de Pol, M. Quantifying individual variation in reaction norms: how study design affects the accuracy, precision and power of random regression models. \u003cem\u003eMethods Ecol. Evol.\u003c/em\u003e \u003cb\u003e3\u003c/b\u003e, 268\u0026ndash;280 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBates, D. et al. lme4: Linear Mixed-Effects Models using \u0026lsquo;Eigen\u0026rsquo; and S4. (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStoffel, M. A., Nakagawa, S. \u0026amp; Schielzeth, H. rptR: repeatability estimation and variance decomposition by generalized linear mixed-effects models. \u003cem\u003eMethods Ecol. Evol.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 1639\u0026ndash;1644 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNakagawa, S. \u0026amp; Schielzeth, H. Repeatability for Gaussian and non-Gaussian data: a practical guide for biologists. \u003cem\u003eBiol. Rev.\u003c/em\u003e \u003cb\u003e85\u003c/b\u003e, 935\u0026ndash;956 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDobson, A. J. \u003cem\u003eAn Introduction to Generalized Linear Models\u003c/em\u003e (Chapman \u0026amp; Hall / CRC, 2002).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eQuinn, G. P. \u0026amp; Keough, M. J. \u003cem\u003eExperimental Design and Data Analysis for Biologists\u003c/em\u003e (Cambridge University Press, 2002).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eR Core Team. R: A Language and Environment for Statistical Computing.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiebal, K., Slocombe, K. E. \u0026amp; Waller, B. M. The language void 10 years on: multimodal primate communication research is still uncommon. \u003cem\u003eEthol. Ecol. Evol.\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, 274\u0026ndash;287 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Leeuwen, E. J. C. Temporal stability of chimpanzee social culture. \u003cem\u003eBiol. Lett.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 20210031 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDal Pesco, F. \u0026amp; Fischer, J. On the evolution of baboon greeting rituals. \u003cem\u003ePhilos. Trans. R Soc. B Biol. Sci.\u003c/em\u003e \u003cb\u003e375\u003c/b\u003e, 20190420 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLuef, E. M. \u0026amp; Pika, S. Social relationships and greetings in wild chimpanzees (Pan troglodytes): use of signal combinations. \u003cem\u003ePrimates\u003c/em\u003e \u003cb\u003e60\u003c/b\u003e, 507\u0026ndash;515 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGrafe, T. U. et al. Multimodal Communication in a Noisy Environment: A Case Study of the Bornean Rock Frog Staurois parvus. \u003cem\u003ePLOS ONE\u003c/em\u003e. \u003cb\u003e7\u003c/b\u003e, e37965 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLehmann, J. \u0026amp; Boesch, C. Sociality of the dispersing sex: the nature of social bonds in West African female chimpanzees, \u003cem\u003ePan troglodytes\u003c/em\u003e. \u003cem\u003eAnim. Behav.\u003c/em\u003e \u003cb\u003e77\u003c/b\u003e, 377\u0026ndash;387 (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBernstein, I. S. \u0026amp; Gordon, T. P. The Function of Aggression in Primate Societies: Uncontrolled aggression may threaten human survival, but aggression may be vital to the establishment and regulation of primate societies and sociality. \u003cem\u003eAm. Sci.\u003c/em\u003e \u003cb\u003e62\u003c/b\u003e, 304\u0026ndash;311 (1974).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eScott, N. M. Gesture Use by Chimpanzees (an troglodytes): Differences Between Sexes in Inter- and Intra-Sexual Interactions. \u003cem\u003eAm. J. Primatol.\u003c/em\u003e \u003cb\u003e75\u003c/b\u003e, 555\u0026ndash;567 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLakoff, R. Linguistic Theory and the Real World. \u003cem\u003eLang. Learn.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 309\u0026ndash;338 (1975).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evon Rudolf, C. \u0026amp; Burkart, J. M. Schaik, C. P. Evolutionary precursors of social norms in chimpanzees: a new approach. \u003cem\u003eBiol. Philos.\u003c/em\u003e \u003cb\u003e26\u003c/b\u003e, 1\u0026ndash;30 (2011). van.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBell, A. M., Hankison, S. J. \u0026amp; Laskowski, K. L. The repeatability of behaviour: a meta-analysis. \u003cem\u003eAnim. Behav.\u003c/em\u003e \u003cb\u003e77\u003c/b\u003e, 771\u0026ndash;783 (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVerderane, M. P. \u0026amp; Izar, P. Maternal care styles in primates: considering a New World species. \u003cem\u003ePsicol. USP\u003c/em\u003e. \u003cb\u003e30\u003c/b\u003e, e190055 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGoodall, J. The chimpanzees of Gombe: patterns of behaviour. \u003cem\u003eNo Title\u003c/em\u003e (1986).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFedurek, P. et al. The function of chimpanzee greeting calls is modulated by their acoustic variation. \u003cem\u003eAnim. Behav.\u003c/em\u003e \u003cb\u003e174\u003c/b\u003e, 279\u0026ndash;289 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFlack, J. C., Jeannotte, L. A. \u0026amp; de Waal, F. B. M. Play Signaling and the Perception of Social Rules by Juvenile Chimpanzees (Pan troglodytes). \u003cem\u003eJ. Comp. Psychol.\u003c/em\u003e \u003cb\u003e118\u003c/b\u003e, 149\u0026ndash;159 (2004).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWolf, M., van Doorn, G. S. \u0026amp; Weissing, F. J. Evolutionary emergence of responsive and unresponsive personalities. \u003cem\u003eProc. Natl. Acad. Sci.\u003c/em\u003e 105, 15825\u0026ndash;15830 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVan Gils, J. A. State-dependent Bayesian foraging on spatially autocorrelated food distributions. \u003cem\u003eOikos\u003c/em\u003e \u003cb\u003e119\u003c/b\u003e, 237\u0026ndash;244 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMathot, K. J. et al. Disentangling the roles of frequency-vs. state-dependence in generating individual differences in behavioural plasticity. \u003cem\u003eEcol. Lett.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 1254\u0026ndash;1262 (2011).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"behavioural reaction norms, context, individual variation, multimodal communication, multicomponent signalling, multisensory signalling, Pan troglodytes, plasticity","lastPublishedDoi":"10.21203/rs.3.rs-6901931/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6901931/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eHuman communication is remarkable for its flexibility, a trait largely reflected in its multimodal nature and shared to some extent with nonhuman primates. Although individual differences in social behaviour have known evolutionary implications, their role in shaping primate communication remains largely unexplored. This study adopts a multimodal framework to partition variation in chimpanzees\u0026rsquo; use of multicomponent and multisensory communicative strategies into socio-environmental, between-individual, and within-individual sources. Results showed that research setting and signaller\u0026rsquo;s sex affected communicative expression. Importantly, we also detected consistent between-individual differences in both strategies, independent of age, sex, or setting. While only multicomponent signal use was predicted by behavioural context at the population level, only for the use of multisensory acts did individuals vary in how they adjusted to context. These findings reveal profound flexibility in chimpanzee communication, highlighting individual-specific patterns and supporting a gradual evolutionary pathway toward the complexity of human multimodal communication.\u003c/p\u003e","manuscriptTitle":"Flexible use of multimodal communicative strategies in adult chimpanzees (Pan troglodytes)","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-18 12:41:25","doi":"10.21203/rs.3.rs-6901931/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-07-09T07:02:09+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-28T13:48:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-06-27T03:56:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"112713705343907697357349348197021455098","date":"2025-06-23T14:43:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"8968804551926973869790657123737107469","date":"2025-06-19T15:53:08+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-06-19T14:16:30+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-19T14:08:43+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-06-18T02:40:56+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-06-16T09:41:04+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-06-16T05:50:58+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"01f0dd39-20db-44de-a371-87043c8ea208","owner":[],"postedDate":"July 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":51259356,"name":"Biological sciences/Evolution/Anthropology/Biological anthropology"},{"id":51259357,"name":"Biological sciences/Zoology/Animal behaviour"}],"tags":[],"updatedAt":"2025-09-15T16:05:27+00:00","versionOfRecord":{"articleIdentity":"rs-6901931","link":"https://doi.org/10.1038/s41598-025-14835-x","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-09-09 15:56:57","publishedOnDateReadable":"September 9th, 2025"},"versionCreatedAt":"2025-07-18 12:41:25","video":"","vorDoi":"10.1038/s41598-025-14835-x","vorDoiUrl":"https://doi.org/10.1038/s41598-025-14835-x","workflowStages":[]},"version":"v1","identity":"rs-6901931","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6901931","identity":"rs-6901931","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}