{"paper_id":"267bf797-4666-41ed-bc72-b2c595056864","body_text":"Wild and zoo-housed orangutans differ in how they explore objects | 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 Wild and zoo-housed orangutans differ in how they explore objects Isabelle B. Laumer, Shubhangi Kansal, Anais van Cauwenberghe, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5594384/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 30 Apr, 2025 Read the published version in Scientific Reports → Version 1 posted 3 You are reading this latest preprint version Abstract In human infants, exploratory object manipulations (henceforth called “EOM”) stimulate cognitive development and affect cognitive performance in later life. Zoo-housed great apes are frequently used to study the evolution of human cognition, however, it is unknown how the zoo environment affects their daily expression of EOM. We investigated how wild and zoo-housed Sumatran orangutans differ in their daily EOM throughout life. We collected ~ 12’000 EOM events by 51 wild and zoo-housed individuals of all ages. Zoo-housed orangutans showed significantly higher EOM rates than wild orangutans. Exploratory actions were more diverse in zoos than in the wild, even with objects available in both settings. Zoo-housed orangutans also showed a larger repertoire of exploratory actions and a higher probability of multi-object exploration, including tool use. There was no difference between settings at which age individuals first showed specific exploratory actions. Our results show that the zoo environment significantly affects EOM in orangutans and that the species’ exploratory potential exceeds its natural expression. This may have important implications for cognitive performance, as zoo-housed individuals are likely to have a broader range of affordances to draw from when confronted with novel problems. These results highlight the potential of captive-wild comparisons to study cognitive development and evolution. Biological sciences/Developmental biology Biological sciences/Evolution exploration exploratory object manipulation captive-wild comparison cognitive development cognitive aging great ape Figures Figure 1 Figure 2 Figure 3 Introduction Exploratory object manipulation (i.e., the deliberate manipulation and visual inspection of objects ( 1 – 5 ), henceforth referred to as “EOM”) enables human infants to learn about objects and their physical properties (e.g. texture, weight, gravity, etc. ( 1 , 3 , 4 , 6 – 10 )). EOM is also an important vehicle of cognitive stimulation ( 11 ) as it is linked to motor and cognitive development ( 12 – 20 ). Neurological studies further provide evidence for a link between EOM and cognitive processing ( 21 – 23 ). In human infants, manual, visual, and oral EOM peaks around the age of two years and then decreases ( 3 , 24 – 26 ). However, the length of EOM events increases ( 3 , 20 , 27 , 28 ), and EOM becomes more diverse and complex with increasing age (e.g., a larger number and different exploratory actions are combined; ( 1 , 3 , 24 , 28 )). These patterns are indicative of learning processes as actions become more fine-tuned and more goal-directed over time ( 29 ). Similar to humans, EOM also leads to the development of knowledge and sensorimotor skills in non-human animals (e.g. ( 30 – 35 )). Studies in various mammalian and bird species have shown that EOM correlates positively with problem-solving ability and innovation probability ( 36 – 43 ). This suggests that knowledge and experience gained during EOM can be transferred across contexts and applied to novel problems. Thus, similar to humans, EOM also appears to be a key mechanism in non-human animals for translating cognitive potential into actual knowledge and skills as well as for fostering cognitive performance in several domains (e.g., ( 5 , 44 )). In primates, EOM tendency is closely linked with phylogeny in that strepsirrhine primates show the lowest levels of EOM tendency, followed by Platyrrhines, non-hominid Catarrhines, and great apes ( 45 ). Non-human great apes (henceforth called “great apes”) show the most diverse and complex EOMs of all non-human primates in terms of rates of bimanual manipulation, object-object (i.e., two or more are combined), and object-substrate manipulations (i.e., an object is actively combined with a substrate, other than the plain floor) as well as the use of different body parts during manipulation ( 3 , 46 – 49 ). Humans demonstrate even more complex and diverse manipulations ( 50 , 51 ). Despite these differences, manipulations present across multiple taxa develop in a highly similar sequence, suggesting relatively conserved neurodevelopmental processes ( 52 , 53 ). Furthermore, like humans, wild great apes show a peak in exploratory tendency during the early dependency period ( 5 , 33 ) and an increase of EOM diversity with increasing age ( 5 ). In large-brained species with pronounced developmental plasticity, cognition is shaped by the social and ecological environment ( 54 , 55 ). In humans, the conditions under which one grows up influence cognitive development (e.g., ( 56 – 58 )), and cognitive performance is shaped by specific experiences during development ( 59 ). Children growing up with severely limited social and physical stimulation, later show deficits in several cognitive domains, along with physiological and anatomical changes in the brain (e.g., ( 57 , 60 )). Similarly, rhesus monkeys ( Macaca mulatta ) that grow up in socially deprived conditions perform worse in various cognitive tasks and have significantly altered brain structures (( 61 ), see ( 62 ) for similar findings in rats). High frequency and quality of social inputs during early childhood on the other hand positively affect cognitive development in humans ( 63 ). Accordingly, enculturation (i.e., being raised with close human contact) positively affects performance in physical and social cognition tasks in great apes ( 64 – 69 ). Effects of the ecological and social environment on the development of cognition appear to be, at least to some degree, mediated through exploration ( 42 , 68 , 70 ). Zoo-housed great apes are a frequently used model to study human cognition. Compared to their wild conspecifics, they are easier to access and observe, and can be tested under controlled conditions, which allows for minimizing environmental and social influences for the time of testing. As cognitive studies on zoo-housed great apes often draw conclusions on the evolutionary level, it is crucial to increase our understanding of how the zoo environment affects their cognitive performance. Wild and zoo-housed individuals are exposed to different physical and social environments, which result in differences in the cognitive demands they face ( 71 – 74 ). On the one hand, wild great apes face cognitive challenges that are absent in captivity such as the need to locate food sources in the habitat, defend their territories, or hide from predators ( 56 ). On the other hand, zoo-housed great apes grow up in close contact with humans and thus likely receive more varied social stimulation ( 68 ). Furthermore, zoo-housed great apes are provided with carefully designed, at times even computer-assisted enrichment and thus objects that vary in their properties (e.g., shape, size, color, material) and can be manipulated in myriads of ways (e.g. ( 73 , 75 – 77 )). Zoo-housed orangutans thus likely gain a broad range of affordances from EOM, that they may transfer to novel situations (i.e., when they manipulate a novel object or are confronted with novel problems, e.g. ( 78 )). Zoo-housed individuals also have increased amounts of time and energy available for exploration because their food is regularly provided and they do not have to travel to find food or mates. So far, it has remained largely unclear how these factors affect great apes’ cognitive performance, mainly because performing cognitive tests with wild great apes is challenging for a variety of factors such as their high levels of neophobia ( 79 , 80 ) and high protection status due to associated risks such as disease transmission from humans to the test subjects (e.g., ( 81 )). Using EOM as a measure of cognitive performance may help to quantify cognitive differences in wild and captive great apes. Even though the developmental trajectories for the development of the overall frequency of EOM are to some degree hard-wired ( 5 , 33 ), there is evidence that captivity increases overall exploration tendency in great apes (e.g., ( 79 , 82 , 83 )). However, so far, detailed comparisons of the development of EOM in wild versus zoo-housed great apes across ages are lacking. Because individuals can experience cognitive stimulation through EOM throughout the day, to be able to quantify potential differences in cognitive stimulation through EOM, one needs to compare exploration during their daily lives in their habitat (henceforward called “daily EOM behavior”). Because the effects of captivity on cognition may change throughout an individual’s life, it is important to study cognitive differences across a large age range. To date, systematic comparisons of wild and captive animals’ daily EOM behavior and comparisons across large age ranges are missing. Orangutans are the ideal study species to investigate the effect of the environment on exploration as they show a high tendency to explore, which allows for studying the details of the behavior (e.g. ( 5 , 30 )). Furthermore, whilst the mainly arboreal lifestyle of wild orangutans (e.g. ( 84 )) requires them to hold onto branches and thus likely limits exploration, zoo-housed orangutans spend more time on the ground (e.g. ( 85 )) and thus often have both hands free for exploration. Furthermore, because wild orangutans ( Pongo sp. ) are challenging to study because of their low numbers and resulting high protection status, and because their habitats present challenges to human observers, it is crucial to understand whether zoo-housed orangutans differ significantly from wild orangutans. Thus, we compared different aspects of daily EOM behavior in wild and zoo-housed orangutans across all ages and predicted that: P1a) The age trajectories of the overall frequency of EOM are similar in zoo-housed and wild orangutans. P1b) Zoo-housed orangutans of all ages show more frequent EOM and longer EOM bouts than wild orangutans. P2) Zoo-housed orangutans of all ages show a greater variety of actions, use more body parts to explore, and are more likely to explore multiple objects. Furthermore, zoo-housed orangutans show a higher diversity of actions compared to wild orangutans, also when exploring the same objects. P3) The ages at which individuals first show each behavior will be similar across the two settings. In our study, we defined EOM as the prolonged, non-repetitive, usually destructive manipulations of objects (with or without a tool), excluding feeding and nest building, while the visual and tactile foci (manual and/or oral) of the individual are on the object. We analysed a balanced data set containing 12’000 EOM events of wild and zoo-housed Sumatran orangutans. Because most EOM happens during the immature period (below the age of 16)( 5 ) and to prevent overall trends across all ages from overwriting effects during the immature period, we present separate analyses on the immature period in the supplementary information. Results 1a) Developmental trajectory of EOM frequency Age-dependent EOM rates significantly differed between wild and zoo-housed orangutans (full-null model comparison, likelihood ratio test (LRT): χ2 = 114.362, df = 4, P < 0.001), and there was a significant interaction between setting and age squared (χ2 = 7.410, df = 1, P = 0.006), indicating setting-specific age effects; SI, section D1, Table S5). Inspecting the data and the fitted model revealed that zoo-housed individuals generally explored more frequently than wild individuals. Furthermore, in wild orangutans, EOM rates peaked at around two to three years, whereas in zoo-housed orangutans, statistically no such peak was detected (Fig. 1 a). However, notably, the fitted model was associated with large uncertainly (wide confidence intervals in Fig. 1 a) and revealed large fitted values for 0.5 to 2-year-old zoo animals. 1b) Duration of EOM The age-dependent EOM duration per EOM event was not significantly different between settings as indicated by a non-significant full-null model comparison (LRT: χ2 = 4.839, df = 3, P = 0.184, see Fig. 1 b and SI, D, Table S6 for more details; median zoo : 30s [10s, 80s], median wild : 30s [15s, 60s]). See SI, D, 1b.2 for the results of the models on EOM duration in immature individuals only, where also no significant difference between the settings was found. 2a) EOM diversity: actions When looking at exploration events including all object types, the age-dependent number of exploratory actions performed per EOM event differed significantly between wild and zoo-housed orangutans (LRT: χ2 = 17.019, df = 3, P < 0.001). However, the interaction between setting and age squared was not significant in this model (χ2 = 1.251, df = 1, P = 0.263; SI, section D2a, Table S8). After the removal of this non-significant interaction, we also did not find a significant effect of age squared (χ2 = 1.525, df = 1, P = 0.217; SI, section D2a, Table S9) which indicated linear age trajectories and led us to remove age squared from the model. In the resulting model, also the interaction between age and setting was not significant (χ2 = 0.049, df = 1, P = 0.825; SI, section D2a, Table S10) and we thus removed it from the model. The final model revealed significant effects of setting (χ2 = 15.620, df = 1, P < 0.001; SI, section C3a, Table S11) and age (χ2 = 7.140, df = 1, P = 0.008; SI, section D2a, Table S11). The number of exploratory actions per EOM event increased with age in a similar fashion in wild and zoo-housed orangutans, but zoo-housed individuals performed significantly more exploratory actions per event (Fig. 2 a; median number of actions zoo = 2 [1, 3], median number of actions wild = 2 [1, 2]). The results of the models on exploration action diversity in immature individuals only showed a significant difference in the same direction between the settings (SI, section D, 2a.2). 2b) EOM action diversity: EOM events involving objects available in both settings When looking at exploration events including natural objects that occur at both settings, the age-dependent number of exploratory actions performed per EOM event was significantly different between zoo-housed and wild orangutans (LRT: χ2 = 10.000, df = 3, P = 0.019; see SI, D, S16). After the iterative removal all non-significant effects involving interactions and age squared (see SI, D, Table S16-S18) we found significant effects of setting (χ2 = 8.462, df = 1, P = 0.004; see SI, D, Table S19) and age (χ2 = 5.299, df = 1, P = 0.021; see SI, D, Table S19) in that zoo-housed orangutans used more exploratory actions when exploring objects that are available at both settings (see Fig. 2 b). Accordingly, the average number of exploratory actions was higher in zoo-housed than in wild orangutans (median number of actions zoo = 2 [1, 3], median number of actions wild = 2 [1, 2]). See SI, D, 2b.2 for the results of the models on exploration action diversity in immature individuals only, where a significant difference between the settings in the same direction was found. 2c) EOM diversity: body parts The age-dependent trajectory of the body parts used per EOM event was significantly different between wild and zoo-housed orangutans (LRT: χ2 = 11.935, df = 3, P = 0.008; SI, section D2c, Table S24). The interaction between age squared and setting was not significant in this model (χ2 = 0.042, df = 1, P = 0.838). After the removal of this interaction, the final model revealed a significant interaction between age and setting (χ2 = 8.600, df = 1, P = 0.003; SI, D2c, Table S24) and a significant effect of age squared (χ2 = 11.304, df = 1, P = 0.001; SI, D2c, Table S25). Whilst in wild individuals the number of body parts used to explore peaked at the age of four to eight, in zoo-housed orangutans it peaked at about 16 years of age (see Fig. 2 c; median number of body parts zoo : 1 [1, 2]; median number of body parts wild : 1 [1, 2]). See SI, D2c. 2 for the results of the models on exploration body part diversity in immature individuals only, where a significant difference in the same direction between the settings was found. 2d) EOM diversity: objects There was a significant difference between the settings in the age-dependent probability that EOM events involved more than one object (LRT: χ2 = 47.983, df = 3, P < 0.001). The interaction between age squared and setting was significant in this model (χ2 = 5.924, df = 1, P = 0.015); for the model results see SI, section D4.1, Table S28). In zoo-housed orangutans, the probability that EOM events involved multiple objects increased until the age of 16 and then decreased, while overall the probability of EOM events involving multiple objects was higher in zoo-housed compared to wild orangutans (see Fig. 2 d). Wild orangutans showed an overall low probability to involve multiple objects in their exploration throughout all ages. Among the immature individuals, we found a significant difference in the same direction between the settings for the probability that EOM events involved more than one object (SI, D4.2). 3) Age of first occurrence of specific EOM We did not find a significant difference between settings in the age at which individuals first showed exploratory actions that occurred at both settings (paired-samples t-test: t( 60 ) = − 0.8112, p = 0.421; see Fig. 3a). However, overall zoo-housed orangutans showed a larger repertoire of EOMs (n = 123) than wild orangutans (n = 63; for a detailed list see SI, section D3a, Table S32). Additionally, we also found a larger number of tool-orientated EOM in zoo-housed individuals compared to wild individuals (n zoo =13 vs. n wild =3, Table S32, Fig. 3a (indicated by circles around the respective dots)). Almost all EOM were observed for the first time below the age of ten years in both settings (see Fig. 3b). The cumulative number of EOM as a function of age in zoo orangutans showed a higher asymptote (estimate = 0.472, SE = 0.129, z = 3.654, p < 0.001; see Fig. 3b and SI, section D3b, Table S33) but the slope of the increase did not differ significantly between settings (estimate=-0.044, SE = 0.210, z=-0.210, p = 0.834; see Fig. 3b and SI, section D3b, Table S33). However, note that cumulative repertoire sizes were determined pooled across all individuals for the analysis, and therefore do not allow for conclusions on the level of individuals’ repertoire development. Discussion To assess potential effects of the zoo environment on cognition, we investigated differences in EOM between zoo-housed and wild orangutans across a wide age range. Taken together, we found overall similar age-dependent patterns of daily EOM rates for zoo-housed and wild orangutans. However, compared to wild orangutans, zoo-housed immature orangutans showed the highest EOM rates later in development and continued to show higher EOM rates throughout late immaturity and adulthood. We did not find a difference in individuals’ EOM duration between the two settings. Zoo-housed orangutans used a greater variety of exploratory actions and body parts to explore. Furthermore, zoo-housed orangutans showed more tool-orientated behaviors and a higher probability of EOM involving multiple objects. Despite larger overall repertoires of exploratory actions at the zoos, there was no difference between the settings as to at what age different exploratory actions were first shown. Similar to studies in human and non-human primates ( 3 , 5 , 24 – 26 , 33 ), we found overall similar age trajectories of EOM rates (P1a) with a steep increase during the early dependency period and a subsequent decrease in both settings. This similarity across the vastly different settings suggests that the age-dependent motivation to engage in EOM is not affected by the immediate socio-ecological conditions nor the conditions experienced during development and thus may have hard-wired components. Furthermore, in both settings, EOM became more diverse with increasing age (P2), which is indicative of learning ( 29 ). This pattern suggests that just as human infants, wild and zoo-housed immature orangutans learn about objects through EOM. As predicted (P1b), zoo-housed orangutans showed more frequent EOM at all ages in their daily lives which is in line with previous studies in captive great apes (e.g., ( 68 , 79 )). Living in safe and stable captive conditions, animals may experience increased free time and energy to devote to EOM while not being occupied with time-intensive foraging or vigilance (e.g., (82, 87)). Wild great apes have been found to be less curious than zoo-housed great apes ( 79 ), which is commonly attributed to differences in levels of neophobia (88). Whilst studies on curiosity commonly measure individuals’ reactions to novel objects, our study looked at EOM performed with the orangutans’ standard enrichment (the enrichment scheme mostly entails objects that are provided regularly; for an overview see SI, table S1 b). The difference in EOM tendencies of familiar objects between the settings cannot be caused by differences in neophobia and may thus indicate that the difference in curiosity found between wild and captive orangutans may also stem from a heightened tendency to explore in zoo-housed orangutans. Overall, our findings are in line with a study on wild orangutans’ curiosity which showed that visual exploration of novel objects increases with increasing food availability (89). The difference in EOM rates between the settings was less pronounced for immatures below four years of age, most likely because in both settings they are still suckling and thus energetically buffered, and their activity budgets are less affected by the need to forage. Interestingly, at the age of eight when wild Sumatran orangutans reach weaning and their younger sibling is born (for Suaq, the average weaning age is 8.1 years, Schuppli unpublished data) their EOM rates dropped to reach near zero values, whereas zoo-housed immatures’ EOM rate peaked despite much shorter interbirth intervals of ca. 5.5 years at zoos (90, 91) and thus earlier weaning. The fact that the drop in EOM frequency coincides with weaning in the wild but not in the zoos supports the theory that exploration is affected by individuals’ need to forage themselves. In contrast to our prediction (second part of prediction P1b), EOM durations did not differ between settings, indicating that, even though wild orangutans overall showed lower rates of EOM, when exploring, orangutans at both settings were equally persistent in doing so. As we predicted, zoo-housed orangutans showed more diverse EOM (in terms of number of actions performed and body parts used to explore; P2) than their wild counterparts. In undisturbed wild habitats (such as the Suaq research area), orangutans encounter a stable array of natural objects, ranging from tree substrate (e.g., bark, leaves, etc.) to epiphytes, fungi, as well as bird nests. While these objects can be inherently complex and their configurations vary, they only allow for certain exploratory behaviors. Modern zoos, including the zoos where we collected our data, provide specifically designed enrichment consisting of objects of different shapes and materials which likely allows for a large range of exploratory actions (e.g., ( 73 , 75 – 77 )). Interestingly, when exploring natural objects that occur at both settings, zoo-housed individuals also showed a greater diversity of exploratory actions (second part of prediction P2). The latter result implies that the difference in EOM diversity across settings is more than a direct result of differences in the objects that are being explored. Instead, this suggests that individuals of the two settings differ in the set of affordances (i.e., the range of actions) that they can draw from when exploring, most likely because they transfer their repertoires of actions across contexts, i.e., explored objects. Ultimately, these differences in affordances may also affect cognitive performance in other domains, such as problem-solving. Indeed, studies show that zoo-housed animals often show better task performance in problem-solving experiments than their wild counterparts (( 70 ), but see e.g. (92)), a phenomenon known as the ‘captivity bias’ ((87, 93). This is also reflected by higher rates of innovative behavior in captivity compared to wild settings (94–97). However, aside from being the result of opportunities to explore different objects, differences in cognitive performance between wild and zoo-housed individuals may also be caused by social factors. Studies have shown that great apes that have experienced extensive interactions with humans exhibit enhanced cognitive performance in the physical (64, 98, 99) and in the social domain (e.g. ( 67 ) but see (100)). An experiment with captive and rehabilitated orangutans with different rearing backgrounds found that the degree of human orientation predicted the motivation to explore an apparatus which affected their problem-solving skills ( 68 ), suggesting that the effects of the social and physical environment on cognitive performance interact with each other. To pin down in what aspects and to what extent differences in daily EOM behavior translate into differences in cognitive performance, systematic standardized cognitive tests in both settings are needed which are to date lacking for great apes (but see, e.g., (70, 101) for findings on other species). Zoo-housed orangutans also showed a higher probability of exploring multiple objects in one event (P2). Zoo enrichment likely allows for more complex and sustained combinatory object-object and object-substrate manipulations than naturally occurring objects ( 73 , 75 – 77 ). Combinatory object manipulations seem to be linked to abilities in the domain of physical cognition, as tool-using species show the most complex structured object combinations (e.g. corvids and parrots ( 44 ), primates ( 49 )). Combinatory actions have also been shown to be relevant to the development of tool-use abilities in humans (102). Even though we found evidence for more multi-object manipulations in zoo-housed orangutans, we could not investigate the exact nature of these manipulations because our data only entail information on how many distinct objects were explored (but not how many objects of the same type). Therefore, we do not know whether the actions performed were combinatory or not. However, our results showed that zoo-housed orangutans have a larger repertoire of tool-orientated behaviors than wild orangutans. The greater availability of objects that allow for more complex, including more combinatory manipulations, may stimulate cognitive development ( 71 – 74 ). Most exploratory actions were first shown by individuals below the age of ten years in both settings. Furthermore, there was no difference between the settings in the ages of first occurrence for exploratory actions that were present at both settings (P3), which supports the conclusion of previous studies that the neurodevelopmental sequence of manipulative skills is largely fixed in primates (see ( 52 , 53 )). However, zoo-housed orangutans showed an overall higher number of exploratory actions than wild orangutans (P2) which could be a result of having different object types available to explore. Interestingly, several of our findings showed that certain aspects of EOM change throughout life, thus developmental effect on EOM are not restricted to the immature period. In both settings, body part diversity during exploration decreased after a certain age whereas action diversity increased with age. This may suggest that orangutans become more skilled as they age and therefore can perform a more diverse range of actions while using a smaller number of body parts. If EOM is directly linked to cognitive performance, our results suggest that cognitive performance, similarly as in humans (e.g., (103, 104)), in great apes also changes throughout adulthood. Conclusion Taken together, our results show that the zoo environment significantly affects EOM in orangutans as well as that aspects of EOM and the effect of captivity on EOM change throughout life. Our findings also strengthen existing evidence that EOM reflects cognitive development and might function as a means of cognitive stimulation, not just in humans but across great apes. The comparison across settings suggests that the conditions that individuals experience at the zoo amplify their daily EOM behavior. Given the link between EOM and certain areas of cognitive performance (e.g. problem-solving and innovation probability ( 36 – 43 )), our findings add to the studies that suggest that in some aspects, cognitive performance measured in zoo-housed individuals may not be directly comparable to cognitive performance of their wild counterparts ( 64 – 67 ). On the one hand, this calls for caution when drawing conclusions on the evolutionary level, such as on the evolution of cognition. This is especially true for studies that compare cognitive performance across species by testing in captivity because the strength of the effect of captivity on EOM and cognitive performance may vary across different species. On the other hand, differences in EOM and cognitive performance between captive and wild great apes allow for important novel insights into the proximate factors affecting cognition which are in themselves informative about the evolution of cognition. Furthermore, captive studies may help to elucidate a species' cognitive potential, i.e., the level of cognitive abilities individuals can develop upon the evolved cognitive potential they are born with. Methods Subjects The data on wild Sumatran orangutans were collected from 2007 to 2020 at the Suaq Balimbing research area located in the Gunung Leuser National Park in South Aceh, Indonesia (3°42’N, 97°26’E). The data on zoo-housed Sumatran orangutans were collected from 2021 to 2023 at Leipzig Zoo and Dresden Zoo in Germany, and Zoo Zurich and Basel Zoo in Switzerland (for information on enclosure size and enrichment schedule see SI, Table S1 b). All-occurrence event data (that were used for models 2–4; see below) included 6478 EOM events performed by 33 wild orangutans on 363 observation days between the ages of 0.5 and 76.7 years at Suaq Balimbing in Indonesia and 6640 EOM events performed by 24 zoo-housed orangutans on 134 observation days between the ages of 0.6 and 49.4. Note that sample sizes varied for the different analyses, as not all details were reported for all events (for subject information see SI, section A, Table S1 a). Note that a subset of the EOM data in wild immature orangutans (3200 EOM events collected on 13 immatures between the ages of 0.5 and 13 years) has already been published previously ( 5 ). Following van Noordwijk et al. 2018, we call individuals below the age of 16 “immatures”, below the age of 8 years “dependent immatures” (at Suaq the average weaning age is 8.1 years; Schuppli, unpublished data), and between the age of 8 and 16 years “independent immatures”. Individuals above the age of 16 are referred to as “adults”. Ethical guidelines The data collection was purely observational and without any interaction with the study animals. The research protocols for the data collection in the wild were approved by the Ministry of research and technology (RISTEK; research permit no. 152/ SIP/FRP/SM/V/2012 and following permits) and complied with the legal requirements of Indonesia. Data collection at the zoos did not lead to any deviation from the usual husbandry protocols, was approved by the ethics joint committee of the Max Planck Institute for Evolutionary Anthropology, Leipzig Zoo, and was carried out in accordance with relevant guidelines and regulations at Leipzig Zoo, Dresden Zoo, Basel Zoo, and Zoo Zurich. Our research protocols complied with the EAZA minimum standards for the accommodation and care of animals in zoos and aquaria. The research adhered to all German and Swiss laws regarding animal keeping. Data collection procedure The data were collected following an established protocol for orangutan behavioral data collection(ab.mpg.de/571325/standarddatacollectionrules_suaq_detailed_jan204.pdf ) which consisted of a) instantaneous scan sampling at two-minute intervals of the individual’s activity and b) all-occurrence sampling of behaviors of interest, including EOM. We collected the data on wild orangutans during focal animal follows, of which 50.9% were full-day follows (i.e., from when the focal animal got out of its nest in the morning until it went into its evening nest) and 49.1% partial follows (when the focal animal was either found and then followed until the evening nest or followed from the morning nest until it was lost). The zoo data were collected from the moment the orangutans entered their enclosure in the morning until they left it in the evening (or from the beginning until the end of the visitor hours, in case the orangutans remained in the enclosure for 24 hours). The zoo-housed orangutans were observed from the visitors’ area or an observation tower (at Leipzig Zoo), which provides a view of the indoor enclosure from above. EOM was defined as the prolonged, non-repetitive, usually destructive manipulations of objects (with or without a tool), while the visual and tactile foci (manual and/or oral) of the individual are on the object (105) (see movie S1 for an example of EOM in the wild and in zoo-housed orangutans). Following Pisula (106), we excluded feeding with or without a tool (defined as such by actual ingestion), but included object play (i.e., manipulations during which the visual and tactile foci were desynchronized, and were characterized by repetitive movements), as it is at times difficult to reliably distinguish exploratory from non-exploratory object play (but see (107, 108)). Further we did not include nesting behavior. The exception was if the orangutan was deconstructing a nest or was manipulating twigs that may have been unsuccessful attempts to start constructing a nest. This was done as it is hard to judge the individual’s intention in these instances. We coded the duration of EOM events from the start of the manipulation to the time the orangutan stopped manipulating. To be counted as two separate EOM events, there had to be at least ten seconds between two EOM events during which no EOM was observed.Data were collected by 25 experienced observers (see SI, section A2 for details on inter-observer reliability). Data preparation To investigate EOM frequencies, we counted the number of EOM events per observation day (model 1a). We included data from 65 wild and 24 zoo-housed orangutans (collected on 650 and 142 observation days respectively, note that we also included days where the EOM rate was zero). To investigate EOM duration (model 1b), we used the duration of all EOM events that were timed. To investigate exploratory action diversity (model 2a), we used the number of different exploratory actions performed during EOM events (SI, section B, Table S2 a for a list of all explorative actions and definitions). We also assessed exploratory action diversity for EOM events that included natural objects only that occurred in both settings (SI, section B, Table S2 b; model 2b). To investigate the diversity of body parts involved during EOM (model 2c), we used the number of body parts involved (see SI, section B, Table S3 for a list of all body parts). To assess EOM object diversity (model 2d), we coded whether or not multiple distinct objects (i.e., two or more) were manipulated during an event (see SI, Table S2 b). Statistical analysis To assess how the age-dependent development of various aspects of EOM differed between settings, we fitted general and Generalized Linear Mixed Models (LMM and GLMM; (109)) using a significance criterium of P < 0.05. In all models, we included setting and age as fixed effects predictors. To account for the possibility that EOM rates peaked at intermediate ages, we also included age squared into the fixed effects part of each model. As the age-dependent trajectory of EOM rate could vary between settings, we also included the interaction between age and setting and age squared and setting in the fixed effects part of all models. We further included random intercepts effect of the individual ID, and in models 1b – 2d we included follow number (i.e., the id of the observation day) nested in subject as a further random intercepts effect (as in the all-occurrence data, events associated with individuals other than the focal of the follow were recorded which resulted in data on multiple individuals under the same follow number). When identifiable, to account for individual variation in age trajectories and to avoid overconfident models and increased type I error rates, we also included random slopes (110, 111) of age and age squared. To assess the overall effects of the predictors, we first compared each full model with a respective null model which lacked setting and its interactions with age and age squared in the fixed effects part (also in the zero inflation part in model 1a) but was otherwise identical to the full model via a likelihood ratio test (LRT, (112)). In case a full-null model comparison revealed significance, we first determined the significance of the interaction between setting and age squared, and if this did not reveal significance, we removed it from the model and then checked the significance of the effect age squared. If this also did not reveal significance, we removed it from the model, too, and checked the significance of the interaction between age and setting. In case this interaction did not reveal significance, we also removed it from the model (see (113)). This iterative removal of non-significant interactions served in obtaining unconditional estimates of the respective lower-order terms (114). 1. EOM frequency and duration To investigate differences in age-dependent EOM rates between the settings (model 1a), we used the total daily number of EOM events as the response and the observation time during which orangutans were visible, divided by six and then log-transformed as an offset term (the mean follow duration was roughly six hours; dividing by six was done to ease model convergence; (115)). To avoid zero-inflation and overdispersion issues, we chose a model with a zero-inflated negative binomial error structure. In the zero-inflation part, we included the fixed effects of species, age, and age squared. The model comprised random slopes of age and age squared within individual, but we had to drop parameters for the correlations among random intercepts and slopes as a model including them did not converge (for final full model structure and sample sizes see SI, section C, Table S4a, b). To compare age-dependent differences in EOM duration between the settings (model 1b), we used a linear mixed effects model (LMM) fitted with Gaussian error distribution and identity link. To achieve roughly normally distributed and homogeneous residuals (assessed by visual inspection of a qq-plot and residuals plotted against fitted values) we log-transformed (base e) the response (EOM duration). 2) EOM diversity – actions, body parts, and objects used To investigate differences in the age-dependent development in EOM diversity (model 2a-2d), we used GLMMs with the number of different exploratory actions used during EOM (exploratory actions that included all object types (model 2a), exploratory actions during EOM events that involve only natural objects occurring at both settings (model 2b), the number of body parts used during EOM (model 2c), and whether or not multiple objects were explored during an event (i.e. one object versus more than one object, coded as 0 and 1; model 2d) as a response variable. For models 2a – 2d, the response variable was the number of unique elements per EOM event. Since the response variable in models 2a – 2c did not comprise zeros, we fitted these models with a zero-truncated negative binomial error distribution and log link function. All four models comprised random slopes of age and age squared within individual. Model 2b also included the object as a random intercepts effect because it aimed at comparing EOM action diversity within the same object between the two settings. Initially, we also included parameters for the correlations among random intercepts and slopes. However, in each model we either had one or several absolute correlation parameters being close to one (which is indicative of them being unidentifiable (116) or the model did not converge when we included them. We hence report results for models not comprising such correlation parameters. For the final full model structure of models 2a – 2d and the respective sample sizes see SI, section C, Table S4. 3) Age of first occurrence of specific object EOM To model the cumulative repertoire of exploratory behaviors as a function of age (model 3) we used a specifically tailored model. The data analyzed with this model consisted of the data aggregated such that they were a series of cumulative repertoire sizes and the ages at which these were reached (e.g., 12 unique behaviors at an age of 1.62 years, 28 behaviors at an age of 1.76 years, etc.) separately for wild and zoo-housed individuals but pooled across all individuals of a given setting). To these data we then fitted an exponential function (for detailed information see SI, section D3). We assumed R to be Poisson distributed given the fitted model (for sample sizes see SI, section C, Table S4a,b). 4) General considerations We fitted all models in R (version 4.3.1 or higher; (117)), using the function glmmTMB of the equally named package (version 1.1.8; (Brooks et al., 2017) for models 1a, 2a, 2b, and 2c; the function lmer of the package lme4 (version 1.1–34; (118)) for model 1b and the function glmer of the same package for model 2d. Before fitting the models, we log- (base e) and then z-transformed age to a mean of zero and a standard deviation of one to ease model convergence. In the case of models 1a and 2a-d, we determined the significance of individual fixed effects of the final full model by dropping them from the model and comparing the resulting simplified model with the more complex one using a likelihood ratio test (R function drop1). For model 1b, we determined the significance of individual fixed effects of the final full model by means of the Satterthwaite approximation to degrees of freedom (function lmer of the package lmerTest; version 3.1-3; (119); see also (120)). We estimated 95% confidence intervals of model estimates and fitted values by means of parametric bootstraps (N = 1000 bootstraps; function simulate of the package glmmTMB or function bootMer of the package lme4). For information on model stability, collinearity and overdispersion checks, see SI, section A3. For models 1b-2d, we fitted equivalent models comprising only data collected when individuals were less than 16 years old. This was done because fine-grained effects during this period of development can be overwritten by overall trends across all ages. The details and results of these models are shown in the SI, section C and D. We report information on median and 1st and 3rd quartile throughout the manuscript like this: median [1st quartile, 3rd quartile]. Declarations Author contributions Conceptualization: CS Data collection at Suaq Balimbing: CS, AvC, TR Data collection at the Zoos: AvC, SK, IBL Data entry: CS, AvC, SK Data analysis: IBL, RM, CS Writing of the manuscript: IBL Editing the manuscript: CS Commenting on the manuscript: RM, DH, TMS, TR Research Administration & Planning: TMS, TR Funding statement This project was funded by a Freigeist Fellowship of the Volkswagen Foundation and a Postdoc Mobility Grant by the Swiss National Science Foundation (grant number P400PM_180844) to CS. Funding was also obtained from the Max Planck Institute for Animal Behavior, the University of Zürich, the SUAQ Foundation, the Stiftung Mensch und Tier Freiburg i.Br., the Leakey Foundation (Primate Research Fund), the A.H. Schultz-Foundation, and the Claraz Foundation. Acknowledgements We acknowledge all students, volunteers, and local field assistants involved in the collection of standard behavioral data at Suaq. Special thanks go to Emma Lokuciejewski, Frances Luhn, Francois Lamarque, Glenn Honstetter, Ivan Lenzi, Luise Staatsmann, Mulati Mikeliban, Nele Kater, Nora Slania, Paulina Kukofka, Revathe Thillaikumar, Richard Young, Beatrice Ehmann, Ellen Meulmann, Lara Nellissen, Luz Carvajal, Natalie Oliver-Caldwell, and Sofia Forss, who contributed to collecting the data on EOM that were used in this study. We gratefully acknowledge the Direktur Tata Kelola Perijinan Riset dan Inovasi dan Otoritas Ilmiah (BRIN), the Balai Besar Taman Nasional Gunung Leuser (TNGL), the local government in South Aceh, Director General of Natural Resources and Ecosystem Conservation (KSDAE) Ministry of Environment and Forestry (KLHK) for their permission and support to conduct this research. We thank the Yayasan Ekosistem Lestari (YEL) and its Orangutan Conservation Program (SOCP) for hosting our project at the Suaq Balimbing monitoring station. We thank the Fakultas Biologi of Universitas Nasional (UNAS) in Jakarta for their collaboration and support, in particular, Dr. Sri Utami Atmoko. We thank the curators, staff, and helpers at the zoo for their support in conducting our data collection: Adrian Baumeyer (Zoo Basel), Dr. Claudia Rudolf von Rohr (Zoo Zurich), Dr. Leyla Davis (Zoo Zurich), Dr. Wolfgang Ludwig (Zoo Dresden), Roman Richter (Zoo Dresden), and Matthias Hendel (Zoo Dresden). We thank our colleagues at the Max Planck Institute for Evolutionary Anthropology in Leipzig, and especially Hanna Petschauer and Dr. Daniel Hanus for their support with data collected at Leipzig Zoo. Supplementary material To view general considerations for this article, please visit <LINK>. Data availability The data is available as a supplementary file. Declaration of interests The authors declare no competing interests. References Fenson L, Schell RE. The origins of exploratory play. Early Child Development and Care. 1985;19(1-2):3-24. Muentener P, Herrig E, Schulz L. The efficiency of infants' exploratory play is related to longer-term cognitive development. Frontiers in psychology. 2018;9:635. Power TG. Play and exploration in children and animals: Psychology Press; 1999. Ruff HA, Saltarelli LM, Capozzoli M, Dubiner K. The differentiation of activity in infants' exploration of objects. Developmental psychology. 1992;28(5):851. Schuppli C, Van Cauwenberghe A, Setia TM, Haun D. The ontogeny of exploratory object manipulation behaviour in wild orangutans. Evolutionary Human Sciences. 2021;3:e39. Malachowski LG, Needham AW. Infants exploring objects: a cascades perspective. Advances in child development and behavior. 64: Elsevier; 2023. p. 39-68. Needham A. Learning in infants’ object perception, object-directed action, and tool use. Learning and the infant mind. 2009:208-26. Oakes LM, Baumgartner HA, editors. Manual object exploration and learning about object features in human infants. 2012 IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL); 2012 7-9 Nov. 2012; San Diego, CA, USA: IEEE. Swirbul MS, Herzberg O, Tamis-LeMonda CS. Object play in the everyday home environment generates rich opportunities for infant learning. Infant Behavior and Development. 2022;67:101712. Bondi D, Robazza C, Lange‐Küttner C, Pietrangelo T. Fine motor skills and motor control networking in developmental age. American Journal of Human Biology. 2022;34(8):e23758. Piaget J. The stages of the intellectual development of the child. Bulletin of the Menninger Clinic. 1962;63(26(3)):120–8. Baumgartner HA, Oakes LM. Investigating the relation between infants' manual activity with objects and their perception of dynamic events. Infancy. 2013;18(6):983-1006. Needham A. Improvements in object exploration skills may facilitate the development of object segregation in early infancy. Journal of Cognition and Development. 2000;1(2):131-56. Soska KC, Adolph KE, Johnson SP. Systems in development: motor skill acquisition facilitates three-dimensional object completion. Developmental psychology. 2010;46(1):129. Wilcox T, Woods R, Chapa C, McCurry S. Multisensory exploration and object individuation in infancy. Developmental psychology. 2007;43(2):479. Case-Smith J, Bigsby R, Clutter J. Perceptual-motor coupling in the development of grasp. The American Journal of Occupational Therapy. 1998;52(2):102-10. Lobo MA, Galloway JC. Postural and object‐oriented experiences advance early reaching, object exploration, and means–end behavior. Child development. 2008;79(6):1869-90. Babik I, Galloway JC, Lobo MA. Early exploration of one’s own body, exploration of objects, and motor, language, and cognitive development relate dynamically across the first two years of life. Developmental psychology. 2022;58(2):222. Fagan MK, Iverson JM. The influence of mouthing on infant vocalization. Infancy. 2007;11(2):191-202. Muentener P, Herrig E, Schulz L. The efficiency of infants' exploratory play is related to longer-term cognitive development. Frontiers in psychology. 2018;9:291931. Begus K, Southgate V, Gliga T. Neural mechanisms of infant learning: differences in frontal theta activity during object exploration modulate subsequent object recognition. Biology letters. 2015;11(5):20150041. Orekhova E, Stroganova T, Posikera I, Elam M. EEG theta rhythm in infants and preschool children. Clinical neurophysiology. 2006;117(5):1047-62. Degroot A, Wolff MC, Nomikos GG. Acute exposure to a novel object during consolidation enhances cognition. Neuroreport. 2005;16(1):63-7. Belsky J, Most RK. From exploration to play: A cross-sectional study of infant free play behavior. Developmental psychology. 1981;17(5):630. Bock J. Farming, foraging, and children’s play in the Okavango Delta, Botswana. The nature of play: Great apes and humans. 2005:254-81. Power TG, Chapieski ML, McGrath MP. Assessment of individual differences in infant exploration and play. Developmental Psychology. 1985;21(6):974. Belsky J, Goode MK, Most RK. Maternal stimulation and infant exploratory competence: Cross-sectional, correlational, and experimental analyses. Child development. 1980:1168-78. McCall RB. Exploratory manipulation and play in the human infant. Monographs of the society for research in child development. 1974:1-88. Corbetta D, DiMercurio A, Wiener RF, Connell JP, Clark M. How perception and action fosters exploration and selection in infant skill acquisition. Advances in child development and behavior. 2018;55:1-29. Bard KA. Sensorimotor cognition in young feral orangutans (Pongo pygmaeus). Primates. 1995;36:297-321. Chevalier-Skolnikoff S. Tool use in Cebus: Its relation to object manipulation, the brain, and ecological adaptations. Behavioral and Brain Sciences. 1989;12(3):610-27. Hayashi M, Takeshita H, Matsuzawa T. Cognitive development in apes and humans assessed by object manipulation. Cognitive development in chimpanzees. 2006:395-410. Lamon N, Neumann C, Zuberbühler K. Development of object manipulation in wild chimpanzees. Animal behaviour. 2018;135:121-30. Butler RA. Discrimination learning by rhesus monkeys to visual-exploration motivation. Journal of Comparative and Physiological Psychology. 1953;46(2):95. Guillette LM, Reddon AR, Hurd PL, Sturdy CB. Exploration of a novel space is associated with individual differences in learning speed in black-capped chickadees, Poecile atricapillus. Behavioural Processes. 2009;82(3):265-70. Auersperg AM, Von Bayern AM, Gajdon GK, Huber L, Kacelnik A. Flexibility in problem solving and tool use of kea and New Caledonian crows in a multi access box paradigm. PloS one. 2011;6(6):e20231. Benson-Amram S, Dantzer B, Stricker G, Swanson EM, Holekamp KE. Brain size predicts problem-solving ability in mammalian carnivores. Proceedings of the National Academy of Sciences. 2016;113(9):2532-7. Griffin AS, Guez D. Innovation and problem solving: a review of common mechanisms. Behavioural Processes. 2014;109:121-34. Overington SE, Cauchard L, Côté K-A, Lefebvre L. Innovative foraging behaviour in birds: what characterizes an innovator? Behavioural processes. 2011;87(3):274-85. Reader SM, Laland KN. Social intelligence, innovation, and enhanced brain size in primates. Proceedings of the National Academy of Sciences. 2002;99(7):4436-41. Webster SJ, Lefebvre L. Problem solving and neophobia in a columbiform–passeriform assemblage in Barbados. Animal Behaviour. 2001;62(1):23-32. Damerius LA, Graber SM, Willems EP, van Schaik CP. Curiosity boosts orang-utan problem-solving ability. Animal Behaviour. 2017;134:57-70. Ebel SJ, Call J. The interplay of prior experience and motivation in great ape problem-solving (Gorilla gorilla, Pan paniscus, Pan troglodytes, and Pongo abelii). Journal of Comparative Psychology. 2018;132(3):294. Auersperg AM, Van Horik JO, Bugnyar T, Kacelnik A, Emery NJ, von Bayern AM. Combinatory actions during object play in psittaciformes (Diopsittaca nobilis, Pionites melanocephala, Cacatua goffini) and corvids (Corvus corax, C. monedula, C. moneduloides). Journal of Comparative Psychology. 2015;129(1):62. Vauclair J. Phylogenetic approach to object manipulation in human and ape infants. Human Development. 1984;27(5-6):321-8. Byrne RW, Corp N, Byrne JM. Manual dexterity in the gorilla: bimanual and digit role differentiation in a natural task. Animal Cognition. 2001;4:347-61. Hayashi M, Matsuzawa T. Cognitive development in object manipulation by infant chimpanzees. Animal cognition. 2003;6:225-33. Heldstab SA, Kosonen ZK, Koski SE, Burkart JM, van Schaik CP, Isler K. Manipulation complexity in primates coevolved with brain size and terrestriality. Scientific reports. 2016;6(1):24528. Torigoe T. Comparison of object manipulation among 74 species of non-human primates. Primates. 1985;26:182-94. Redshaw M. Cognitive development in human and gorilla infants. Journal of human Evolution. 1978;7(2):133-41. Vauclair J, Bard KA. Development of manipulations with objects in ape and human infants. Journal of Human Evolution. 1983;12(7):631-45. Heldstab SA, Isler K, Schuppli C, van Schaik CP. When ontogeny recapitulates phylogeny: Fixed neurodevelopmental sequence of manipulative skills among primates. Science advances. 2020;6(30):eabb4685. Workman AD, Charvet CJ, Clancy B, Darlington RB, Finlay BL. Modeling transformations of neurodevelopmental sequences across mammalian species. Journal of Neuroscience. 2013;33(17):7368-83. Wittig RM, Crockford C. Chimpanzees–investigating cognition in the wild. Field and laboratory methods in animal cognition. 2018:119e45. Bohn M, Eckert J, Hanus D, Lugauer B, Holtmann J, Haun DB. Great ape cognition is structured by stable cognitive abilities and predicted by developmental conditions. Nature Ecology & Evolution. 2023:1-12. Boesch C. What makes us human (Homo sapiens)? The challenge of cognitive cross-species comparison. Journal of Comparative Psychology. 2007;121(3):227. Nelson III CA, Zeanah CH, Fox NA, Marshall PJ, Smyke AT, Guthrie D. Cognitive recovery in socially deprived young children: The Bucharest Early Intervention Project. Science. 2007;318(5858):1937-40. Bauer PM, Hanson JL, Pierson RK, Davidson RJ, Pollak SD. Cerebellar volume and cognitive functioning in children who experienced early deprivation. Biological psychiatry. 2009;66(12):1100-6. Reyes-García V, Pyhälä A, Díaz-Reviriego I, Duda R, Fernández-Llamazares Á, Gallois S, et al. Schooling, local knowledge and working memory: A study among three contemporary hunter-gatherer societies. PloS one. 2016;11(1):e0145265. Eluvathingal TJ, Chugani HT, Behen ME, Juhász C, Muzik O, Maqbool M, et al. Abnormal brain connectivity in children after early severe socioemotional deprivation: a diffusion tensor imaging study. Pediatrics. 2006;117(6):2093-100. Sánchez MM, Hearn EF, Do D, Rilling JK, Herndon JG. Differential rearing affects corpus callosum size and cognitive function of rhesus monkeys. Brain research. 1998;812(1-2):38-49. Schrijver NC, Pallier PN, Brown VJ, Würbel H. Double dissociation of social and environmental stimulation on spatial learning and reversal learning in rats. Behavioural brain research. 2004;152(2):307-14. Maggi S, Irwin LJ, Siddiqi A, Hertzman C. The social determinants of early child development: an overview. Journal of paediatrics and child health. 2010;46(11):627-35. Furlong E, Boose K, Boysen S. Raking it in: the impact of enculturation on chimpanzee tool use. Animal cognition. 2008;11:83-97. Yocom AM, Boysen ST. Comprehension of functional support by enculturated chimpanzees Pan troglodytes. Current Zoology. 2011;57(4):429-40. Russell JL, Lyn H, Schaeffer JA, Hopkins WD. The role of socio‐communicative rearing environments in the development of social and physical cognition in apes. Developmental science. 2011;14(6):1459-70. Tomasello M, Savage‐Rumbaugh S, Kruger AC. Imitative learning of actions on objects by children, chimpanzees, and enculturated chimpanzees. Child development. 1993;64(6):1688-705. Damerius LA, Forss SI, Kosonen ZK, Willems EP, Burkart JM, Call J, et al. Orientation toward humans predicts cognitive performance in orang-utans. Scientific Reports. 2017;7(1):40052. Forss S, Motes-Rodrigo A, Hrubesch C, Tennie C. Chimpanzees’(Pan troglodytes) problem-solving skills are influenced by housing facility and captive care duration. PeerJ. 2020;8:e10263. Benson-Amram S, Weldele ML, Holekamp KE. A comparison of innovative problem-solving abilities between wild and captive spotted hyaenas, Crocuta crocuta. Animal Behaviour. 2013;85(2):349-56. Clark FE. Cognitive enrichment and welfare: Current approaches and future directions. Anim Behav Cogn. 2017;4(1):52-71. Meehan CL, Mench JA. The challenge of challenge: can problem solving opportunities enhance animal welfare? Applied Animal Behaviour Science. 2007;102(3-4):246-61. Tarou LR, Kuhar CW, Adcock D, Bloomsmith MA, Maple TL. Computer-assisted enrichment for zoo-housed orangutans (Pongo pygmaeus). Animal Welfare. 2004;13(4):445-53. Milgram NW, Siwak-Tapp CT, Araujo J, Head E. Neuroprotective effects of cognitive enrichment. Ageing research reviews. 2006;5(3):354-69. Tripp JK. Increasing activity in captive orangutans: provision of manipulable and edible materials. Zoo biology. 1985;4(3):225-34. Shariman P, Ruppert N. Effect of environmental enrichment on activities of captive orangutans at Taiping Zoo, Malaysia. Malayan Nature Journal. 2017;69(4):327-35. Scheel B, editor Designing digital enrichment for orangutans. Proceedings of the Fifth International Conference on Animal-Computer Interaction; 2018. Hacques G, Komar J, Seifert L. Learning and transfer of perceptual-motor skill: Relationship with gaze and behavioral exploration. Attention, Perception, & Psychophysics. 2021;83:2303-19. Forss SI, Schuppli C, Haiden D, Zweifel N, Van Schaik CP. Contrasting responses to novelty by wild and captive orangutans. American Journal of Primatology. 2015;77(10):1109-21. Kalan AK, Hohmann G, Arandjelovic M, Boesch C, McCarthy MS, Agbor A, et al. Novelty response of wild African apes to camera traps. Current Biology. 2019;29(7):1211-7. e3. Dunay E, Apakupakul K, Leard S, Palmer JL, Deem SL. Pathogen transmission from humans to great apes is a growing threat to primate conservation. EcoHealth. 2018;15:148-62. Kummer H, Goodall J. Conditions of innovative behaviour in primates. Philosophical Transactions of the Royal Society of London B, Biological Sciences. 1985;308(1135):203-14. Kluen E, Rönkä K, Thorogood R. Prior experience of captivity affects behavioural responses to ‘novel’environments. PeerJ. 2022;10:e13905. Ashbury AM, Posa MRC, Dunkel LP, Spillmann B, Atmoko SSU, van Schaik CP, et al. Why do orangutans leave the trees? Terrestrial behavior among wild Bornean orangutans (Pongo pygmaeus wurmbii) at Tuanan, Central Kalimantan. American journal of primatology. 2015;77(11):1216-29. Pearson EL, Davis JM, Litchfield CA. A case study of orangutan and siamang behavior within a mixed-species zoo exhibit. Journal of Applied Animal Welfare Science. 2010;13(4):330-46. Ashbury A.M. LF, Permana A., Rahmaeti T., Utami-Atmoko S., Crofoot M.C., Schuppli C. Wild orangutans maintain sleep homeostasis through napping, counterbalancing socioecological factors that interfere with their sleep. Current Biology. in revision;https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5131665. Haslam M. ‘Captivity bias’ in animal tool use and its implications for the evolution of hominin technology. Philosophical Transactions of the Royal Society B: Biological Sciences. 2013;368(1630):20120421. Forss SI, Koski SE, van Schaik CP. Explaining the paradox of neophobic explorers: the social information hypothesis. International Journal of Primatology. 2017;38:799-822. Schuppli C, Nellissen L, Carvajal L, Ashbury AM, Oliver-Caldwell N, Rahmaeti T, et al. Ecological, social, and intrinsic factors affecting wild orangutans’ curiosity, assessed using a field experiment. Scientific Reports. 2023;13(1):13184. Anderson HB, Thompson ME, Knott CD, Perkins L. Fertility and mortality patterns of captive Bornean and Sumatran orangutans: is there a species difference in life history? Journal of Human Evolution. 2008;54(1):34-42. Shumaker RW, Wich SA, Perkins L. Reproductive life history traits of female orangutans (Pongo spp.). Primate reproductive aging. 36: Karger Publishers; 2008. p. 147-61. McCune KB, Jablonski P, Lee S-i, Ha RR. Captive jays exhibit reduced problem-solving performance compared to wild conspecifics. Royal Society open science. 2019;6(1):181311. Meulman EJ, Sanz CM, Visalberghi E, van Schaik CP. The role of terrestriality in promoting primate technology. Evolutionary Anthropology: Issues, News, and Reviews. 2012;21(2):58-68. Motes‐Rodrigo A, Tennie C. Captive great apes tend to innovate simple tool behaviors quickly. American Journal of Primatology. 2022;84(10):e23311. Lehner SR, Burkart JM, van Schaik CP. An evaluation of the geographic method for recognizing innovations in nature, using zoo orangutans. Primates. 2010;51:101-18. Manrique HM, Völter CJ, Call J. Repeated innovation in great apes. Animal Behaviour. 2013;85(1):195-202. van Schaik CP, Burkart J, Damerius L, Forss SI, Koops K, van Noordwijk MA, et al. The reluctant innovator: orangutans and the phylogeny of creativity. Philosophical Transactions of the Royal Society B: Biological Sciences. 2016;371(1690):20150183. Bard KA, Gardner KH. Influences on development in infant chimpanzees: Enculturation, temperament, and cognition. 1996. Gardner RA, Gardner BT, Van Cantfort TE. Teaching sign language to chimpanzees: Suny Press; 1989. Tomasello M, Call J. The role of humans in the cognitive development of apes revisited. Animal Cognition. 2004;7:213-5. Rössler T, Mioduszewska B, O’Hara M, Huber L, Prawiradilaga DM, Auersperg AM. Using an Innovation Arena to compare wild-caught and laboratory Goffin’s cockatoos. Scientific Reports. 2020;10(1):8681. Connolly K, Dalgleish M. The emergence of a tool-using skill in infancy. Developmental Psychology. 1989;25(6):894. Salthouse TA. Continuity of cognitive change across adulthood. Psychonomic bulletin & review. 2016;23:932-9. Murman DL, editor The impact of age on cognition. Seminars in hearing; 2015: Thieme Medical Publishers. Schuppli C, Van Cauwenberghe A, Mirta Setia T, Haun D. The ontogeny of exploratory object manipulation behaviour in wild orangutans. Evolutionary Human Sciences. 2021:1-32. Pisula W. Play and exploration in animals—a comparative analysis. Polish Psychological Bulletin. 2008(2). Hutt C, editor Exploration and play in children. Symposia of the Zoological Society of London; 1966: London. Pellegrini AD, Gustafson K. Boys’ and girls’ uses of objects for exploration, play, and tools in early childhood. The nature of play: Great apes and humans. 2005:113-35. Baayen R. Analyzing linguistic data. Cambridge, UK. Cambridge University Press New York; 2008. Barr DJ, Levy R, Scheepers C, Tily HJ. Random effects structure for confirmatory hypothesis testing: Keep it maximal. Journal of memory and language. 2013;68(3):255-78. Schielzeth H, Forstmeier W. Conclusions beyond support: overconfident estimates in mixed models. Behavioral ecology. 2009;20(2):416-20. Dobson A. An Introduction to Generalized Linear Models, Baton Rouge, LA. Chapman and Hall/CRC Press; 2002. Aiken L. Multiple regression: Testing and interpreting interactions. Sage Publications google schola. 1991;2:513-31. Aiken LS, West SG, Reno RR. Multiple regression: Testing and interpreting interactions: sage; 1991. McCullagh P, Nelder J. 1989: Generalized linear models, London: Chapman and Hall. 1989. Matuschek H, Kliegl R, Vasishth S, Baayen H, Bates D. Balancing Type I error and power in linear mixed models. Journal of memory and language. 2017;94:305-15. Team RC. R: a language and environment for statistical computing. R Foundation for statistical Computing, Vienna, Austria [Methodology Reference]. European Environment Agency https://www R-project org/ Accessed. 2023;12. Bates D, Mächler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. arXiv preprint arXiv:14065823. 2014. Kuznetsova A, Brockhoff PB, Christensen RH. lmerTest package: tests in linear mixed effects models. Journal of statistical software. 2017;82:1-26. Luke SG. Evaluating significance in linear mixed-effects models in R. Behavior research methods. 2017;49:1494-502. Additional Declarations No competing interests reported. Supplementary Files Data.xlsx SupplementaryinformationLaumeretal..docx MovieS1.mp4 Cite Share Download PDF Status: Published Journal Publication published 30 Apr, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Accepted 08 Apr, 2025 Submission checks completed at journal 07 Apr, 2025 First submitted to journal 28 Mar, 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-5594384\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Article\",\"associatedPublications\":[],\"authors\":[{\"id\":439303139,\"identity\":\"f3aa555e-864a-4387-9695-126a7ef84f50\",\"order_by\":0,\"name\":\"Isabelle B. Laumer\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAzElEQVRIiWNgGAWjYBACNnYILccgAWEwNhDUwgyhjYnXwgDVkthAtBY+ZuZnUjcq7qX3z25g3Vzwi0G2n7DD2Mykc84U5864c4Dt9sw+BuOZhKwB+sVMOrctIbfhRgLbbd4ehsQNBwhqYf8mnfsvIV0epmU/YS08QFsaEhIMQFp4fgBtIewXnmLrnGMJhhvvHGy7PbNBwngGIVvk29s33s6pSZCXu9187HbBHxvZ/gZC1iAAYwMzY5sE8erBgJnhD4k6RsEoGAWjYEQAABUBPe7Tvn16AAAAAElFTkSuQmCC\",\"orcid\":\"\",\"institution\":\"Development and Evolution of Cognition Research Group, Max Planck Institute of Animal Behavior\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Isabelle\",\"middleName\":\"B.\",\"lastName\":\"Laumer\",\"suffix\":\"\"},{\"id\":439303140,\"identity\":\"4cb8e257-e2a6-45d1-8399-f8d5a65d6224\",\"order_by\":1,\"name\":\"Shubhangi Kansal\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Leipzig University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Shubhangi\",\"middleName\":\"\",\"lastName\":\"Kansal\",\"suffix\":\"\"},{\"id\":439303141,\"identity\":\"b8c5dbd1-80a2-4d2e-ab8b-416b772820ce\",\"order_by\":2,\"name\":\"Anais van Cauwenberghe\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Development and Evolution of Cognition Research Group, Max Planck Institute of Animal Behavior\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Anais\",\"middleName\":\"van\",\"lastName\":\"Cauwenberghe\",\"suffix\":\"\"},{\"id\":439303143,\"identity\":\"189b54b9-d8f9-4698-b5a7-f121b8aea511\",\"order_by\":3,\"name\":\"Tri Rahmaeti\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Department of Biology, Graduate Program, Faculty of Biology and Agriculture, Universitas Nasional\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Tri\",\"middleName\":\"\",\"lastName\":\"Rahmaeti\",\"suffix\":\"\"},{\"id\":439303147,\"identity\":\"dc915013-b145-471f-8ad6-0ac5ec011c3d\",\"order_by\":4,\"name\":\"Tatang Mitra Setia\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Fakultas Biologi, Universitas Nasional\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Tatang\",\"middleName\":\"Mitra\",\"lastName\":\"Setia\",\"suffix\":\"\"},{\"id\":439303154,\"identity\":\"36e3c375-647f-46d6-9980-d6ca5e607672\",\"order_by\":5,\"name\":\"Roger Mundry\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Georg-August-Universität Göttingen, Johann-Friedrich-Blumenbach Institute, Department for Primate Cognition\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Roger\",\"middleName\":\"\",\"lastName\":\"Mundry\",\"suffix\":\"\"},{\"id\":439303159,\"identity\":\"804f1470-4aeb-4866-9545-e6fab7d2d2e9\",\"order_by\":6,\"name\":\"Daniel Haun\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Department of Comparative Cultural Psychology, Max Planck Institute for Evolutionary Biology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Daniel\",\"middleName\":\"\",\"lastName\":\"Haun\",\"suffix\":\"\"},{\"id\":439303160,\"identity\":\"a3c3e70e-904a-4482-a206-f320cf501ed2\",\"order_by\":7,\"name\":\"Caroline Schuppli\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Development and Evolution of Cognition Research Group, Max Planck Institute of Animal Behavior\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Caroline\",\"middleName\":\"\",\"lastName\":\"Schuppli\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2024-12-06 14:23:19\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-5594384/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-5594384/v1\",\"draftVersion\":[],\"editorialEvents\":[{\"content\":\"https://doi.org/10.1038/s41598-025-97926-z\",\"type\":\"published\",\"date\":\"2025-04-30T15:57:29+00:00\"}],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":80145955,\"identity\":\"3d0505c3-0874-490f-b835-e0070bda7a8b\",\"added_by\":\"auto\",\"created_at\":\"2025-04-08 12:31:46\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":88683,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003ea) EOM frequency.\\u003c/strong\\u003e Number of EOM events per day (i.e., 12 hours; as this roughly corresponds to the wake time of a wild orangutan (86)) as a function of age at both settings (wild: purple; zoo-housed: yellow). The points depict the EOM rate per a given age, per setting, and their area corresponds to the total duration of observations (range of actual observation durations per dot: 0.08 to 22.21 hours). The dashed lines and shaded areas depict the fitted model and its 95% confidence limits. \\u003cstrong\\u003e1b\\u003c/strong\\u003e) \\u003cstrong\\u003eEOM duration.\\u003c/strong\\u003e Duration of EOM events as a function of age in wild (purple) and zoo-housed orangutans (yellow). The points depict the average duration of EOM events per age and setting and their area corresponds to the number of observations (range: 1 to 275).\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5594384/v1/857308254146f7a891cc25b9.png\"},{\"id\":80145438,\"identity\":\"bcc9e807-8ce4-470b-94d2-2530043bd2c9\",\"added_by\":\"auto\",\"created_at\":\"2025-04-08 12:23:41\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":173421,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003eExploration action diversity for all EOM events (2a), exploration action diversity for all events involving only natural objects (2b), exploration body part diversity (2c), and probability to involve multiple objects in explorations (2d) as a function of age in wild (purple) and zoo-housed orangutans (yellow). Dots show the average response per age and setting, whereby the area of the dots corresponds to the number exploration events per dot (ranges: a: 1 to 273; b: 1 to 70, c: 1 to 276; d: 1 to 280). Dashed lines and shaded areas in all plots depict the fitted model and its 95% confidence limits.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5594384/v1/c5380d9adcf2ab27a2b26fef.png\"},{\"id\":80145951,\"identity\":\"2e0047ea-752a-441c-8714-eb06ba86f24a\",\"added_by\":\"auto\",\"created_at\":\"2025-04-08 12:31:41\",\"extension\":\"png\",\"order_by\":3,\"title\":\"Figure 3\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":68187,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003ea) Age at first occurrence of exploratory actions. \\u003c/strong\\u003eAge of first occurrence of all exploratory actions in wild and zoo-housed orangutans. Note that the corresponding analysis (see above) only included actions that were performed by individuals in both settings. Dots surrounded by circles indicate exploratory actions that include the use of a tool. Dashed lines connect the same respective exploratory action. \\u003cstrong\\u003e3b) Age trajectories for repertoires of exploratory actions.\\u003c/strong\\u003e Cumulative number of exploratory actions as a function of age in wild (purple) and zoo-housed orangutans (yellow). Solid lines depict the fitted model and dashed lines indicate the respective asymptote, i.e., the estimated total repertoire size at adult age.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"Figure3.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5594384/v1/187b476fbcf168b90b6e84db.png\"},{\"id\":81987917,\"identity\":\"bcec21b3-d1e2-4845-b34c-aebbecb4a79e\",\"added_by\":\"auto\",\"created_at\":\"2025-05-05 16:06:56\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":1145426,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5594384/v1/dffc0ca7-2b91-40ce-b258-b3f97ff06bda.pdf\"},{\"id\":80145449,\"identity\":\"2003f274-0158-483a-a00f-e9caa65e404e\",\"added_by\":\"auto\",\"created_at\":\"2025-04-08 12:23:43\",\"extension\":\"xlsx\",\"order_by\":3,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":1762741,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"Data.xlsx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5594384/v1/7e4fc4cb7abf39da3cb58755.xlsx\"},{\"id\":80145447,\"identity\":\"d09d150e-a9cf-452b-81b7-e0b0b8f37f80\",\"added_by\":\"auto\",\"created_at\":\"2025-04-08 12:23:43\",\"extension\":\"docx\",\"order_by\":4,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":261444,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"SupplementaryinformationLaumeretal..docx\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5594384/v1/6fb812c7153ef03536dd0aa8.docx\"},{\"id\":80145956,\"identity\":\"8fc08841-f3ac-49bf-a75b-bd5f1ec6d482\",\"added_by\":\"auto\",\"created_at\":\"2025-04-08 12:31:47\",\"extension\":\"mp4\",\"order_by\":5,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"supplement\",\"size\":85359288,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"MovieS1.mp4\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-5594384/v1/02f8014fce2b14c46421bcf8.mp4\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"Wild and zoo-housed orangutans differ in how they explore objects\",\"fulltext\":[{\"header\":\"Introduction\",\"content\":\"\\u003cp\\u003eExploratory object manipulation (i.e., the deliberate manipulation and visual inspection of objects (\\u003cspan additionalcitationids=\\\"CR2 CR3 CR4\\\" citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e), henceforth referred to as \\u0026ldquo;EOM\\u0026rdquo;) enables human infants to learn about objects and their physical properties (e.g. texture, weight, gravity, etc. (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e4\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR7 CR8 CR9\\\" citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e6\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR10\\\" class=\\\"CitationRef\\\"\\u003e10\\u003c/span\\u003e)). EOM is also an important vehicle of cognitive stimulation (\\u003cspan citationid=\\\"CR11\\\" class=\\\"CitationRef\\\"\\u003e11\\u003c/span\\u003e) as it is linked to motor and cognitive development (\\u003cspan additionalcitationids=\\\"CR13 CR14 CR15 CR16 CR17 CR18 CR19\\\" citationid=\\\"CR12\\\" class=\\\"CitationRef\\\"\\u003e12\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e). Neurological studies further provide evidence for a link between EOM and cognitive processing (\\u003cspan additionalcitationids=\\\"CR22\\\" citationid=\\\"CR21\\\" class=\\\"CitationRef\\\"\\u003e21\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR23\\\" class=\\\"CitationRef\\\"\\u003e23\\u003c/span\\u003e). In human infants, manual, visual, and oral EOM peaks around the age of two years and then decreases (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR25\\\" citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e). However, the length of EOM events increases (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR20\\\" class=\\\"CitationRef\\\"\\u003e20\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR27\\\" class=\\\"CitationRef\\\"\\u003e27\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e), and EOM becomes more diverse and complex with increasing age (e.g., a larger number and different exploratory actions are combined; (\\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e1\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR28\\\" class=\\\"CitationRef\\\"\\u003e28\\u003c/span\\u003e)). These patterns are indicative of learning processes as actions become more fine-tuned and more goal-directed over time (\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eSimilar to humans, EOM also leads to the development of knowledge and sensorimotor skills in non-human animals (e.g. (\\u003cspan additionalcitationids=\\\"CR31 CR32 CR33 CR34\\\" citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR35\\\" class=\\\"CitationRef\\\"\\u003e35\\u003c/span\\u003e)). Studies in various mammalian and bird species have shown that EOM correlates positively with problem-solving ability and innovation probability (\\u003cspan additionalcitationids=\\\"CR37 CR38 CR39 CR40 CR41 CR42\\\" citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR43\\\" class=\\\"CitationRef\\\"\\u003e43\\u003c/span\\u003e). This suggests that knowledge and experience gained during EOM can be transferred across contexts and applied to novel problems. Thus, similar to humans, EOM also appears to be a key mechanism in non-human animals for translating cognitive potential into actual knowledge and skills as well as for fostering cognitive performance in several domains (e.g., (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e)).\\u003c/p\\u003e \\u003cp\\u003eIn primates, EOM tendency is closely linked with phylogeny in that strepsirrhine primates show the lowest levels of EOM tendency, followed by Platyrrhines, non-hominid Catarrhines, and great apes (\\u003cspan citationid=\\\"CR45\\\" class=\\\"CitationRef\\\"\\u003e45\\u003c/span\\u003e). Non-human great apes (henceforth called \\u0026ldquo;great apes\\u0026rdquo;) show the most diverse and complex EOMs of all non-human primates in terms of rates of bimanual manipulation, object-object (i.e., two or more are combined), and object-substrate manipulations (i.e., an object is actively combined with a substrate, other than the plain floor) as well as the use of different body parts during manipulation (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR47 CR48\\\" citationid=\\\"CR46\\\" class=\\\"CitationRef\\\"\\u003e46\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR49\\\" class=\\\"CitationRef\\\"\\u003e49\\u003c/span\\u003e). Humans demonstrate even more complex and diverse manipulations (\\u003cspan citationid=\\\"CR50\\\" class=\\\"CitationRef\\\"\\u003e50\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR51\\\" class=\\\"CitationRef\\\"\\u003e51\\u003c/span\\u003e). Despite these differences, manipulations present across multiple taxa develop in a highly similar sequence, suggesting relatively conserved neurodevelopmental processes (\\u003cspan citationid=\\\"CR52\\\" class=\\\"CitationRef\\\"\\u003e52\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR53\\\" class=\\\"CitationRef\\\"\\u003e53\\u003c/span\\u003e). Furthermore, like humans, wild great apes show a peak in exploratory tendency during the early dependency period (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e) and an increase of EOM diversity with increasing age (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eIn large-brained species with pronounced developmental plasticity, cognition is shaped by the social and ecological environment (\\u003cspan citationid=\\\"CR54\\\" class=\\\"CitationRef\\\"\\u003e54\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR55\\\" class=\\\"CitationRef\\\"\\u003e55\\u003c/span\\u003e). In humans, the conditions under which one grows up influence cognitive development (e.g., (\\u003cspan additionalcitationids=\\\"CR57\\\" citationid=\\\"CR56\\\" class=\\\"CitationRef\\\"\\u003e56\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR58\\\" class=\\\"CitationRef\\\"\\u003e58\\u003c/span\\u003e)), and cognitive performance is shaped by specific experiences during development (\\u003cspan citationid=\\\"CR59\\\" class=\\\"CitationRef\\\"\\u003e59\\u003c/span\\u003e). Children growing up with severely limited social and physical stimulation, later show deficits in several cognitive domains, along with physiological and anatomical changes in the brain (e.g., (\\u003cspan citationid=\\\"CR57\\\" class=\\\"CitationRef\\\"\\u003e57\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR60\\\" class=\\\"CitationRef\\\"\\u003e60\\u003c/span\\u003e)). Similarly, rhesus monkeys (\\u003cem\\u003eMacaca mulatta\\u003c/em\\u003e) that grow up in socially deprived conditions perform worse in various cognitive tasks and have significantly altered brain structures ((\\u003cspan citationid=\\\"CR61\\\" class=\\\"CitationRef\\\"\\u003e61\\u003c/span\\u003e), see (\\u003cspan citationid=\\\"CR62\\\" class=\\\"CitationRef\\\"\\u003e62\\u003c/span\\u003e) for similar findings in rats). High frequency and quality of social inputs during early childhood on the other hand positively affect cognitive development in humans (\\u003cspan citationid=\\\"CR63\\\" class=\\\"CitationRef\\\"\\u003e63\\u003c/span\\u003e). Accordingly, enculturation (i.e., being raised with close human contact) positively affects performance in physical and social cognition tasks in great apes (\\u003cspan additionalcitationids=\\\"CR65 CR66 CR67 CR68\\\" citationid=\\\"CR64\\\" class=\\\"CitationRef\\\"\\u003e64\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR69\\\" class=\\\"CitationRef\\\"\\u003e69\\u003c/span\\u003e). Effects of the ecological and social environment on the development of cognition appear to be, at least to some degree, mediated through exploration (\\u003cspan citationid=\\\"CR42\\\" class=\\\"CitationRef\\\"\\u003e42\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR68\\\" class=\\\"CitationRef\\\"\\u003e68\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR70\\\" class=\\\"CitationRef\\\"\\u003e70\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eZoo-housed great apes are a frequently used model to study human cognition. Compared to their wild conspecifics, they are easier to access and observe, and can be tested under controlled conditions, which allows for minimizing environmental and social influences for the time of testing. As cognitive studies on zoo-housed great apes often draw conclusions on the evolutionary level, it is crucial to increase our understanding of how the zoo environment affects their cognitive performance. Wild and zoo-housed individuals are exposed to different physical and social environments, which result in differences in the cognitive demands they face (\\u003cspan additionalcitationids=\\\"CR72 CR73\\\" citationid=\\\"CR71\\\" class=\\\"CitationRef\\\"\\u003e71\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR74\\\" class=\\\"CitationRef\\\"\\u003e74\\u003c/span\\u003e). On the one hand, wild great apes face cognitive challenges that are absent in captivity such as the need to locate food sources in the habitat, defend their territories, or hide from predators (\\u003cspan citationid=\\\"CR56\\\" class=\\\"CitationRef\\\"\\u003e56\\u003c/span\\u003e). On the other hand, zoo-housed great apes grow up in close contact with humans and thus likely receive more varied social stimulation (\\u003cspan citationid=\\\"CR68\\\" class=\\\"CitationRef\\\"\\u003e68\\u003c/span\\u003e). Furthermore, zoo-housed great apes are provided with carefully designed, at times even computer-assisted enrichment and thus objects that vary in their properties (e.g., shape, size, color, material) and can be manipulated in myriads of ways (e.g. (\\u003cspan citationid=\\\"CR73\\\" class=\\\"CitationRef\\\"\\u003e73\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR76\\\" citationid=\\\"CR75\\\" class=\\\"CitationRef\\\"\\u003e75\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR77\\\" class=\\\"CitationRef\\\"\\u003e77\\u003c/span\\u003e)). Zoo-housed orangutans thus likely gain a broad range of affordances from EOM, that they may transfer to novel situations (i.e., when they manipulate a novel object or are confronted with novel problems, e.g. (\\u003cspan citationid=\\\"CR78\\\" class=\\\"CitationRef\\\"\\u003e78\\u003c/span\\u003e)).\\u003c/p\\u003e \\u003cp\\u003eZoo-housed individuals also have increased amounts of time and energy available for exploration because their food is regularly provided and they do not have to travel to find food or mates. So far, it has remained largely unclear how these factors affect great apes\\u0026rsquo; cognitive performance, mainly because performing cognitive tests with wild great apes is challenging for a variety of factors such as their high levels of neophobia (\\u003cspan citationid=\\\"CR79\\\" class=\\\"CitationRef\\\"\\u003e79\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR80\\\" class=\\\"CitationRef\\\"\\u003e80\\u003c/span\\u003e) and high protection status due to associated risks such as disease transmission from humans to the test subjects (e.g., (\\u003cspan citationid=\\\"CR81\\\" class=\\\"CitationRef\\\"\\u003e81\\u003c/span\\u003e)).\\u003c/p\\u003e \\u003cp\\u003eUsing EOM as a measure of cognitive performance may help to quantify cognitive differences in wild and captive great apes. Even though the developmental trajectories for the development of the overall frequency of EOM are to some degree hard-wired (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e), there is evidence that captivity increases overall exploration tendency in great apes (e.g., (\\u003cspan citationid=\\\"CR79\\\" class=\\\"CitationRef\\\"\\u003e79\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR82\\\" class=\\\"CitationRef\\\"\\u003e82\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR83\\\" class=\\\"CitationRef\\\"\\u003e83\\u003c/span\\u003e)). However, so far, detailed comparisons of the development of EOM in wild versus zoo-housed great apes across ages are lacking. Because individuals can experience cognitive stimulation through EOM throughout the day, to be able to quantify potential differences in cognitive stimulation through EOM, one needs to compare exploration during their daily lives in their habitat (henceforward called \\u0026ldquo;daily EOM behavior\\u0026rdquo;). Because the effects of captivity on cognition may change throughout an individual\\u0026rsquo;s life, it is important to study cognitive differences across a large age range. To date, systematic comparisons of wild and captive animals\\u0026rsquo; daily EOM behavior and comparisons across large age ranges are missing.\\u003c/p\\u003e \\u003cp\\u003eOrangutans are the ideal study species to investigate the effect of the environment on exploration as they show a high tendency to explore, which allows for studying the details of the behavior (e.g. (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR30\\\" class=\\\"CitationRef\\\"\\u003e30\\u003c/span\\u003e)). Furthermore, whilst the mainly arboreal lifestyle of wild orangutans (e.g. (\\u003cspan citationid=\\\"CR84\\\" class=\\\"CitationRef\\\"\\u003e84\\u003c/span\\u003e)) requires them to hold onto branches and thus likely limits exploration, zoo-housed orangutans spend more time on the ground (e.g. (\\u003cspan citationid=\\\"CR85\\\" class=\\\"CitationRef\\\"\\u003e85\\u003c/span\\u003e)) and thus often have both hands free for exploration. Furthermore, because wild orangutans (\\u003cem\\u003ePongo sp.\\u003c/em\\u003e) are challenging to study because of their low numbers and resulting high protection status, and because their habitats present challenges to human observers, it is crucial to understand whether zoo-housed orangutans differ significantly from wild orangutans. Thus, we compared different aspects of daily EOM behavior in wild and zoo-housed orangutans across all ages and predicted that:\\u003cdiv class=\\\"BlockQuote\\\"\\u003e\\u003cp\\u003eP1a) The age trajectories of the overall frequency of EOM are similar in zoo-housed and wild orangutans.\\u003c/p\\u003e\\u003cp\\u003eP1b) Zoo-housed orangutans of all ages show more frequent EOM and longer EOM bouts than wild orangutans.\\u003c/p\\u003e\\u003cp\\u003eP2) Zoo-housed orangutans of all ages show a greater variety of actions, use more body parts to explore, and are more likely to explore multiple objects. Furthermore, zoo-housed orangutans show a higher diversity of actions compared to wild orangutans, also when exploring the same objects.\\u003c/p\\u003e\\u003cp\\u003eP3) The ages at which individuals first show each behavior will be similar across the two settings.\\u003c/p\\u003e\\u003c/div\\u003e\\u003c/p\\u003e \\u003cp\\u003eIn our study, we defined EOM as the prolonged, non-repetitive, usually destructive manipulations of objects (with or without a tool), excluding feeding and nest building, while the visual and tactile foci (manual and/or oral) of the individual are on the object. We analysed a balanced data set containing 12\\u0026rsquo;000 EOM events of wild and zoo-housed Sumatran orangutans. Because most EOM happens during the immature period (below the age of 16)(\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e) and to prevent overall trends across all ages from overwriting effects during the immature period, we present separate analyses on the immature period in the supplementary information.\\u003c/p\\u003e\"},{\"header\":\"Results\",\"content\":\"\\u003cdiv id=\\\"Sec3\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e1a) Developmental trajectory of EOM frequency\\u003c/h2\\u003e \\u003cp\\u003eAge-dependent EOM rates significantly differed between wild and zoo-housed orangutans (full-null model comparison, likelihood ratio test (LRT): χ2\\u0026thinsp;=\\u0026thinsp;114.362, df\\u0026thinsp;=\\u0026thinsp;4, P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001), and there was a significant interaction between setting and age squared (χ2\\u0026thinsp;=\\u0026thinsp;7.410, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.006), indicating setting-specific age effects; SI, section D1, Table S5). Inspecting the data and the fitted model revealed that zoo-housed individuals generally explored more frequently than wild individuals. Furthermore, in wild orangutans, EOM rates peaked at around two to three years, whereas in zoo-housed orangutans, statistically no such peak was detected (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ea). However, notably, the fitted model was associated with large uncertainly (wide confidence intervals in Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003ea) and revealed large fitted values for 0.5 to 2-year-old zoo animals.\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e \\u003c/div\\u003e\\n\\u003ch3\\u003e1b) Duration of EOM\\u003c/h3\\u003e\\n\\u003cp\\u003eThe age-dependent EOM duration per EOM event was not significantly different between settings as indicated by a non-significant full-null model comparison (LRT: χ2\\u0026thinsp;=\\u0026thinsp;4.839, df\\u0026thinsp;=\\u0026thinsp;3, P\\u0026thinsp;=\\u0026thinsp;0.184, see Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig1\\\" class=\\\"InternalRef\\\"\\u003e1\\u003c/span\\u003eb and SI, D, Table S6 for more details; median \\u003csub\\u003ezoo\\u003c/sub\\u003e: 30s [10s, 80s], median \\u003csub\\u003ewild\\u003c/sub\\u003e: 30s [15s, 60s]). See SI, D, 1b.2 for the results of the models on EOM duration in immature individuals only, where also no significant difference between the settings was found.\\u003c/p\\u003e\\n\\u003ch3\\u003e2a) EOM diversity: actions\\u003c/h3\\u003e\\n\\u003cp\\u003eWhen looking at exploration events including all object types, the age-dependent number of exploratory actions performed per EOM event differed significantly between wild and zoo-housed orangutans (LRT: χ2\\u0026thinsp;=\\u0026thinsp;17.019, df\\u0026thinsp;=\\u0026thinsp;3, P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001). However, the interaction between setting and age squared was not significant in this model (χ2\\u0026thinsp;=\\u0026thinsp;1.251, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.263; SI, section D2a, Table S8). After the removal of this non-significant interaction, we also did not find a significant effect of age squared (χ2\\u0026thinsp;=\\u0026thinsp;1.525, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.217; SI, section D2a, Table S9) which indicated linear age trajectories and led us to remove age squared from the model. In the resulting model, also the interaction between age and setting was not significant (χ2\\u0026thinsp;=\\u0026thinsp;0.049, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.825; SI, section D2a, Table S10) and we thus removed it from the model. The final model revealed significant effects of setting (χ2\\u0026thinsp;=\\u0026thinsp;15.620, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001; SI, section C3a, Table S11) and age (χ2\\u0026thinsp;=\\u0026thinsp;7.140, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.008; SI, section D2a, Table S11). The number of exploratory actions per EOM event increased with age in a similar fashion in wild and zoo-housed orangutans, but zoo-housed individuals performed significantly more exploratory actions per event (Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ea; median number of actions \\u003csub\\u003ezoo\\u003c/sub\\u003e = 2 [1, 3], median number of actions \\u003csub\\u003ewild\\u003c/sub\\u003e = 2 [1, 2]). The results of the models on exploration action diversity in immature individuals only showed a significant difference in the same direction between the settings (SI, section D, 2a.2).\\u003c/p\\u003e \\u003cp\\u003e \\u003c/p\\u003e\\n\\u003ch3\\u003e2b) EOM action diversity: EOM events involving objects available in both settings\\u003c/h3\\u003e\\n\\u003cp\\u003eWhen looking at exploration events including natural objects that occur at both settings, the age-dependent number of exploratory actions performed per EOM event was significantly different between zoo-housed and wild orangutans (LRT: χ2\\u0026thinsp;=\\u0026thinsp;10.000, df\\u0026thinsp;=\\u0026thinsp;3, P\\u0026thinsp;=\\u0026thinsp;0.019; see SI, D, S16). After the iterative removal all non-significant effects involving interactions and age squared (see SI, D, Table S16-S18) we found significant effects of setting (χ2\\u0026thinsp;=\\u0026thinsp;8.462, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.004; see SI, D, Table S19) and age (χ2\\u0026thinsp;=\\u0026thinsp;5.299, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.021; see SI, D, Table S19) in that zoo-housed orangutans used more exploratory actions when exploring objects that are available at both settings (see Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003eb). Accordingly, the average number of exploratory actions was higher in zoo-housed than in wild orangutans (median number of actions \\u003csub\\u003ezoo\\u003c/sub\\u003e = 2 [1, 3], median number of actions \\u003csub\\u003ewild\\u003c/sub\\u003e = 2 [1, 2]). See SI, D, 2b.2 for the results of the models on exploration action diversity in immature individuals only, where a significant difference between the settings in the same direction was found.\\u003c/p\\u003e\\n\\u003ch3\\u003e2c) EOM diversity: body parts\\u003c/h3\\u003e\\n\\u003cp\\u003eThe age-dependent trajectory of the body parts used per EOM event was significantly different between wild and zoo-housed orangutans (LRT: χ2\\u0026thinsp;=\\u0026thinsp;11.935, df\\u0026thinsp;=\\u0026thinsp;3, P\\u0026thinsp;=\\u0026thinsp;0.008; SI, section D2c, Table S24). The interaction between age squared and setting was not significant in this model (χ2\\u0026thinsp;=\\u0026thinsp;0.042, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.838). After the removal of this interaction, the final model revealed a significant interaction between age and setting (χ2\\u0026thinsp;=\\u0026thinsp;8.600, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.003; SI, D2c, Table S24) and a significant effect of age squared (χ2\\u0026thinsp;=\\u0026thinsp;11.304, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.001; SI, D2c, Table S25). Whilst in wild individuals the number of body parts used to explore peaked at the age of four to eight, in zoo-housed orangutans it peaked at about 16 years of age (see Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ec; median number of body parts \\u003csub\\u003ezoo\\u003c/sub\\u003e: 1 [1, 2]; median number of body parts \\u003csub\\u003ewild\\u003c/sub\\u003e: 1 [1, 2]). See SI, D2c. 2 for the results of the models on exploration body part diversity in immature individuals only, where a significant difference in the same direction between the settings was found.\\u003c/p\\u003e \\u003cdiv id=\\\"Sec8\\\" class=\\\"Section2\\\"\\u003e \\u003ch2\\u003e2d) EOM diversity: objects\\u003c/h2\\u003e \\u003cp\\u003eThere was a significant difference between the settings in the age-dependent probability that EOM events involved more than one object (LRT: χ2\\u0026thinsp;=\\u0026thinsp;47.983, df\\u0026thinsp;=\\u0026thinsp;3, P\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001). The interaction between age squared and setting was significant in this model (χ2\\u0026thinsp;=\\u0026thinsp;5.924, df\\u0026thinsp;=\\u0026thinsp;1, P\\u0026thinsp;=\\u0026thinsp;0.015); for the model results see SI, section D4.1, Table S28). In zoo-housed orangutans, the probability that EOM events involved multiple objects increased until the age of 16 and then decreased, while overall the probability of EOM events involving multiple objects was higher in zoo-housed compared to wild orangutans (see Fig.\\u0026nbsp;\\u003cspan refid=\\\"Fig2\\\" class=\\\"InternalRef\\\"\\u003e2\\u003c/span\\u003ed). Wild orangutans showed an overall low probability to involve multiple objects in their exploration throughout all ages. Among the immature individuals, we found a significant difference in the same direction between the settings for the probability that EOM events involved more than one object (SI, D4.2).\\u003c/p\\u003e \\u003c/div\\u003e\\n\\u003ch3\\u003e3) Age of first occurrence of specific EOM\\u003c/h3\\u003e\\n\\u003cp\\u003eWe did not find a significant difference between settings in the age at which individuals first showed exploratory actions that occurred at both settings (paired-samples t-test: t(\\u003cspan citationid=\\\"CR60\\\" class=\\\"CitationRef\\\"\\u003e60\\u003c/span\\u003e)\\u0026thinsp;=\\u0026thinsp;\\u0026minus;\\u0026thinsp;0.8112, p\\u0026thinsp;=\\u0026thinsp;0.421; see Fig.\\u0026nbsp;3a). However, overall zoo-housed orangutans showed a larger repertoire of EOMs (n\\u0026thinsp;=\\u0026thinsp;123) than wild orangutans (n\\u0026thinsp;=\\u0026thinsp;63; for a detailed list see SI, section D3a, Table S32). Additionally, we also found a larger number of tool-orientated EOM in zoo-housed individuals compared to wild individuals (n\\u003csub\\u003ezoo\\u003c/sub\\u003e=13 vs. n\\u003csub\\u003ewild\\u003c/sub\\u003e=3, Table S32, Fig.\\u0026nbsp;3a (indicated by circles around the respective dots)).\\u003c/p\\u003e \\u003cp\\u003eAlmost all EOM were observed for the first time below the age of ten years in both settings (see Fig.\\u0026nbsp;3b). The cumulative number of EOM as a function of age in zoo orangutans showed a higher asymptote (estimate\\u0026thinsp;=\\u0026thinsp;0.472, SE\\u0026thinsp;=\\u0026thinsp;0.129, z\\u0026thinsp;=\\u0026thinsp;3.654, p\\u0026thinsp;\\u0026lt;\\u0026thinsp;0.001; see Fig.\\u0026nbsp;3b and SI, section D3b, Table S33) but the slope of the increase did not differ significantly between settings (estimate=-0.044, SE\\u0026thinsp;=\\u0026thinsp;0.210, z=-0.210, p\\u0026thinsp;=\\u0026thinsp;0.834; see Fig.\\u0026nbsp;3b and SI, section D3b, Table S33). However, note that cumulative repertoire sizes were determined pooled across all individuals for the analysis, and therefore do not allow for conclusions on the level of individuals\\u0026rsquo; repertoire development.\\u003c/p\\u003e \"},{\"header\":\"Discussion\",\"content\":\"\\u003cp\\u003eTo assess potential effects of the zoo environment on cognition, we investigated differences in EOM between zoo-housed and wild orangutans across a wide age range. Taken together, we found overall similar age-dependent patterns of daily EOM rates for zoo-housed and wild orangutans. However, compared to wild orangutans, zoo-housed immature orangutans showed the highest EOM rates later in development and continued to show higher EOM rates throughout late immaturity and adulthood. We did not find a difference in individuals\\u0026rsquo; EOM duration between the two settings. Zoo-housed orangutans used a greater variety of exploratory actions and body parts to explore. Furthermore, zoo-housed orangutans showed more tool-orientated behaviors and a higher probability of EOM involving multiple objects. Despite larger overall repertoires of exploratory actions at the zoos, there was no difference between the settings as to at what age different exploratory actions were first shown.\\u003c/p\\u003e \\u003cp\\u003eSimilar to studies in human and non-human primates (\\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e3\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR25\\\" citationid=\\\"CR24\\\" class=\\\"CitationRef\\\"\\u003e24\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR26\\\" class=\\\"CitationRef\\\"\\u003e26\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR33\\\" class=\\\"CitationRef\\\"\\u003e33\\u003c/span\\u003e), we found overall similar age trajectories of EOM rates (P1a) with a steep increase during the early dependency period and a subsequent decrease in both settings. This similarity across the vastly different settings suggests that the age-dependent motivation to engage in EOM is not affected by the immediate socio-ecological conditions nor the conditions experienced during development and thus may have hard-wired components. Furthermore, in both settings, EOM became more diverse with increasing age (P2), which is indicative of learning (\\u003cspan citationid=\\\"CR29\\\" class=\\\"CitationRef\\\"\\u003e29\\u003c/span\\u003e). This pattern suggests that just as human infants, wild and zoo-housed immature orangutans learn about objects through EOM.\\u003c/p\\u003e \\u003cp\\u003eAs predicted (P1b), zoo-housed orangutans showed more frequent EOM at all ages in their daily lives which is in line with previous studies in captive great apes (e.g., (\\u003cspan citationid=\\\"CR68\\\" class=\\\"CitationRef\\\"\\u003e68\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR79\\\" class=\\\"CitationRef\\\"\\u003e79\\u003c/span\\u003e)). Living in safe and stable captive conditions, animals may experience increased free time and energy to devote to EOM while not being occupied with time-intensive foraging or vigilance (e.g., (82, 87)). Wild great apes have been found to be less curious than zoo-housed great apes (\\u003cspan citationid=\\\"CR79\\\" class=\\\"CitationRef\\\"\\u003e79\\u003c/span\\u003e), which is commonly attributed to differences in levels of neophobia (88). Whilst studies on curiosity commonly measure individuals\\u0026rsquo; reactions to novel objects, our study looked at EOM performed with the orangutans\\u0026rsquo; standard enrichment (the enrichment scheme mostly entails objects that are provided regularly; for an overview see SI, table \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003eb). The difference in EOM tendencies of familiar objects between the settings cannot be caused by differences in neophobia and may thus indicate that the difference in curiosity found between wild and captive orangutans may also stem from a heightened tendency to explore in zoo-housed orangutans. Overall, our findings are in line with a study on wild orangutans\\u0026rsquo; curiosity which showed that visual exploration of novel objects increases with increasing food availability (89).\\u003c/p\\u003e \\u003cp\\u003eThe difference in EOM rates between the settings was less pronounced for immatures below four years of age, most likely because in both settings they are still suckling and thus energetically buffered, and their activity budgets are less affected by the need to forage. Interestingly, at the age of eight when wild Sumatran orangutans reach weaning and their younger sibling is born (for Suaq, the average weaning age is 8.1 years, Schuppli unpublished data) their EOM rates dropped to reach near zero values, whereas zoo-housed immatures\\u0026rsquo; EOM rate peaked despite much shorter interbirth intervals of ca. 5.5 years at zoos (90, 91) and thus earlier weaning. The fact that the drop in EOM frequency coincides with weaning in the wild but not in the zoos supports the theory that exploration is affected by individuals\\u0026rsquo; need to forage themselves. In contrast to our prediction (second part of prediction P1b), EOM durations did not differ between settings, indicating that, even though wild orangutans overall showed lower rates of EOM, \\u003cem\\u003ewhen\\u003c/em\\u003e exploring, orangutans at both settings were equally persistent in doing so.\\u003c/p\\u003e \\u003cp\\u003eAs we predicted, zoo-housed orangutans showed more diverse EOM (in terms of number of actions performed and body parts used to explore; P2) than their wild counterparts. In undisturbed wild habitats (such as the Suaq research area), orangutans encounter a stable array of natural objects, ranging from tree substrate (e.g., bark, leaves, etc.) to epiphytes, fungi, as well as bird nests. While these objects can be inherently complex and their configurations vary, they only allow for certain exploratory behaviors. Modern zoos, including the zoos where we collected our data, provide specifically designed enrichment consisting of objects of different shapes and materials which likely allows for a large range of exploratory actions (e.g., (\\u003cspan citationid=\\\"CR73\\\" class=\\\"CitationRef\\\"\\u003e73\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR76\\\" citationid=\\\"CR75\\\" class=\\\"CitationRef\\\"\\u003e75\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR77\\\" class=\\\"CitationRef\\\"\\u003e77\\u003c/span\\u003e)). Interestingly, when exploring natural objects that occur at both settings, zoo-housed individuals also showed a greater diversity of exploratory actions (second part of prediction P2). The latter result implies that the difference in EOM diversity across settings is more than a direct result of differences in the objects that are being explored. Instead, this suggests that individuals of the two settings differ in the set of affordances (i.e., the range of actions) that they can draw from when exploring, most likely because they transfer their repertoires of actions across contexts, i.e., explored objects.\\u003c/p\\u003e \\u003cp\\u003eUltimately, these differences in affordances may also affect cognitive performance in other domains, such as problem-solving. Indeed, studies show that zoo-housed animals often show better task performance in problem-solving experiments than their wild counterparts ((\\u003cspan citationid=\\\"CR70\\\" class=\\\"CitationRef\\\"\\u003e70\\u003c/span\\u003e), but see e.g. (92)), a phenomenon known as the \\u0026lsquo;captivity bias\\u0026rsquo; ((87, 93). This is also reflected by higher rates of innovative behavior in captivity compared to wild settings (94\\u0026ndash;97). However, aside from being the result of opportunities to explore different objects, differences in cognitive performance between wild and zoo-housed individuals may also be caused by social factors. Studies have shown that great apes that have experienced extensive interactions with humans exhibit enhanced cognitive performance in the physical (64, 98, 99) and in the social domain (e.g. (\\u003cspan citationid=\\\"CR67\\\" class=\\\"CitationRef\\\"\\u003e67\\u003c/span\\u003e) but see (100)). An experiment with captive and rehabilitated orangutans with different rearing backgrounds found that the degree of human orientation predicted the motivation to explore an apparatus which affected their problem-solving skills (\\u003cspan citationid=\\\"CR68\\\" class=\\\"CitationRef\\\"\\u003e68\\u003c/span\\u003e), suggesting that the effects of the social and physical environment on cognitive performance interact with each other. To pin down in what aspects and to what extent differences in daily EOM behavior translate into differences in cognitive performance, systematic standardized cognitive tests in both settings are needed which are to date lacking for great apes (but see, e.g., (70, 101) for findings on other species).\\u003c/p\\u003e \\u003cp\\u003eZoo-housed orangutans also showed a higher probability of exploring multiple objects in one event (P2). Zoo enrichment likely allows for more complex and sustained combinatory object-object and object-substrate manipulations than naturally occurring objects (\\u003cspan citationid=\\\"CR73\\\" class=\\\"CitationRef\\\"\\u003e73\\u003c/span\\u003e, \\u003cspan additionalcitationids=\\\"CR76\\\" citationid=\\\"CR75\\\" class=\\\"CitationRef\\\"\\u003e75\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR77\\\" class=\\\"CitationRef\\\"\\u003e77\\u003c/span\\u003e). Combinatory object manipulations seem to be linked to abilities in the domain of physical cognition, as tool-using species show the most complex structured object combinations (e.g. corvids and parrots (\\u003cspan citationid=\\\"CR44\\\" class=\\\"CitationRef\\\"\\u003e44\\u003c/span\\u003e), primates (\\u003cspan citationid=\\\"CR49\\\" class=\\\"CitationRef\\\"\\u003e49\\u003c/span\\u003e)). Combinatory actions have also been shown to be relevant to the development of tool-use abilities in humans (102). Even though we found evidence for more multi-object manipulations in zoo-housed orangutans, we could not investigate the exact nature of these manipulations because our data only entail information on how many distinct objects were explored (but not how many objects of the same type). Therefore, we do not know whether the actions performed were combinatory or not. However, our results showed that zoo-housed orangutans have a larger repertoire of tool-orientated behaviors than wild orangutans. The greater availability of objects that allow for more complex, including more combinatory manipulations, may stimulate cognitive development (\\u003cspan additionalcitationids=\\\"CR72 CR73\\\" citationid=\\\"CR71\\\" class=\\\"CitationRef\\\"\\u003e71\\u003c/span\\u003e\\u0026ndash;\\u003cspan citationid=\\\"CR74\\\" class=\\\"CitationRef\\\"\\u003e74\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eMost exploratory actions were first shown by individuals below the age of ten years in both settings. Furthermore, there was no difference between the settings in the ages of first occurrence for exploratory actions that were present at both settings (P3), which supports the conclusion of previous studies that the neurodevelopmental sequence of manipulative skills is largely fixed in primates (see (\\u003cspan citationid=\\\"CR52\\\" class=\\\"CitationRef\\\"\\u003e52\\u003c/span\\u003e, \\u003cspan citationid=\\\"CR53\\\" class=\\\"CitationRef\\\"\\u003e53\\u003c/span\\u003e)). However, zoo-housed orangutans showed an overall higher number of exploratory actions than wild orangutans (P2) which could be a result of having different object types available to explore.\\u003c/p\\u003e \\u003cp\\u003eInterestingly, several of our findings showed that certain aspects of EOM change throughout life, thus developmental effect on EOM are not restricted to the immature period. In both settings, body part diversity during exploration decreased after a certain age whereas action diversity increased with age. This may suggest that orangutans become more skilled as they age and therefore can perform a more diverse range of actions while using a smaller number of body parts. If EOM is directly linked to cognitive performance, our results suggest that cognitive performance, similarly as in humans (e.g., (103, 104)), in great apes also changes throughout adulthood.\\u003c/p\\u003e\"},{\"header\":\"Conclusion\",\"content\":\"\\u003cp\\u003eTaken together, our results show that the zoo environment significantly affects EOM in orangutans as well as that aspects of EOM and the effect of captivity on EOM change throughout life. Our findings also strengthen existing evidence that EOM reflects cognitive development and might function as a means of cognitive stimulation, not just in humans but across great apes.\\u003c/p\\u003e \\u003cp\\u003eThe comparison across settings suggests that the conditions that individuals experience at the zoo amplify their daily EOM behavior. Given the link between EOM and certain areas of cognitive performance (e.g. problem-solving and innovation probability (\\u003cspan additionalcitationids=\\\"CR37 CR38 CR39 CR40 CR41 CR42\\\" citationid=\\\"CR36\\\" class=\\\"CitationRef\\\"\\u003e36\\u003c/span\\u003e–\\u003cspan citationid=\\\"CR43\\\" class=\\\"CitationRef\\\"\\u003e43\\u003c/span\\u003e)), our findings add to the studies that suggest that in some aspects, cognitive performance measured in zoo-housed individuals may not be directly comparable to cognitive performance of their wild counterparts (\\u003cspan additionalcitationids=\\\"CR65 CR66\\\" citationid=\\\"CR64\\\" class=\\\"CitationRef\\\"\\u003e64\\u003c/span\\u003e–\\u003cspan citationid=\\\"CR67\\\" class=\\\"CitationRef\\\"\\u003e67\\u003c/span\\u003e). On the one hand, this calls for caution when drawing conclusions on the evolutionary level, such as on the evolution of cognition. This is especially true for studies that compare cognitive performance across species by testing in captivity because the strength of the effect of captivity on EOM and cognitive performance may vary across different species. On the other hand, differences in EOM and cognitive performance between captive and wild great apes allow for important novel insights into the proximate factors affecting cognition which are in themselves informative about the evolution of cognition. Furthermore, captive studies may help to elucidate a species' cognitive potential, i.e., the level of cognitive abilities individuals can develop upon the evolved cognitive potential they are born with.\\u003c/p\\u003e \\u003cdiv id=\\\"Sec12\\\" class=\\\"Section2\\\"\\u003e \\u003cdiv id=\\\"Sec13\\\" class=\\\"Section3\\\"\\u003e \\u003c/div\\u003e \\u003c/div\\u003e \"},{\"header\":\"Methods\",\"content\":\"\\u003ch2\\u003eSubjects\\u003c/h2\\u003e\\u003cp\\u003eThe data on wild Sumatran orangutans were collected from 2007 to 2020 at the Suaq Balimbing research area located in the Gunung Leuser National Park in South Aceh, Indonesia (3°42’N, 97°26’E). The data on zoo-housed Sumatran orangutans were collected from 2021 to 2023 at Leipzig Zoo and Dresden Zoo in Germany, and Zoo Zurich and Basel Zoo in Switzerland (for information on enclosure size and enrichment schedule see SI, Table \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003eb). All-occurrence event data (that were used for models 2–4; see below) included 6478 EOM events performed by 33 wild orangutans on 363 observation days between the ages of 0.5 and 76.7 years at Suaq Balimbing in Indonesia and 6640 EOM events performed by 24 zoo-housed orangutans on 134 observation days between the ages of 0.6 and 49.4. Note that sample sizes varied for the different analyses, as not all details were reported for all events (for subject information see SI, section A, Table \\u003cspan refid=\\\"MOESM1\\\" class=\\\"InternalRef\\\"\\u003eS1\\u003c/span\\u003ea). Note that a subset of the EOM data in wild immature orangutans (3200 EOM events collected on 13 immatures between the ages of 0.5 and 13 years) has already been published previously (\\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e5\\u003c/span\\u003e). Following van Noordwijk et al. 2018, we call individuals below the age of 16 “immatures”, below the age of 8 years “dependent immatures” (at Suaq the average weaning age is 8.1 years; Schuppli, unpublished data), and between the age of 8 and 16 years “independent immatures”. Individuals above the age of 16 are referred to as “adults”.\\u003c/p\\u003e\\u003ch2\\u003eEthical guidelines\\u003c/h2\\u003e\\u003cp\\u003eThe data collection was purely observational and without any interaction with the study animals. The research protocols for the data collection in the wild were approved by the Ministry of research and technology (RISTEK; research permit no. 152/ SIP/FRP/SM/V/2012 and following permits) and complied with the legal requirements of Indonesia. Data collection at the zoos did not lead to any deviation from the usual husbandry protocols, was approved by the ethics joint committee of the Max Planck Institute for Evolutionary Anthropology, Leipzig Zoo, and was carried out in accordance with relevant guidelines and regulations at Leipzig Zoo, Dresden Zoo, Basel Zoo, and Zoo Zurich. Our research protocols complied with the EAZA minimum standards for the accommodation and care of animals in zoos and aquaria. The research adhered to all German and Swiss laws regarding animal keeping.\\u003c/p\\u003e\\u003ch2\\u003eData collection procedure\\u003c/h2\\u003e\\u003cp\\u003eThe data were collected following an established protocol for orangutan behavioral data collection(ab.mpg.de/571325/standarddatacollectionrules_suaq_detailed_jan204.pdf ) which consisted of a) instantaneous scan sampling at two-minute intervals of the individual’s activity and b) all-occurrence sampling of behaviors of interest, including EOM. We collected the data on wild orangutans during focal animal follows, of which 50.9% were full-day follows (i.e., from when the focal animal got out of its nest in the morning until it went into its evening nest) and 49.1% partial follows (when the focal animal was either found and then followed until the evening nest or followed from the morning nest until it was lost). The zoo data were collected from the moment the orangutans entered their enclosure in the morning until they left it in the evening (or from the beginning until the end of the visitor hours, in case the orangutans remained in the enclosure for 24 hours). The zoo-housed orangutans were observed from the visitors’ area or an observation tower (at Leipzig Zoo), which provides a view of the indoor enclosure from above. EOM was defined as the prolonged, non-repetitive, usually destructive manipulations of objects (with or without a tool), while the visual and tactile foci (manual and/or oral) of the individual are on the object (105) (see movie S1 for an example of EOM in the wild and in zoo-housed orangutans). Following Pisula (106), we excluded feeding with or without a tool (defined as such by actual ingestion), but included object play (i.e., manipulations during which the visual and tactile foci were desynchronized, and were characterized by repetitive movements), as it is at times difficult to reliably distinguish exploratory from non-exploratory object play (but see (107, 108)). Further we did not include nesting behavior. The exception was if the orangutan was deconstructing a nest or was manipulating twigs that may have been unsuccessful attempts to start constructing a nest. This was done as it is hard to judge the individual’s intention in these instances. We coded the duration of EOM events from the start of the manipulation to the time the orangutan stopped manipulating. To be counted as two separate EOM events, there had to be at least ten seconds between two EOM events during which no EOM was observed.Data were collected by 25 experienced observers (see SI, section A2 for details on inter-observer reliability).\\u003c/p\\u003e\\u003ch2\\u003eData preparation\\u003c/h2\\u003e\\u003cp\\u003eTo investigate EOM frequencies, we counted the number of EOM events per observation day (model 1a). We included data from 65 wild and 24 zoo-housed orangutans (collected on 650 and 142 observation days respectively, note that we also included days where the EOM rate was zero). To investigate EOM duration (model 1b), we used the duration of all EOM events that were timed. To investigate exploratory action diversity (model 2a), we used the number of different exploratory actions performed during EOM events (SI, section B, Table \\u003cspan refid=\\\"MOESM2\\\" class=\\\"InternalRef\\\"\\u003eS2\\u003c/span\\u003ea for a list of all explorative actions and definitions). We also assessed exploratory action diversity for EOM events that included natural objects only that occurred in both settings (SI, section B, Table \\u003cspan refid=\\\"MOESM2\\\" class=\\\"InternalRef\\\"\\u003eS2\\u003c/span\\u003eb; model 2b). To investigate the diversity of body parts involved during EOM (model 2c), we used the number of body parts involved (see SI, section B, Table \\u003cspan refid=\\\"MOESM3\\\" class=\\\"InternalRef\\\"\\u003eS3\\u003c/span\\u003e for a list of all body parts). To assess EOM object diversity (model 2d), we coded whether or not multiple distinct objects (i.e., two or more) were manipulated during an event (see SI, Table \\u003cspan refid=\\\"MOESM2\\\" class=\\\"InternalRef\\\"\\u003eS2\\u003c/span\\u003eb).\\u003c/p\\u003e\\u003ch2\\u003eStatistical analysis\\u003c/h2\\u003e\\u003cp\\u003eTo assess how the age-dependent development of various aspects of EOM differed between settings, we fitted general and Generalized Linear Mixed Models (LMM and GLMM; (109)) using a significance criterium of P \\u0026lt; 0.05. In all models, we included setting and age as fixed effects predictors. To account for the possibility that EOM rates peaked at intermediate ages, we also included age squared into the fixed effects part of each model. As the age-dependent trajectory of EOM rate could vary between settings, we also included the interaction between age and setting and age squared and setting in the fixed effects part of all models. We further included random intercepts effect of the individual ID, and in models 1b – 2d we included follow number (i.e., the id of the observation day) nested in subject as a further random intercepts effect (as in the all-occurrence data, events associated with individuals other than the focal of the follow were recorded which resulted in data on multiple individuals under the same follow number). When identifiable, to account for individual variation in age trajectories and to avoid overconfident models and increased type I error rates, we also included random slopes (110, 111) of age and age squared.\\u003c/p\\u003e\\u003cp\\u003eTo assess the overall effects of the predictors, we first compared each full model with a respective null model which lacked setting and its interactions with age and age squared in the fixed effects part (also in the zero inflation part in model 1a) but was otherwise identical to the full model via a likelihood ratio test (LRT, (112)). In case a full-null model comparison revealed significance, we first determined the significance of the interaction between setting and age squared, and if this did not reveal significance, we removed it from the model and then checked the significance of the effect age squared. If this also did not reveal significance, we removed it from the model, too, and checked the significance of the interaction between age and setting. In case this interaction did not reveal significance, we also removed it from the model (see (113)). This iterative removal of non-significant interactions served in obtaining unconditional estimates of the respective lower-order terms (114).\\u003c/p\\u003e\\u003ch2\\u003e1. EOM frequency and duration\\u003c/h2\\u003e\\u003cp\\u003eTo investigate differences in age-dependent EOM rates between the settings (model 1a), we used the total daily number of EOM events as the response and the observation time during which orangutans were visible, divided by six and then log-transformed as an offset term (the mean follow duration was roughly six hours; dividing by six was done to ease model convergence; (115)). To avoid zero-inflation and overdispersion issues, we chose a model with a zero-inflated negative binomial error structure. In the zero-inflation part, we included the fixed effects of species, age, and age squared. The model comprised random slopes of age and age squared within individual, but we had to drop parameters for the correlations among random intercepts and slopes as a model including them did not converge (for final full model structure and sample sizes see SI, section C, Table S4a, b).\\u003c/p\\u003e\\u003cp\\u003eTo compare age-dependent differences in EOM duration between the settings (model 1b), we used a linear mixed effects model (LMM) fitted with Gaussian error distribution and identity link. To achieve roughly normally distributed and homogeneous residuals (assessed by visual inspection of a qq-plot and residuals plotted against fitted values) we log-transformed (base e) the response (EOM duration).\\u003c/p\\u003e\\u003ch2\\u003e2) EOM diversity – actions, body parts, and objects used\\u003c/h2\\u003e\\u003cp\\u003eTo investigate differences in the age-dependent development in EOM diversity (model 2a-2d), we used GLMMs with the number of different exploratory actions used during EOM (exploratory actions that included all object types (model 2a), exploratory actions during EOM events that involve only natural objects occurring at both settings (model 2b), the number of body parts used during EOM (model 2c), and whether or not multiple objects were explored during an event (i.e. one object versus more than one object, coded as 0 and 1; model 2d) as a response variable. For models 2a – 2d, the response variable was the number of unique elements per EOM event.\\u003c/p\\u003e\\u003cp\\u003eSince the response variable in models 2a – 2c did not comprise zeros, we fitted these models with a zero-truncated negative binomial error distribution and log link function. All four models comprised random slopes of age and age squared within individual. Model 2b also included the object as a random intercepts effect because it aimed at comparing EOM action diversity within the same object between the two settings. Initially, we also included parameters for the correlations among random intercepts and slopes. However, in each model we either had one or several absolute correlation parameters being close to one (which is indicative of them being unidentifiable (116) or the model did not converge when we included them. We hence report results for models not comprising such correlation parameters. For the final full model structure of models 2a – 2d and the respective sample sizes see SI, section C, Table S4.\\u003c/p\\u003e\\u003ch2\\u003e3) Age of first occurrence of specific object EOM\\u003c/h2\\u003e\\u003cp\\u003eTo model the cumulative repertoire of exploratory behaviors as a function of age (model 3) we used a specifically tailored model. The data analyzed with this model consisted of the data aggregated such that they were a series of cumulative repertoire sizes and the ages at which these were reached (e.g., 12 unique behaviors at an age of 1.62 years, 28 behaviors at an age of 1.76 years, etc.) separately for wild and zoo-housed individuals but pooled across all individuals of a given setting). To these data we then fitted an exponential function (for detailed information see SI, section D3). We assumed R to be Poisson distributed given the fitted model (for sample sizes see SI, section C, Table S4a,b).\\u003c/p\\u003e\\u003ch2\\u003e4) General considerations\\u003c/h2\\u003e\\u003cp\\u003eWe fitted all models in R (version 4.3.1 or higher; (117)), using the function glmmTMB of the equally named package (version 1.1.8; (Brooks et al., 2017) for models 1a, 2a, 2b, and 2c; the function lmer of the package lme4 (version 1.1–34; (118)) for model 1b and the function glmer of the same package for model 2d. Before fitting the models, we log- (base e) and then z-transformed age to a mean of zero and a standard deviation of one to ease model convergence. In the case of models 1a and 2a-d, we determined the significance of individual fixed effects of the final full model by dropping them from the model and comparing the resulting simplified model with the more complex one using a likelihood ratio test (R function drop1). For model 1b, we determined the significance of individual fixed effects of the final full model by means of the Satterthwaite approximation to degrees of freedom (function lmer of the package lmerTest; version 3.1-3; (119); see also (120)). We estimated 95% confidence intervals of model estimates and fitted values by means of parametric bootstraps (N = 1000 bootstraps; function simulate of the package glmmTMB or function bootMer of the package lme4). For information on model stability, collinearity and overdispersion checks, see SI, section A3.\\u003c/p\\u003e\\u003cp\\u003eFor models 1b-2d, we fitted equivalent models comprising only data collected when individuals were less than 16 years old. This was done because fine-grained effects during this period of development can be overwritten by overall trends across all ages. The details and results of these models are shown in the SI, section C and D.\\u003c/p\\u003e\\u003cp\\u003eWe report information on median and 1st and 3rd quartile throughout the manuscript like this: median [1st quartile, 3rd quartile].\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eAuthor contributions\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eConceptualization: CS\\u003c/p\\u003e\\n\\u003cp\\u003eData collection at Suaq Balimbing: CS, AvC, TR\\u003c/p\\u003e\\n\\u003cp\\u003eData collection at the Zoos: AvC, SK, IBL\\u003c/p\\u003e\\n\\u003cp\\u003eData entry: CS, AvC, SK\\u003c/p\\u003e\\n\\u003cp\\u003eData analysis: IBL, RM, CS\\u003c/p\\u003e\\n\\u003cp\\u003eWriting of the manuscript: IBL\\u003c/p\\u003e\\n\\u003cp\\u003eEditing the manuscript: CS\\u003c/p\\u003e\\n\\u003cp\\u003eCommenting on the manuscript: RM, DH, TMS, TR\\u003c/p\\u003e\\n\\u003cp\\u003eResearch Administration \\u0026amp; Planning: TMS, TR\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eFunding statement\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis project was funded by a Freigeist Fellowship of the Volkswagen Foundation and a Postdoc Mobility Grant by the Swiss National Science Foundation (grant number P400PM_180844) to CS. Funding was also obtained from the Max Planck Institute for Animal Behavior, the University of Zürich, the SUAQ Foundation, the Stiftung Mensch und Tier Freiburg i.Br., the Leakey Foundation (Primate Research Fund), the A.H. Schultz-Foundation, and the Claraz Foundation.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAcknowledgements\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eWe acknowledge all students, volunteers, and local field assistants involved in the collection of standard behavioral data at Suaq. Special thanks go to Emma Lokuciejewski, Frances Luhn, Francois Lamarque, Glenn Honstetter, Ivan Lenzi, Luise Staatsmann, Mulati Mikeliban, Nele Kater, Nora Slania, Paulina Kukofka, Revathe Thillaikumar, Richard Young, Beatrice Ehmann, Ellen Meulmann, Lara Nellissen, Luz Carvajal, Natalie Oliver-Caldwell, and Sofia Forss, who contributed to collecting the data on\\u0026nbsp;EOM\\u0026nbsp;that were used in this study. We gratefully acknowledge the Direktur Tata Kelola Perijinan Riset dan Inovasi dan Otoritas Ilmiah (BRIN), the Balai Besar Taman Nasional Gunung Leuser (TNGL), the local government in South Aceh, Director General of Natural Resources and Ecosystem Conservation (KSDAE) Ministry of Environment and Forestry (KLHK) for their permission and support to conduct this research. We thank the Yayasan Ekosistem Lestari (YEL) and its Orangutan Conservation Program (SOCP) for hosting our project at the Suaq Balimbing monitoring station. We thank the Fakultas Biologi of Universitas Nasional (UNAS) in Jakarta for their collaboration and support, in particular, Dr. Sri Utami Atmoko. We thank the curators, staff, and helpers at the zoo for their support in conducting our data collection: Adrian Baumeyer (Zoo Basel), Dr. Claudia Rudolf von Rohr (Zoo Zurich), Dr. Leyla Davis (Zoo Zurich), Dr. Wolfgang Ludwig (Zoo Dresden), Roman Richter (Zoo Dresden), and Matthias Hendel (Zoo Dresden). We thank our colleagues at the Max Planck Institute for Evolutionary Anthropology in Leipzig, and especially Hanna Petschauer and Dr. Daniel Hanus for their support with data collected at Leipzig Zoo.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eSupplementary material\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eTo view general considerations for this article, please visit \\u0026lt;LINK\\u0026gt;.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eData availability\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe data is available as a supplementary file.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eDeclaration of interests\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThe authors declare no competing interests.\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\n\\u003cli\\u003eFenson L, Schell RE. The origins of exploratory play. Early Child Development and Care. 1985;19(1-2):3-24.\\u003c/li\\u003e\\n\\u003cli\\u003eMuentener P, Herrig E, Schulz L. The efficiency of infants' exploratory play is related to longer-term cognitive development. Frontiers in psychology. 2018;9:635.\\u003c/li\\u003e\\n\\u003cli\\u003ePower TG. Play and exploration in children and animals: Psychology Press; 1999.\\u003c/li\\u003e\\n\\u003cli\\u003eRuff HA, Saltarelli LM, Capozzoli M, Dubiner K. The differentiation of activity in infants' exploration of objects. Developmental psychology. 1992;28(5):851.\\u003c/li\\u003e\\n\\u003cli\\u003eSchuppli C, Van Cauwenberghe A, Setia TM, Haun D. The ontogeny of exploratory object manipulation behaviour in wild orangutans. Evolutionary Human Sciences. 2021;3:e39.\\u003c/li\\u003e\\n\\u003cli\\u003eMalachowski LG, Needham AW. Infants exploring objects: a cascades perspective. Advances in child development and behavior. 64: Elsevier; 2023. p. 39-68.\\u003c/li\\u003e\\n\\u003cli\\u003eNeedham A. Learning in infants\\u0026rsquo; object perception, object-directed action, and tool use. Learning and the infant mind. 2009:208-26.\\u003c/li\\u003e\\n\\u003cli\\u003eOakes LM, Baumgartner HA, editors. Manual object exploration and learning about object features in human infants. 2012 IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL); 2012 7-9 Nov. 2012; San Diego, CA, USA: IEEE.\\u003c/li\\u003e\\n\\u003cli\\u003eSwirbul MS, Herzberg O, Tamis-LeMonda CS. Object play in the everyday home environment generates rich opportunities for infant learning. Infant Behavior and Development. 2022;67:101712.\\u003c/li\\u003e\\n\\u003cli\\u003eBondi D, Robazza C, Lange‐K\\u0026uuml;ttner C, Pietrangelo T. Fine motor skills and motor control networking in developmental age. American Journal of Human Biology. 2022;34(8):e23758.\\u003c/li\\u003e\\n\\u003cli\\u003ePiaget J. The stages of the intellectual development of the child. Bulletin of the Menninger Clinic. 1962;63(26(3)):120\\u0026ndash;8.\\u003c/li\\u003e\\n\\u003cli\\u003eBaumgartner HA, Oakes LM. Investigating the relation between infants' manual activity with objects and their perception of dynamic events. Infancy. 2013;18(6):983-1006.\\u003c/li\\u003e\\n\\u003cli\\u003eNeedham A. Improvements in object exploration skills may facilitate the development of object segregation in early infancy. Journal of Cognition and Development. 2000;1(2):131-56.\\u003c/li\\u003e\\n\\u003cli\\u003eSoska KC, Adolph KE, Johnson SP. Systems in development: motor skill acquisition facilitates three-dimensional object completion. Developmental psychology. 2010;46(1):129.\\u003c/li\\u003e\\n\\u003cli\\u003eWilcox T, Woods R, Chapa C, McCurry S. Multisensory exploration and object individuation in infancy. Developmental psychology. 2007;43(2):479.\\u003c/li\\u003e\\n\\u003cli\\u003eCase-Smith J, Bigsby R, Clutter J. Perceptual-motor coupling in the development of grasp. The American Journal of Occupational Therapy. 1998;52(2):102-10.\\u003c/li\\u003e\\n\\u003cli\\u003eLobo MA, Galloway JC. Postural and object‐oriented experiences advance early reaching, object exploration, and means\\u0026ndash;end behavior. Child development. 2008;79(6):1869-90.\\u003c/li\\u003e\\n\\u003cli\\u003eBabik I, Galloway JC, Lobo MA. Early exploration of one\\u0026rsquo;s own body, exploration of objects, and motor, language, and cognitive development relate dynamically across the first two years of life. Developmental psychology. 2022;58(2):222.\\u003c/li\\u003e\\n\\u003cli\\u003eFagan MK, Iverson JM. The influence of mouthing on infant vocalization. Infancy. 2007;11(2):191-202.\\u003c/li\\u003e\\n\\u003cli\\u003eMuentener P, Herrig E, Schulz L. The efficiency of infants' exploratory play is related to longer-term cognitive development. Frontiers in psychology. 2018;9:291931.\\u003c/li\\u003e\\n\\u003cli\\u003eBegus K, Southgate V, Gliga T. Neural mechanisms of infant learning: differences in frontal theta activity during object exploration modulate subsequent object recognition. Biology letters. 2015;11(5):20150041.\\u003c/li\\u003e\\n\\u003cli\\u003eOrekhova E, Stroganova T, Posikera I, Elam M. EEG theta rhythm in infants and preschool children. Clinical neurophysiology. 2006;117(5):1047-62.\\u003c/li\\u003e\\n\\u003cli\\u003eDegroot A, Wolff MC, Nomikos GG. Acute exposure to a novel object during consolidation enhances cognition. Neuroreport. 2005;16(1):63-7.\\u003c/li\\u003e\\n\\u003cli\\u003eBelsky J, Most RK. From exploration to play: A cross-sectional study of infant free play behavior. Developmental psychology. 1981;17(5):630.\\u003c/li\\u003e\\n\\u003cli\\u003eBock J. Farming, foraging, and children\\u0026rsquo;s play in the Okavango Delta, Botswana. The nature of play: Great apes and humans. 2005:254-81.\\u003c/li\\u003e\\n\\u003cli\\u003ePower TG, Chapieski ML, McGrath MP. Assessment of individual differences in infant exploration and play. Developmental Psychology. 1985;21(6):974.\\u003c/li\\u003e\\n\\u003cli\\u003eBelsky J, Goode MK, Most RK. Maternal stimulation and infant exploratory competence: Cross-sectional, correlational, and experimental analyses. Child development. 1980:1168-78.\\u003c/li\\u003e\\n\\u003cli\\u003eMcCall RB. Exploratory manipulation and play in the human infant. Monographs of the society for research in child development. 1974:1-88.\\u003c/li\\u003e\\n\\u003cli\\u003eCorbetta D, DiMercurio A, Wiener RF, Connell JP, Clark M. How perception and action fosters exploration and selection in infant skill acquisition. Advances in child development and behavior. 2018;55:1-29.\\u003c/li\\u003e\\n\\u003cli\\u003eBard KA. Sensorimotor cognition in young feral orangutans (Pongo pygmaeus). Primates. 1995;36:297-321.\\u003c/li\\u003e\\n\\u003cli\\u003eChevalier-Skolnikoff S. Tool use in Cebus: Its relation to object manipulation, the brain, and ecological adaptations. Behavioral and Brain Sciences. 1989;12(3):610-27.\\u003c/li\\u003e\\n\\u003cli\\u003eHayashi M, Takeshita H, Matsuzawa T. Cognitive development in apes and humans assessed by object manipulation. Cognitive development in chimpanzees. 2006:395-410.\\u003c/li\\u003e\\n\\u003cli\\u003eLamon N, Neumann C, Zuberb\\u0026uuml;hler K. Development of object manipulation in wild chimpanzees. Animal behaviour. 2018;135:121-30.\\u003c/li\\u003e\\n\\u003cli\\u003eButler RA. Discrimination learning by rhesus monkeys to visual-exploration motivation. Journal of Comparative and Physiological Psychology. 1953;46(2):95.\\u003c/li\\u003e\\n\\u003cli\\u003eGuillette LM, Reddon AR, Hurd PL, Sturdy CB. Exploration of a novel space is associated with individual differences in learning speed in black-capped chickadees, Poecile atricapillus. Behavioural Processes. 2009;82(3):265-70.\\u003c/li\\u003e\\n\\u003cli\\u003eAuersperg AM, Von Bayern AM, Gajdon GK, Huber L, Kacelnik A. Flexibility in problem solving and tool use of kea and New Caledonian crows in a multi access box paradigm. PloS one. 2011;6(6):e20231.\\u003c/li\\u003e\\n\\u003cli\\u003eBenson-Amram S, Dantzer B, Stricker G, Swanson EM, Holekamp KE. Brain size predicts problem-solving ability in mammalian carnivores. Proceedings of the National Academy of Sciences. 2016;113(9):2532-7.\\u003c/li\\u003e\\n\\u003cli\\u003eGriffin AS, Guez D. Innovation and problem solving: a review of common mechanisms. Behavioural Processes. 2014;109:121-34.\\u003c/li\\u003e\\n\\u003cli\\u003eOverington SE, Cauchard L, C\\u0026ocirc;t\\u0026eacute; K-A, Lefebvre L. Innovative foraging behaviour in birds: what characterizes an innovator? Behavioural processes. 2011;87(3):274-85.\\u003c/li\\u003e\\n\\u003cli\\u003eReader SM, Laland KN. Social intelligence, innovation, and enhanced brain size in primates. Proceedings of the National Academy of Sciences. 2002;99(7):4436-41.\\u003c/li\\u003e\\n\\u003cli\\u003eWebster SJ, Lefebvre L. Problem solving and neophobia in a columbiform\\u0026ndash;passeriform assemblage in Barbados. Animal Behaviour. 2001;62(1):23-32.\\u003c/li\\u003e\\n\\u003cli\\u003eDamerius LA, Graber SM, Willems EP, van Schaik CP. Curiosity boosts orang-utan problem-solving ability. Animal Behaviour. 2017;134:57-70.\\u003c/li\\u003e\\n\\u003cli\\u003eEbel SJ, Call J. The interplay of prior experience and motivation in great ape problem-solving (Gorilla gorilla, Pan paniscus, Pan troglodytes, and Pongo abelii). Journal of Comparative Psychology. 2018;132(3):294.\\u003c/li\\u003e\\n\\u003cli\\u003eAuersperg AM, Van Horik JO, Bugnyar T, Kacelnik A, Emery NJ, von Bayern AM. Combinatory actions during object play in psittaciformes (Diopsittaca nobilis, Pionites melanocephala, Cacatua goffini) and corvids (Corvus corax, C. monedula, C. moneduloides). Journal of Comparative Psychology. 2015;129(1):62.\\u003c/li\\u003e\\n\\u003cli\\u003eVauclair J. Phylogenetic approach to object manipulation in human and ape infants. Human Development. 1984;27(5-6):321-8.\\u003c/li\\u003e\\n\\u003cli\\u003eByrne RW, Corp N, Byrne JM. Manual dexterity in the gorilla: bimanual and digit role differentiation in a natural task. Animal Cognition. 2001;4:347-61.\\u003c/li\\u003e\\n\\u003cli\\u003eHayashi M, Matsuzawa T. Cognitive development in object manipulation by infant chimpanzees. Animal cognition. 2003;6:225-33.\\u003c/li\\u003e\\n\\u003cli\\u003eHeldstab SA, Kosonen ZK, Koski SE, Burkart JM, van Schaik CP, Isler K. Manipulation complexity in primates coevolved with brain size and terrestriality. Scientific reports. 2016;6(1):24528.\\u003c/li\\u003e\\n\\u003cli\\u003eTorigoe T. Comparison of object manipulation among 74 species of non-human primates. Primates. 1985;26:182-94.\\u003c/li\\u003e\\n\\u003cli\\u003eRedshaw M. Cognitive development in human and gorilla infants. Journal of human Evolution. 1978;7(2):133-41.\\u003c/li\\u003e\\n\\u003cli\\u003eVauclair J, Bard KA. Development of manipulations with objects in ape and human infants. Journal of Human Evolution. 1983;12(7):631-45.\\u003c/li\\u003e\\n\\u003cli\\u003eHeldstab SA, Isler K, Schuppli C, van Schaik CP. When ontogeny recapitulates phylogeny: Fixed neurodevelopmental sequence of manipulative skills among primates. Science advances. 2020;6(30):eabb4685.\\u003c/li\\u003e\\n\\u003cli\\u003eWorkman AD, Charvet CJ, Clancy B, Darlington RB, Finlay BL. Modeling transformations of neurodevelopmental sequences across mammalian species. Journal of Neuroscience. 2013;33(17):7368-83.\\u003c/li\\u003e\\n\\u003cli\\u003eWittig RM, Crockford C. Chimpanzees\\u0026ndash;investigating cognition in the wild. Field and laboratory methods in animal cognition. 2018:119e45.\\u003c/li\\u003e\\n\\u003cli\\u003eBohn M, Eckert J, Hanus D, Lugauer B, Holtmann J, Haun DB. Great ape cognition is structured by stable cognitive abilities and predicted by developmental conditions. Nature Ecology \\u0026amp; Evolution. 2023:1-12.\\u003c/li\\u003e\\n\\u003cli\\u003eBoesch C. What makes us human (Homo sapiens)? The challenge of cognitive cross-species comparison. Journal of Comparative Psychology. 2007;121(3):227.\\u003c/li\\u003e\\n\\u003cli\\u003eNelson III CA, Zeanah CH, Fox NA, Marshall PJ, Smyke AT, Guthrie D. Cognitive recovery in socially deprived young children: The Bucharest Early Intervention Project. Science. 2007;318(5858):1937-40.\\u003c/li\\u003e\\n\\u003cli\\u003eBauer PM, Hanson JL, Pierson RK, Davidson RJ, Pollak SD. Cerebellar volume and cognitive functioning in children who experienced early deprivation. Biological psychiatry. 2009;66(12):1100-6.\\u003c/li\\u003e\\n\\u003cli\\u003eReyes-Garc\\u0026iacute;a V, Pyh\\u0026auml;l\\u0026auml; A, D\\u0026iacute;az-Reviriego I, Duda R, Fern\\u0026aacute;ndez-Llamazares \\u0026Aacute;, Gallois S, et al. Schooling, local knowledge and working memory: A study among three contemporary hunter-gatherer societies. PloS one. 2016;11(1):e0145265.\\u003c/li\\u003e\\n\\u003cli\\u003eEluvathingal TJ, Chugani HT, Behen ME, Juh\\u0026aacute;sz C, Muzik O, Maqbool M, et al. Abnormal brain connectivity in children after early severe socioemotional deprivation: a diffusion tensor imaging study. Pediatrics. 2006;117(6):2093-100.\\u003c/li\\u003e\\n\\u003cli\\u003eS\\u0026aacute;nchez MM, Hearn EF, Do D, Rilling JK, Herndon JG. Differential rearing affects corpus callosum size and cognitive function of rhesus monkeys. Brain research. 1998;812(1-2):38-49.\\u003c/li\\u003e\\n\\u003cli\\u003eSchrijver NC, Pallier PN, Brown VJ, W\\u0026uuml;rbel H. Double dissociation of social and environmental stimulation on spatial learning and reversal learning in rats. Behavioural brain research. 2004;152(2):307-14.\\u003c/li\\u003e\\n\\u003cli\\u003eMaggi S, Irwin LJ, Siddiqi A, Hertzman C. The social determinants of early child development: an overview. Journal of paediatrics and child health. 2010;46(11):627-35.\\u003c/li\\u003e\\n\\u003cli\\u003eFurlong E, Boose K, Boysen S. Raking it in: the impact of enculturation on chimpanzee tool use. Animal cognition. 2008;11:83-97.\\u003c/li\\u003e\\n\\u003cli\\u003eYocom AM, Boysen ST. Comprehension of functional support by enculturated chimpanzees Pan troglodytes. Current Zoology. 2011;57(4):429-40.\\u003c/li\\u003e\\n\\u003cli\\u003eRussell JL, Lyn H, Schaeffer JA, Hopkins WD. The role of socio‐communicative rearing environments in the development of social and physical cognition in apes. Developmental science. 2011;14(6):1459-70.\\u003c/li\\u003e\\n\\u003cli\\u003eTomasello M, Savage‐Rumbaugh S, Kruger AC. Imitative learning of actions on objects by children, chimpanzees, and enculturated chimpanzees. Child development. 1993;64(6):1688-705.\\u003c/li\\u003e\\n\\u003cli\\u003eDamerius LA, Forss SI, Kosonen ZK, Willems EP, Burkart JM, Call J, et al. Orientation toward humans predicts cognitive performance in orang-utans. Scientific Reports. 2017;7(1):40052.\\u003c/li\\u003e\\n\\u003cli\\u003eForss S, Motes-Rodrigo A, Hrubesch C, Tennie C. Chimpanzees\\u0026rsquo;(Pan troglodytes) problem-solving skills are influenced by housing facility and captive care duration. PeerJ. 2020;8:e10263.\\u003c/li\\u003e\\n\\u003cli\\u003eBenson-Amram S, Weldele ML, Holekamp KE. A comparison of innovative problem-solving abilities between wild and captive spotted hyaenas, Crocuta crocuta. Animal Behaviour. 2013;85(2):349-56.\\u003c/li\\u003e\\n\\u003cli\\u003eClark FE. Cognitive enrichment and welfare: Current approaches and future directions. Anim Behav Cogn. 2017;4(1):52-71.\\u003c/li\\u003e\\n\\u003cli\\u003eMeehan CL, Mench JA. The challenge of challenge: can problem solving opportunities enhance animal welfare? Applied Animal Behaviour Science. 2007;102(3-4):246-61.\\u003c/li\\u003e\\n\\u003cli\\u003eTarou LR, Kuhar CW, Adcock D, Bloomsmith MA, Maple TL. Computer-assisted enrichment for zoo-housed orangutans (Pongo pygmaeus). Animal Welfare. 2004;13(4):445-53.\\u003c/li\\u003e\\n\\u003cli\\u003eMilgram NW, Siwak-Tapp CT, Araujo J, Head E. Neuroprotective effects of cognitive enrichment. Ageing research reviews. 2006;5(3):354-69.\\u003c/li\\u003e\\n\\u003cli\\u003eTripp JK. Increasing activity in captive orangutans: provision of manipulable and edible materials. Zoo biology. 1985;4(3):225-34.\\u003c/li\\u003e\\n\\u003cli\\u003eShariman P, Ruppert N. Effect of environmental enrichment on activities of captive orangutans at Taiping Zoo, Malaysia. Malayan Nature Journal. 2017;69(4):327-35.\\u003c/li\\u003e\\n\\u003cli\\u003eScheel B, editor Designing digital enrichment for orangutans. Proceedings of the Fifth International Conference on Animal-Computer Interaction; 2018.\\u003c/li\\u003e\\n\\u003cli\\u003eHacques G, Komar J, Seifert L. Learning and transfer of perceptual-motor skill: Relationship with gaze and behavioral exploration. Attention, Perception, \\u0026amp; Psychophysics. 2021;83:2303-19.\\u003c/li\\u003e\\n\\u003cli\\u003eForss SI, Schuppli C, Haiden D, Zweifel N, Van Schaik CP. Contrasting responses to novelty by wild and captive orangutans. American Journal of Primatology. 2015;77(10):1109-21.\\u003c/li\\u003e\\n\\u003cli\\u003eKalan AK, Hohmann G, Arandjelovic M, Boesch C, McCarthy MS, Agbor A, et al. Novelty response of wild African apes to camera traps. Current Biology. 2019;29(7):1211-7. e3.\\u003c/li\\u003e\\n\\u003cli\\u003eDunay E, Apakupakul K, Leard S, Palmer JL, Deem SL. Pathogen transmission from humans to great apes is a growing threat to primate conservation. EcoHealth. 2018;15:148-62.\\u003c/li\\u003e\\n\\u003cli\\u003eKummer H, Goodall J. Conditions of innovative behaviour in primates. Philosophical Transactions of the Royal Society of London B, Biological Sciences. 1985;308(1135):203-14.\\u003c/li\\u003e\\n\\u003cli\\u003eKluen E, R\\u0026ouml;nk\\u0026auml; K, Thorogood R. Prior experience of captivity affects behavioural responses to \\u0026lsquo;novel\\u0026rsquo;environments. PeerJ. 2022;10:e13905.\\u003c/li\\u003e\\n\\u003cli\\u003eAshbury AM, Posa MRC, Dunkel LP, Spillmann B, Atmoko SSU, van Schaik CP, et al. Why do orangutans leave the trees? Terrestrial behavior among wild Bornean orangutans (Pongo pygmaeus wurmbii) at Tuanan, Central Kalimantan. American journal of primatology. 2015;77(11):1216-29.\\u003c/li\\u003e\\n\\u003cli\\u003ePearson EL, Davis JM, Litchfield CA. A case study of orangutan and siamang behavior within a mixed-species zoo exhibit. Journal of Applied Animal Welfare Science. 2010;13(4):330-46.\\u003c/li\\u003e\\n\\u003cli\\u003eAshbury A.M. LF, Permana A., Rahmaeti T., Utami-Atmoko S., Crofoot M.C., Schuppli C. Wild orangutans maintain sleep homeostasis through napping, counterbalancing socioecological factors that interfere with their sleep. Current Biology. in revision;https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5131665.\\u003c/li\\u003e\\n\\u003cli\\u003eHaslam M. \\u0026lsquo;Captivity bias\\u0026rsquo; in animal tool use and its implications for the evolution of hominin technology. Philosophical Transactions of the Royal Society B: Biological Sciences. 2013;368(1630):20120421.\\u003c/li\\u003e\\n\\u003cli\\u003eForss SI, Koski SE, van Schaik CP. Explaining the paradox of neophobic explorers: the social information hypothesis. International Journal of Primatology. 2017;38:799-822.\\u003c/li\\u003e\\n\\u003cli\\u003eSchuppli C, Nellissen L, Carvajal L, Ashbury AM, Oliver-Caldwell N, Rahmaeti T, et al. Ecological, social, and intrinsic factors affecting wild orangutans\\u0026rsquo; curiosity, assessed using a field experiment. Scientific Reports. 2023;13(1):13184.\\u003c/li\\u003e\\n\\u003cli\\u003eAnderson HB, Thompson ME, Knott CD, Perkins L. Fertility and mortality patterns of captive Bornean and Sumatran orangutans: is there a species difference in life history? Journal of Human Evolution. 2008;54(1):34-42.\\u003c/li\\u003e\\n\\u003cli\\u003eShumaker RW, Wich SA, Perkins L. Reproductive life history traits of female orangutans (Pongo spp.). Primate reproductive aging. 36: Karger Publishers; 2008. p. 147-61.\\u003c/li\\u003e\\n\\u003cli\\u003eMcCune KB, Jablonski P, Lee S-i, Ha RR. Captive jays exhibit reduced problem-solving performance compared to wild conspecifics. Royal Society open science. 2019;6(1):181311.\\u003c/li\\u003e\\n\\u003cli\\u003eMeulman EJ, Sanz CM, Visalberghi E, van Schaik CP. The role of terrestriality in promoting primate technology. Evolutionary Anthropology: Issues, News, and Reviews. 2012;21(2):58-68.\\u003c/li\\u003e\\n\\u003cli\\u003eMotes‐Rodrigo A, Tennie C. Captive great apes tend to innovate simple tool behaviors quickly. American Journal of Primatology. 2022;84(10):e23311.\\u003c/li\\u003e\\n\\u003cli\\u003eLehner SR, Burkart JM, van Schaik CP. An evaluation of the geographic method for recognizing innovations in nature, using zoo orangutans. Primates. 2010;51:101-18.\\u003c/li\\u003e\\n\\u003cli\\u003eManrique HM, V\\u0026ouml;lter CJ, Call J. Repeated innovation in great apes. Animal Behaviour. 2013;85(1):195-202.\\u003c/li\\u003e\\n\\u003cli\\u003evan Schaik CP, Burkart J, Damerius L, Forss SI, Koops K, van Noordwijk MA, et al. The reluctant innovator: orangutans and the phylogeny of creativity. Philosophical Transactions of the Royal Society B: Biological Sciences. 2016;371(1690):20150183.\\u003c/li\\u003e\\n\\u003cli\\u003eBard KA, Gardner KH. Influences on development in infant chimpanzees: Enculturation, temperament, and cognition. 1996.\\u003c/li\\u003e\\n\\u003cli\\u003eGardner RA, Gardner BT, Van Cantfort TE. Teaching sign language to chimpanzees: Suny Press; 1989.\\u003c/li\\u003e\\n\\u003cli\\u003eTomasello M, Call J. The role of humans in the cognitive development of apes revisited. Animal Cognition. 2004;7:213-5.\\u003c/li\\u003e\\n\\u003cli\\u003eR\\u0026ouml;ssler T, Mioduszewska B, O\\u0026rsquo;Hara M, Huber L, Prawiradilaga DM, Auersperg AM. Using an Innovation Arena to compare wild-caught and laboratory Goffin\\u0026rsquo;s cockatoos. Scientific Reports. 2020;10(1):8681.\\u003c/li\\u003e\\n\\u003cli\\u003eConnolly K, Dalgleish M. The emergence of a tool-using skill in infancy. Developmental Psychology. 1989;25(6):894.\\u003c/li\\u003e\\n\\u003cli\\u003eSalthouse TA. Continuity of cognitive change across adulthood. Psychonomic bulletin \\u0026amp; review. 2016;23:932-9.\\u003c/li\\u003e\\n\\u003cli\\u003eMurman DL, editor The impact of age on cognition. Seminars in hearing; 2015: Thieme Medical Publishers.\\u003c/li\\u003e\\n\\u003cli\\u003eSchuppli C, Van Cauwenberghe A, Mirta Setia T, Haun D. The ontogeny of exploratory object manipulation behaviour in wild orangutans. Evolutionary Human Sciences. 2021:1-32.\\u003c/li\\u003e\\n\\u003cli\\u003ePisula W. Play and exploration in animals\\u0026mdash;a comparative analysis. Polish Psychological Bulletin. 2008(2).\\u003c/li\\u003e\\n\\u003cli\\u003eHutt C, editor Exploration and play in children. Symposia of the Zoological Society of London; 1966: London.\\u003c/li\\u003e\\n\\u003cli\\u003ePellegrini AD, Gustafson K. Boys\\u0026rsquo; and girls\\u0026rsquo; uses of objects for exploration, play, and tools in early childhood. The nature of play: Great apes and humans. 2005:113-35.\\u003c/li\\u003e\\n\\u003cli\\u003eBaayen R. Analyzing linguistic data. Cambridge, UK. Cambridge University Press New York; 2008.\\u003c/li\\u003e\\n\\u003cli\\u003eBarr DJ, Levy R, Scheepers C, Tily HJ. Random effects structure for confirmatory hypothesis testing: Keep it maximal. Journal of memory and language. 2013;68(3):255-78.\\u003c/li\\u003e\\n\\u003cli\\u003eSchielzeth H, Forstmeier W. Conclusions beyond support: overconfident estimates in mixed models. Behavioral ecology. 2009;20(2):416-20.\\u003c/li\\u003e\\n\\u003cli\\u003eDobson A. An Introduction to Generalized Linear Models, Baton Rouge, LA. Chapman and Hall/CRC Press; 2002.\\u003c/li\\u003e\\n\\u003cli\\u003eAiken L. Multiple regression: Testing and interpreting interactions. Sage Publications google schola. 1991;2:513-31.\\u003c/li\\u003e\\n\\u003cli\\u003eAiken LS, West SG, Reno RR. Multiple regression: Testing and interpreting interactions: sage; 1991.\\u003c/li\\u003e\\n\\u003cli\\u003eMcCullagh P, Nelder J. 1989: Generalized linear models, London: Chapman and Hall. 1989.\\u003c/li\\u003e\\n\\u003cli\\u003eMatuschek H, Kliegl R, Vasishth S, Baayen H, Bates D. Balancing Type I error and power in linear mixed models. Journal of memory and language. 2017;94:305-15.\\u003c/li\\u003e\\n\\u003cli\\u003eTeam RC. R: a language and environment for statistical computing. R Foundation for statistical Computing, Vienna, Austria [Methodology Reference]. European Environment Agency https://www R-project org/ Accessed. 2023;12.\\u003c/li\\u003e\\n\\u003cli\\u003eBates D, M\\u0026auml;chler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. arXiv preprint arXiv:14065823. 2014.\\u003c/li\\u003e\\n\\u003cli\\u003eKuznetsova A, Brockhoff PB, Christensen RH. lmerTest package: tests in linear mixed effects models. Journal of statistical software. 2017;82:1-26.\\u003c/li\\u003e\\n\\u003cli\\u003eLuke SG. Evaluating significance in linear mixed-effects models in R. Behavior research methods. 2017;49:1494-502.\\u003c/li\\u003e\\n\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"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\":\"info@researchsquare.com\",\"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\":\"exploration, exploratory object manipulation, captive-wild comparison, cognitive development, cognitive aging, great ape\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-5594384/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-5594384/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"\\u003cp\\u003eIn human infants, exploratory object manipulations (henceforth called \\u0026ldquo;EOM\\u0026rdquo;) stimulate cognitive development and affect cognitive performance in later life. Zoo-housed great apes are frequently used to study the evolution of human cognition, however, it is unknown how the zoo environment affects their daily expression of EOM. We investigated how wild and zoo-housed Sumatran orangutans differ in their daily EOM throughout life. We collected\\u0026thinsp;~\\u0026thinsp;12\\u0026rsquo;000 EOM events by 51 wild and zoo-housed individuals of all ages. Zoo-housed orangutans showed significantly higher EOM rates than wild orangutans. Exploratory actions were more diverse in zoos than in the wild, even with objects available in both settings. Zoo-housed orangutans also showed a larger repertoire of exploratory actions and a higher probability of multi-object exploration, including tool use. There was no difference between settings at which age individuals first showed specific exploratory actions. Our results show that the zoo environment significantly affects EOM in orangutans and that the species\\u0026rsquo; exploratory potential exceeds its natural expression. This may have important implications for cognitive performance, as zoo-housed individuals are likely to have a broader range of affordances to draw from when confronted with novel problems. These results highlight the potential of captive-wild comparisons to study cognitive development and evolution.\\u003c/p\\u003e\",\"manuscriptTitle\":\"Wild and zoo-housed orangutans differ in how they explore objects\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-04-08 12:23:31\",\"doi\":\"10.21203/rs.3.rs-5594384/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0},{\"type\":\"decision\",\"content\":\"Accepted\",\"date\":\"2025-04-08T07:01:48+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksComplete\",\"content\":\"\",\"date\":\"2025-04-07T06:49:59+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"submitted\",\"content\":\"Scientific Reports\",\"date\":\"2025-03-28T19:58:01+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"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\":\"80f6bbe4-fd62-4e65-ad9b-1fc1f3c42c56\",\"owner\":[],\"postedDate\":\"April 8th, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"published-in-journal\",\"subjectAreas\":[{\"id\":46765284,\"name\":\"Biological sciences/Developmental biology\"},{\"id\":46765285,\"name\":\"Biological sciences/Evolution\"}],\"tags\":[],\"updatedAt\":\"2025-05-05T16:03:21+00:00\",\"versionOfRecord\":{\"articleIdentity\":\"rs-5594384\",\"link\":\"https://doi.org/10.1038/s41598-025-97926-z\",\"journal\":{\"identity\":\"scientific-reports\",\"isVorOnly\":false,\"title\":\"Scientific Reports\"},\"publishedOn\":\"2025-04-30 15:57:29\",\"publishedOnDateReadable\":\"April 30th, 2025\"},\"versionCreatedAt\":\"2025-04-08 12:23:31\",\"video\":\"\",\"vorDoi\":\"10.1038/s41598-025-97926-z\",\"vorDoiUrl\":\"https://doi.org/10.1038/s41598-025-97926-z\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-5594384\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-5594384\",\"identity\":\"rs-5594384\",\"version\":[\"v1\"]},\"buildId\":\"8U1c8b4HqxoKbykW_rLl7\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}