Current Biology Vol 22 No 10R402

Animal Culture: Chimpanzee than wooden tools more frequently

Conformity?

Culture-like phenomena in wild animals have received much attention, but howgood is the evidence and how similar are they to human culture? New data onchimpanzees suggest their culture may even have an element of conformity.

Carel P. van Schaik

In a variety of species, especiallycetaceans and primates, biologistshave documented geographic variationin communication signals, comfortvariants and often tool-basedsubsistence techniques [1], which theyinterpreted as an expression of culture,i.e. innovative behavior patterns thatindividuals acquire through someform of social learning rather thanindependently. This interpretation,if correct, suggests that humanculture, despite all its differences,shares homologous elements withgreat-ape cultures. In this issue ofCurrent Biology, Luncz, Mundry andBoesch [2] present detailed informationthat bolsters the cultural interpretationof one particularly prominentchimpanzee behavior, tool-assistednut cracking, and may even suggestchimpanzees show a tendencytoward conformity.

The cultural interpretation ofgeographic variation in behavior hasbeen challenged as unwarranted[3]. Most criticism focused on theapproach used, the ethnographicmethod of comparing practicesbetween different areas. This methoddoes not provide direct evidence infavor of a cultural explanation, butinstead tries to exclude alternativeexplanations for geographic variation:animals could either rely onindependent but convergent behaviordue to shared genetic predispositions,or on independent but convergentindividual behavioral plasticity inresponse to potentially subtleecological differences. The criticismrevolves around the extent to whichthese alternative explanations reallycan be excluded.

One may wonder why the questionof whether geographic variation inbehaviors was the result of culturalprocesses was not settled years ago.After all, elegant translocation orcross-fostering experiments couldprovide definitive answers [4]. Thereason is that these experiments are

not feasible or permissible with animalssuch as great apes. Thus, theethnographic method continuesto be necessary to validate acultural interpretation for specificcases, although it is not good atestimating the relative importanceof cultural processes in creatinggeographic variation in behavior ortechnology [5].

An absence of genetic influences hasbeen implied by the presence ofbehavioral differences betweenchimpanzee populations despite a lackof genetic differentiation [6,7]. Dealingwith the role of individual plasticity inresponse to subtle ecologicaldifferences has proven more difficult,however. A classic test casehighlighted the role of culturaltransmission. Chimpanzees at varioussites were seen to use distinctant-dipping techniques [8] — adifference attributed to cultural effects.Subsequent work, however, showedthat different techniques existed sideby side in a single population [9]. Criticsused this observation to suggest thatecological influences could explain thebetween-site variation [3]. This inspiredmorework, which convincingly showedthat details of geographic variationcannot be fully explained by plasticresponses to ecological differences[10,11], and that they are, therefore,possibly cultural. Indeed, manyaspects of ant-dipping are acquiredthrough social learning [12], showingthat a mix of individually and sociallyacquired experience causes thebehavior patterns of an individual. Thisexample shows not only that culturalprocesses explain some of thevariation in chimpanzee behavior, butalso how criticism can lead to morepowerful tests.

Luncz et al. [2] now provide aneven more detailed analysis oftool-assisted nut cracking (Figure 1).They report consistent differences inthe techniques among three adjacentchimpanzee communities inhabitingthe same patch of rainforest. Membersof one community used stone rather

and kept on using them as the seasonprogressed even though nuts couldincreasingly be cracked comfortablywith the more abundantly availablewooden tools.When they usedwoodentools, these were consistently smallerthan those used elsewhere. Subtledifferences also existed between thetwo other communities. Availability ofeither stone or wood hammers did notdiffer among communities, and nutseverywhere were hardest early in theseason. The study by Luncz et al. [2]therefore strongly suggests thatindividual community members weremore similar in their techniquesthan expected based on ecologicalgrounds, although critics might pointout that one possible non-culturalexplanation was not excluded:it could be possible that thepredominant use of stone tools ina community, perhaps because thenuts initially are hardest there, leadsto a higher accumulation of stonetools right around the nut-crackingsites than in other communities.Thus, any chimpanzee, regardlessof its initial preferences, mightindependently bias its techniquetoward the community’smodal pattern,simply because of what tools happento be within reach at the cracking sites.Provided this alternative can berefuted, the study by Luncz et al. [2]strengthens a cultural interpretationof one of the most visible examplesof geographic variation in chimpanzeebehavior, where social learning hasalready been implicated in itsacquisition.Assuming the cultural interpretation

stands, there is another implication.In chimpanzees, adult females aregenerally born in a differentcommunity than where they live asadults. We know that females havealready learned how to crack nutsbefore they disperse. The findings ofLuncz et al. [2], therefore, imply thatthere is some process at workwhereby immigrants modify their ownnut-cracking techniques and convergeon the local pattern. Such conformityis yet another aspect of behavior longthought to be uniquely human, butthe present study adds to the evidencequestioning the strict version ofthis notion: Whiten et al. [13], forinstance, reported experiments withchimpanzees, where they could adoptone of two distinct techniques to

Figure 1. Cracking culture.

A juvenile male chimpanzee uses a stone tooland a root anvil to crack nuts. In the IvoryCoast, chimpanzees show group-dependentnut crackingbehaviorwhich isdifferent amongneighboring communities, even though theylive in close proximity to one another inthe same habitat, suggesting a cultural in-fluence on tool choice. Picture: by MarkLinfield.

DispatchR403

extract food from a machine calledpan-pipes (poke or lift), and found thatwhen seeded with one technique, theindividuals exposed to this techniquereliably adopted the demonstratedtechnique. In each group, some‘corruption’ with the other techniquesubsequently arose, but tended todisappear again over time. Similarself-correcting tendencies were foundin other experiments [14]. Likewise,Perry [15] could show that wildimmature white-faced capuchinmonkeys gradually settled on oneof two distinct techniques to openLuehea fruits, and that this techniquetended to be the one they werepredominantly exposed to. In all thesecases, the techniques were basicallyequally efficient.

Such homogenizing effects arealmost inevitable. First, naıveindividuals are simply most likely toadopt whatever technique they aremost exposed to, regardless of theidentity of the models. This tendencywill over time produce local

homogeneity. It may be adaptivebecause they are most likely to providethe locally optimal solution. We can callthis tendency ‘weak informationalconformity’. Second, learners withprevious knowledge may value whatthey encounter more than what theyalready know themselves if there issome asymmetry, e.g. becausethey are immigrants, as in thesechimpanzees. There is experimentalevidence for this from Norway rats,where animals may overcome apersonal preference when confrontedwith others demonstrating theother option [16]. However, thehomogenizing effect will be weakerthe more individuals tend to stick towhat they know, as shown in captivechimpanzees.

Local homogeneity arises mostreliably when animals preferentially,and thus disproportionately, copywhat is demonstrated by the majorityof role models, rather than merelywhatever they encounter the most [17].This may seem like a subtle difference,but it implies an explicit preferencefor the majority’s variant, whichcould be adaptive because it allowsindividuals to tap more effectivelyinto the wisdom of the crowd [17].We can call this tendency ‘stronginformational conformity’. It is generallythought to be uniquely human, buta recent experiment could demonstratethis effect in both chimpanzees andchildren, where naıve observers weremore likely to copy the action of threerole models seen once each than anaction performed by one role modelthree times [18].

In humans, we see a processthat goes one step further, in thatindividuals prefer to comply withthe local social norm, or are forced todo so by punishment [19]. This is‘normative conformity’. Both stronginformational and normativeconformity produce marked localhomogeneity. Interestingly,experimental evidence for stronginformational conformity amonghumans is mixed, perhaps because weonly feel obliged to comply withnormative conformity for specificcategories of cultural variantsand private information maysometimes trump the wisdom ofthe crowd [17,20].

If future work confirms thesepatterns, it is likely that the distinctivelyhuman normative conformity [19] wasenabled by the strong informational

conformity that our ancestors inheritedfrom the last common ancestor with thetwo chimpanzee species (commonchimpanzees and bonobos), becauseit put in place the mechanism ofattending to what the majority wasdoing. As usual, if we take a closer look,seemingly unique human traits can bedecomposed into shared and uniqueelements with distinct functions andhistories. In the process, we oftenlearn more about human nature andhuman evolution. It may turn out thatthis will also happen in the case ofcultural conformity.

References1. Whiten, A. (2011). The scope of culture in

chimpanzees, humans and ancestral apes.Phil. Transact. Roy. Soc. London 366,997–1007.

2. Luncz, L.V., Mundry, R., and Boesch, C. (2012).Evidence for cultural differences betweenneighboring chimpanzee (Pan troglodytes)communities. Curr. Biol. 22, 922–926.

3. Laland, K.N., and Janik, V.M. (2006). The animalcultures debate. Trends Ecol. Evol. 21,542–547.

4. Slagsvold, T., and Wiebe, K.L. (2011). Sociallearning in birds and its role in shaping theforaging niche. Phil. Trans. Roy. Soc. London366, 969–977.

5. van Schaik, C.P. (2009). Geographicvariation in the behavior of wild great apes: isit really cultural? In The Question of AnimalCulture, K.N. Laland and B.G. Galef, eds.(Cambridge, MA: Harvard University Press),pp. 70–98.

6. Langergraber, K.E., Boesch, C., Inoue, E.,Inoue-Murayama, M., Mitani, J.C., Nishida, T.,Pusey, A., Reynolds, V., Schubert, G.,Wrangham, R.W., et al. (2011). Genetic and‘cultural’ similarity in wild chimpanzees. Proc.Roy. Soc. B 278, 408–416.

7. Gruber, T., Muller, M.N., Strimling, P.,Wrangham, R.W., and Zuberbuhler, K. (2009).Wild chimpanzees rely on cultural knowledgeto solve an experimental honey acquisitiontask. Curr. Biol. 19, 1806–1810.

8. Whiten, A., Goodall, J., McGrew, W.C.,Nishida, T., Reynolds, V., Sugiyama, Y.,Tutin, C.E.G., Wrangham, R.W., and Boesch, C.(1999). Cultures in chimpanzees. Nature 399,682–685.

9. Humle, T., and Matsuzawa, T. (2002).Ant-dipping among chimpanzees of Bossou,Guinea, and some comparisons with othersites. Am. J. Primatol. 58, 133–148.

10. Mobius, Y., Boesch, C., Koops, K.,Matsuzawa, T., and Humle, T. (2008). Culturaldifferences in army ant predation by WestAfrican chimpanzees? A comparative study ofmicroecological variables. Anim. Behav. 76,37–45.

11. Schoning, C., Humle, T., Mobius, Y., andMcGrew, W.C. (2008). The nature of culture:technological variation in chimpanzeepredation on army ants revisited. J. Hum. Evol.55, 48–59.

12. Humle, T., Snowdon, C.T., and Matsuzawa, T.(2009). Social influences on ant-dippingacquisition in the wild chimpanzees(Pan troglodytes verus) of Bossou, Guinea,West Africa. Anim. Cognition 12, S37–S48.

13. Whiten, A., Horner, V., and de Waal, F.B.M.(2005). Conformity to cultural norms of tool usein chimpanzees. Nature 437, 737–740.

14. Whiten, A., Spiteri, A., Horner, V., Bonnie, K.E.,Schapiro, S.J., and de Waal, F.B.M. (2007).Transmission of multiple traditions withinand between chimpanzee groups. Curr. Biol.17, 1–6.

Current Biology Vol 22 No 10R404

15. Perry, S. (2009). Conformism in the foodprocessing techniques of white-faced capuchinmonkeys (Cebus capucinus). Anim. Cognition12, 705–716.

16. Galef, B.G., and Whiskin, E.E. (2008).‘Conformity’ in Norway rats? Anim. Behav. 75,2035–2039.

17. Efferson, C., Lalive, R., Richerson, P.J.,McElreath, R., and Lubell, M. (2008).Conformists and mavericks: the empirics offrequency-dependent cultural transmission.Evol. Hum. Behav. 29, 56–64.

18. Haun, D.B.M., Rekers, Y., and Tomasello, M.(2012). Majority-biased transmissionin chimpanzees and human children,but not orangutans. Curr. Biol. 22,727–731.

19. Schulman, G.I. (1967). Asch conformity studies:conformity to the experimenter and/or thegroup? Sociometry 30, 26–40.

20. McElreath, R., Lubell, M., Richerson, P.J.,Waring, T.M., Baum, W., Edsten, E.,Efferson, C., and Paciotti, B. (2005). Applyingevolutionary models to the laboratory study

of social learning. Evol. Hum. Behav. 26,483–508.

Anthropological Institute and Museum,University of Zurich, 8057-CH Zurich,Switzerland.E-mail: vschaik@aim.uzh.ch

DOI: 10.1016/j.cub.2012.04.001

DNA Replication: Pif1 Pulls the Plugon Stalled Replication Forks

The conserved PIF helicase family appears to function in replication to ensuretermination and passage through regions that slow or arrest replication forkmovement. Findings in fission yeast extend evidence from budding yeast,and argue for universal mechanisms that ensure replication integrity.

Kenji Shimada1,*and Susan M. Gasser1,2

All good things must come to an end,including the replication fork. Andwhenit does, one hopes that the terminationwill be a happy one. Yet there can alsobe dangers along the way, includingreplication pause sites that shouldnot prematurely signal termination.Replication fork pause sites arenumerous and include 5S and tRNAgenes, centromeres, silent chromatinloci, highly transcribed RNApolymerase II (PolII) genes andreplication fork barriers (RFBs) inthe rDNA [1–5]. It is critical that thereplication machinery remainsengaged while the impediments tofork progression are cleared away.Moreover, since stalled replicationforks are susceptible to doublestrand breaks, mechanisms areneeded to prevent this by stabilizingthe replication fork. Such mechanismsprevent spontaneous chromosomebreakage and rearrangements,which are harbingers of oncogenictransformation.

Two recent papers shed light on themolecular machinery that ensures botha safe transition through fork pausingsites and a proper termination ofreplication forks as they converge [6,7].Specifically, they highlight the role ofthe Pif1 family of DNA helicases,which were originally shown in buddingyeast to promote fork passage at

hard-to-replicate sites, and to ensurethe completion of replication whenreplication forks meet [6,7].

Pif1 belongs to a superfamily of50 to 30 directed helicases, found inall eukaryotes, as well as someprokaryotes [8]. In both budding andfission yeast, Pif1 helicase has roles inboth nuclear and mitochondrial DNAreplication, and the two roles can beseparated by genetic manipulation ofthe PIF1 gene [9,10]. Saccharomycescerevisiae actually harbors two Pif1family members, Rrm3 as well as Pif1.The two have distinct and sometimesopposing roles, even though bothcontribute to proper genomeduplication [11]. For example, Rrm3facilitates replication fork passage atrDNA RFBs, while Pif1 promotes forkstalling at these sites [11]. Moreover,Rrm3 promotes replication forkprogression at telomeres, while Pif1promotes fork progression throughG-quadruplex motifs, and at the sametime antagonizes telomerase-mediatedelongation, apparently by displacingtelomerase from its template [12–14].Since the Schizosaccharomycespombe genome encodes only one Pif1helicase family member (pfh1+), thequestion arose which of thefork-related activities characterized inS. cerevisiae would be maintained inthis distantly related yeast.

The unique Pif1 helicase in fissionyeast, Pfh1, has an essential function inboth chromosomal and mitochondrial

replication in S. pombe [10]. Therefore,to examine its role in genomicreplication, the two laboratoriesemployed special methods todown-regulate, but not completelyablate, this essential gene. On onehand, Sabouri et al. [6] depleted Pfh1by shutting off its expression witha thiamine-repressed promoter,while Steinacher et al. [7] used afunction-specific mutant, the pfh1-mt*allele, in which Pfh1 retained itsmitochondrial function but wasexcluded from the nucleus [10]. Thelatter studied replication fork pausingprimarily on replicating plasmids in thefission yeast cells, while Sabouri et al.studied endogenous chromosomalloci. Nonetheless, the two studiescame to a consensus: Pfh1, likethe S. cerevisiae Rrm3, facilitatesprogression when the fork slows orencounters a barrier and plays animportant role in ensuring propertermination at converging replicationforks.Sabouri and colleagues [6] mapped

Pfh1 distribution genome-wide inunperturbed cells using chromatinimmunoprecipitation. They showedthat Pfh1 occupancy is specificallyenriched at the rDNA RFB, themating-type locus RFB (RTS1), 5SrRNA and tRNA genes, all of which areknown to be replication fork pausesites in S. pombe. They also foundevidence for enrichment at highlytranscribed PolII genes. They thenshowed a significant increase of forkstalling at Phf1-enriched sites, wherethey also detected signs of enhancedDNA damage (gH2AX). Damage andpausing increases when Pfh1 isdepleted, suggesting that Pfh1promotes replication fork passage andsuppresses breaks at hard-to-replicatesites. Interestingly, they also detectedlower but significant Pfh1 binding at