Animal behaviour

Incipient tradition in wild chimpanzees


The adoption of a new form of tool use has been observed to spread along social-network pathways in a chimpanzee community. The finding offers the first direct evidence of cultural diffusion in these animals in the wild.

Social learning — learning from others — is one of the fastest-expanding research fields in animal behaviour1,2. At the fundamental level of evolutionary biology, social learning provides a high-speed 'second inheritance system' that interacts with genetic inheritance to enrich behavioural evolution2. From a more anthropocentric perspective, animal social learning casts light on the evolutionary foundations of the cultural capacities that make our own species so successful. Studies of putative cultural variations in wild chimpanzees2,3,4, the primates with which (together with bonobos) humans last shared a common ancestor, have been particularly influential in our understanding of behavioural evolution. But these observed regional behavioural differences have displayed little change, making it difficult to investigate the workings of social learning. Now, writing in PLoS Biology, Hobaiter et al.5 describe a new form of tool use in the Sonso community of chimpanzees of the Budongo Forest in Uganda, and present a statistical technique for tracing the social transmission of this innovation.

Chimpanzees at Sonso fold wads of leaves in their mouths to fashion a 'leaf sponge' that they dip into tree holes to extract water to drink. In 2011, the researchers observed the dominant male of the group, Nick, creating a sponge of moss gathered from a tree trunk and using it to drink from a small flooded waterhole — a behaviour not previously recorded in the 20-year research programme at the site. He was watched by the dominant female, Nambi. Over the next six days of intensive and often competitive use of the waterhole (which the researchers suspect contained unusual densities of minerals or other desirable content), Nambi and six other chimpanzees began to display the moss-sponging technique (Fig. 1a). More than 20 other individuals drank at the hole or in puddles around it, but either directly with their mouths or using leaf sponges rather than moss sponges.

Figure 1: Tools and trends.

A: Catherine Hobaiter/B: Edwin Van Leeuwen

a, A chimpanzee from the Sonso community using a moss sponge to drink from a waterhole. Hobaiter et al.5 observed that this new tool use spread from one animal to others along the community's social network. b, Another study9 reports the spread of a new and seemingly useless behavioural 'fad' — sticking a blade of grass in the ear — among chimpanzees in a captive sanctuary community.

To establish whether this behavioural spread was due to social learning, the researchers developed a form of network-based diffusion analysis (NBDA). This statistical technique quantifies the extent to which the spread of a new behaviour is consistent with the predictionthat it will follow the social network — a numerical representation of who associates most closely with whom in the community. An impressive recent example of NBDA6 used more than 73,000 observations of 653 humpback whales during the 27-year spread of a 'lobtail' prey-capture technique, which diffused as predicted by the social network, implying social learning. However, NBDA studies have so far used only a static, summary quantification of the social network. Hobaiter et al. took this approach to a new level, which they describe as dynamic NBDA, by incorporating repeatedly updated information on whom each individual was likely to have watched (those within 1 metre of and facing the current sponger). They found strong evidence consistent with social transmission, with an estimated 15-fold enhancement of moss-sponging behaviour for each time a novice observed an existing moss-sponger.

Of course, NBDA and other purely statistical approaches to analysing observational data are essentially correlational and thus do not necessarily imply cause. To interpret such data, one has to try to rule out alternative potential explanations — in this case, for example, that the order of acquisition observed in the chimpanzees resulted not from social learning but from rank-based queuing to gain access to the waterhole, with each lower-ranked individual happening to watch the previous higher ranker before they got their turn. It seems that this possibility can be excluded, because the 'lower rankers' were often offspring of the higher-ranked earlier moss-spongers and so had simultaneous access to the waterhole, yet started to use moss only after watching their mothers. However, caution is still warranted in interpreting these findings in case some subtle alternative factor explains the observed putative evidence for social learning.

Experimental approaches can provide more-robust tests of causality. Indisputable cases of social learning have already been seen in captive primates, including chimpanzees, in studies in which alternative techniques for using tools or otherwise manipulating foraging tasks are seeded and subsequently spread in different groups7. Such approaches are inherently difficult to engineer in the field, but a few attempts have been made (see ref. 8 for a review). Unfortunately, this method has yet to be successfully implemented with wild chimpanzees, which are surprisingly neophobic.

Nevertheless, non-interventional studies of natural behaviour, such as the one presented by Hobaiter et al., are vital to the field. Experimental studies make good sense only when they build on what has first been established in the wild. The innovation recorded by the authors was not dramatic — it was merely a modification of existing leaf-sponging expertise. But the findings are valuable as the first direct evidence of cultural diffusion in this key species, converging with observational evidence from the wild and rigorously controlled experiments in captive animals to consolidate a substantial case for the role of cultural transmission in such cases.

This study follows hot on the heels of another9 documenting the diffusion of a particularly intriguing innovation — placing a blade of grass in the ear — in chimpanzees living in four large enclosures in an African sanctuary (Fig. 1b). That study is unusual because the behaviour seems to be functionless, and thus akin to human cultural phenomena such as fads and fashions. The grass-in-ear behaviour spread from one apparent inventor in 2007 to seven others by 2012, in just one of the four groups. The lack of overt function makes any explanation other than social learning difficult to accept, and underlines the potential potency of this form of learning in this species.

Researchers have also claimed the first documented case of successful transmission of a novel cultural behaviour — fishing for wood-boring ants using peeled bark or other material — between wild chimpanzee communities10. And in other studies of the selection of materials for nut-cracking, researchers concluded that migrating female chimpanzees soon conform to the practices of the group they move into11. A major question for the future is thus what determines the outcome of such migrations between different local cultures. Why do migrants sometimes seed behaviours that diffuse in their new community in the manner demonstrated in the moss-sponging study, whereas others instead abandon previous behaviours and conform to the new local norms? Investigating which factors throw this important switch will add considerably to our understanding of cultural transmission in animals.


  1. 1

    Hoppitt, W. & Laland, K. N. Social Learning: An Introduction to Mechanisms, Methods, and Models (Princeton Univ. Press, 2013).

    Google Scholar 

  2. 2

    Whiten, A. Nature 437, 52–55 (2005).

    CAS  Article  ADS  Google Scholar 

  3. 3

    Nishida, T., Zamma, K., Matsusaka, T., Inaba, A. & McGrew, W. C. Chimpanzee Behavior in the Wild: An Audio-Visual Encylopedia (Springer, 2010).

    Google Scholar 

  4. 4

    Boesch, C. Wild Cultures: A Comparison between Chimpanzee and Human Cultures (Cambridge Univ. Press, 2012).

    Google Scholar 

  5. 5

    Hobaiter, C., Poisot, T., Zuberbühler, K., Hoppitt, W. & Gruber, T. PLoS Biol. 12, e1001960 (2014).

    Article  Google Scholar 

  6. 6

    Allen, J., Weinrich, M., Hoppitt, W. & Rendell, L. Science 340, 485–488 (2013).

    CAS  Article  ADS  Google Scholar 

  7. 7

    Whiten, A. & Mesoudi, A. Phil. Trans. R. Soc. B 363, 3477–3488 (2008).

    Article  Google Scholar 

  8. 8

    van de Waal, E., Borgeaud, C. & Whiten, A. Science 340, 483–485 (2013).

    CAS  Article  ADS  Google Scholar 

  9. 9

    van Leeuwen, E. J. C., Cronin, K. A. & Haun, D. B. M. Anim. Cogn. (2014).

  10. 10

    O'Malley, R. C., Wallauer, W., Murray, C. M. & Goodall, J. Curr. Anthropol. 53, 650–663 (2012).

    Article  Google Scholar 

  11. 11

    Luncz, L. V. & Boesch, C. Am. J. Primatol. 76, 649–657 (2014).

    Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Andrew Whiten.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Whiten, A. Incipient tradition in wild chimpanzees. Nature 514, 178–179 (2014).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.