In the target article, Stibbard-Hawkes re-evaluates the archeological record of human cultural artifacts through the lens of the cultural repertoires of modern hunter-gatherers. He cautions against using artifacts as a barometer of the cognitive capacities of our extinct hominin ancestors, primarily because of the differential durability of cultural items: Modern hunter-gatherers, for example, have robust cultural repertoires, but relatively few of their tools would be durable enough to be recovered many thousands of years in the future.

We agree with this assessment and aim to push the argument further. While the author suggests that some prehistoric populations may have produced symbolic artifacts that were not preserved, we propose that populations could have had the necessary cognitive abilities and still not produce any such artifacts, or even more broadly not show evidence for cumulative culture. We suggest that cultural innovations, such as symbolic artifacts, are enabled not only by the cognitive skills of the individual but also by the broader cultural context of a population and other populations it contacts.

We previously suggested, using cultural evolutionary models, that both punctuated increases in culture and extensive loss of tools may often be stochastic byproducts of processes of cultural evolution (Creanza, Kolodny, & Feldman, Reference Creanza, Kolodny and Feldman2017a; Feldman & Creanza, Reference Feldman and Creanza2018; Greenbaum, Friesem, Hovers, Feldman, & Kolodny, Reference Greenbaum, Friesem, Hovers, Feldman and Kolodny2019b; Greenbaum et al., Reference Greenbaum, Fogarty, Colleran, Berger-Tal, Kolodny and Creanza2019a; Kolodny, Creanza, & Feldman, Reference Kolodny, Creanza and Feldman2015, Reference Kolodny, Creanza and Feldman2016): Cultural traits spread in a population, they facilitate associated innovations and novel combinations, and they occasionally are lost. These processes result in dynamically fluctuating cultural repertoires, even in cognitively indistinguishable populations and in constant environments. This property of cultural evolution continues to be underappreciated in human history. Such dynamics may further affect populations' cultural repertoires when coupled with demographic or environmental change (Ben-Oren, Jaffe, & Kolodny, Reference Ben-Oren, Jaffe and Kolodny2023a; Ben-Oren, Kolodny, & Creanza, Reference Ben-Oren, Kolodny and Creanza2023b; Ben-Oren, Strassberg, Hovers, Kolodny, & Creanza, Reference Ben-Oren, Strassberg, Hovers, Kolodny and Creanza2023c; Creanza, Kolodny, & Feldman, Reference Creanza, Kolodny and Feldman2017b; Fogarty & Creanza, Reference Fogarty and Creanza2017; Henrich, Reference Henrich2004; Powell, Shennan, & Thomas, Reference Powell, Shennan and Thomas2009; Strassberg & Creanza, Reference Strassberg and Creanza2021). The insight of the target article – that many tool lineages would be unlikely to leave an archaeological record – adds another factor that may increase observed between-population cultural differences that are unrelated to cognitive differences.

The possibility that complex cognition does not immediately lead to the rise of complexity in material culture, sometimes not for tens of thousands of years, is supported by the archeological record (Hovers & Belfer-Cohen, Reference Hovers, Belfer-Cohen, Hovers and Kuhn2006, Reference Hovers and Belfer-Cohen2022; Mcbrearty & Brooks, Reference Mcbrearty and Brooks2000; Paige & Perreault, Reference Paige and Perreault2024; Scerri & Will, Reference Scerri and Will2023): Many hallmarks of modernity, including various manifestations of symbolic thought, appear anecdotally thousands of years before they begin to occur regularly, suggesting that the cognitive capacity for symbolic artifacts greatly preceded their establishment in the record. This is true for ochre (Henshilwood, d'Errico, & Van Niekerk, Reference Henshilwood, d'Errico and Van Niekerk2011), jewelry (Vanhaeren, d'Errico, Stringer, & James, Reference Vanhaeren, d'Errico, Stringer and James2006; Zilhão et al., Reference Zilhão, Angelucci, Badal-García, d'Errico, Daniel, Dayet and Zapata2010), as well as the precocious appearance of architectural features in Africa at ~476 kya (Barham et al., Reference Barham, Duller, Candy, Scott, Cartwright, Peterson and Nkombwe2023) or among European Neanderthals at ~176 kya (Jaubert et al., Reference Jaubert, Verheyden, Genty, Soulier, Cheng, Blamart and Santos2016) followed by a long hiatus until architecture appears again in the archaeological record of the Upper Paleolithic after 50 kya.

Such early instances of symbolism, along with other technologies that are considered advanced (e.g., architecture, figurative art), raise the question of why symbolic artifacts were not widely adopted earlier. Differential preservation may be a factor, as suggested by the target article. We suggest that deeper, systemic cultural processes may also play a role: This phenomenon may be explained by a negative ratcheting effect in which some traits are easier or more likely to be lost than to be acquired. One factor that may affect the probability of an invention to occur, as well as to be retained, is its usefulness. It is possible, thus, that symbolic artifacts had greater utility in some populations than others. It has been suggested, for example, that one early function of symbolism was communication and maintenance of social cohesion across large populations. If that is the case, symbolic artifacts might have been more likely to be established when populations became more interconnected (Bar-Yosef, Reference Bar-Yosef1997; Hovers, Reference Hovers1990; Hovers & Belfer-Cohen, Reference Hovers and Belfer-Cohen2022). Arguably, the establishment of symbolic artifacts could create a feedback loop, allowing for higher social cohesiveness, which in turn favored greater use of symbols.

Finally, we suggest that it may be the case that some culturally related cognitive capacities cannot develop in an individual without appropriate cultural scaffolding at early developmental stages. By way of analogy, just as the development of linguistic skills requires exposure to language in infancy (Lenneberg, Reference Lenneberg1967), it could be that symbolic thought requires early exposure to symbols. Chimpanzees, for example, show naturally no evidence of a capacity for language, yet are cognitively able to learn words in sign language once exposed to them and teach them to other individuals (Gardner & Gardner, Reference Gardner and Gardner1969); similarly, the potential for symbolic thought could have existed in early humans but was unrealized in most populations for extended periods in prehistory. More generally, we suggest that beyond the dependency of many cultural traits on pre-existing traits and cultural context, there may be a far more fundamental challenge for the development and multi-generational retention of certain cultural complexes, such as language or the production of symbolic artifacts.

In conclusion, for any cultural trait to appear in the archeological record, it needs to be invented, adopted, and preserved. We suggest that each of these steps is governed by multiple factors other than the cognitive capacity of the individual, meaning that while symbolic artifacts may be proof of sophisticated cognitive abilities, their absence does not rule out the existence of such abilities.