The target paper's main thesis – that cumulative technological culture originates primarily not in social learning but in technical-reasoning skills – is consistent with results obtained with two computer models, both of which show that cumulative cultural evolution is possible in the absence of social learning (albeit at a slower pace) but not in the absence of mental operations akin to reasoning or creative cognition (Gabora Reference Gabora, Nadel and Stein1995; Reference Gabora, Sloutsky, Love and McRae2008). Variants of the target paper's thesis have been proposed elsewhere (see Overmann & Coolidge Reference Overmann and Coolidge2019). A competing theory not addressed in the target paper is that the cultural accumulation of both technical-reasoning skills and enhanced social skills stemmed from the emergence of some more basic cognitive ability. This theory enjoys support from psychology, anthropology, archeology, neuroscience, and genetics, and multiple versions of it (some not mutually inconsistent) have been proposed. Hauser et al. (Reference Hauser, Chomsky and Fitch2002) attribute cumulative culture to the capacity for recursion, as does Corballis (Reference Corballis2011), who also emphasizes mental time travel: the capacity to think about events not occurring in the present. Others have argued that the general cognitive ability underlying cumulative culture was the onset of a self-triggered recall and rehearsal loop (Donald Reference Donald1991), relational reinterpretation (Penn et al. Reference Penn, Holyoak and Povinelli2008b), conceptual fluidity (Mithen Reference Mithen1996b), conceptual blending (Fauconnier & Turner Reference Fauconnier and Turner2008), or something Chomsky (Reference Chomsky, Freidlin, Otero and Subizarreta2008) called “merge.” Our own two-step theory attributes cumulative culture to the onset of representational redescription followed by the capacity to shift between the convergent and divergent modes of thought, culminating in the emergence of an integrated internal model of the world (Gabora Reference Gabora and Leung2018; Reference Gabora2019; Reference Gabora, Gelfand, Chiu and Hong2020; Gabora & Smith Reference Gabora and Smith2018; Reference Gabora, Smith, Henley, Kardas and Rossano2019; Smith et al. Reference Smith, Gabora and Gardner-O'Kearny2018). Thus, although Osiurak and Reynaud (O&R) position their technical-reasoning theory as the only alternative to social explanations for cumulative technological culture, they fail to consider theories that attribute it to the onset of a more general cognitive ability. Many of the arguments O&R put forward in support of their theory are also compatible with, and supportive of, theories that attribute cumulative culture to a cognitive ability that paved the way for complex cognition in both the social and technological domains.

The authors highlight the distinction between sequential mechanical actions and combined mechanical actions, and between combined mechanical actions and genuine innovations (e.g., when they write “innovation in humans might primarily result from technical combinations rather than from novel inventions”). However, novelty does not depend on whether or not the elements are sequential (after all, notes of a song are sequential), nor is it something so simple as whether or not they are combined. The degree of novelty depends on the structure of the combination. The idea that nothing is truly new because innovation merely involves combining pre-existing elements was discredited decades ago with the discovery of emergent properties in concept (or word) combinations (Osherson & Smith Reference Osherson and Smith1981), which have been shown to be not just present, but ubiquitous (Hampton Reference Hampton1987; Storms et al. Reference Storms, De Boeck, Van Mechelen and Ruts1998). Indeed, there is a field dedicated to studying, empirically (e.g., Scotney et al. Reference Scotney, Schwartz, Carbert, Saab and Gabora2020) and mathematically (e.g., Aerts & Gabora Reference Aerts and Gabora2005a; Reference Aerts and Gabora2005b; Aerts & Sozzo Reference Aerts and Sozzo2014; Bruza et al. Reference Bruza, Kitto, Ramm, Sitbon, Song and Blomberg2012) the kinds of structure that emerge in combinations.

Throughout the paper, the authors espouse a sharp distinction between social and asocial learning (e.g., they write, “social vs. asocial learning”). However, consider the following scenarios for how a child learns to peel a banana: (1) by watching a sibling peel a banana, (2) by watching a monkey peel a banana, (3) by watching a cartoon monkey peel a banana, (4) by watching the petals of a smiley-faced cartoon tulip unfold, and (5) by watching the petals of a real tulip unfold. Where did we cross the line between social and asocial? One is forced to view social and asocial learning as ends on a continuum. The authors also assume that imitation and emulation are uniquely associated with social learning, but ask children in a theater or dance class to imitate leaves blowing in the wind and they know exactly what to do. (Indeed, efforts to emulate nature have given rise to much of what constitutes human culture.)

Related to this is a confusion in the paper between individual learning and creative cognition. Individual learning involves obtaining pre-existing information from the environment through asocial means (e.g., learning by oneself the distinctions between different kinds of butterflies), whereas creative cognition involves generating ideas, behavior, or artifacts that did not previously exist (Gabora & Tseng Reference Gabora and Tseng2017). This is important; supplying raw information is not the same as mental operations on this information. Individual learning and creative cognition contribute to cumulative culture in distinct, yet, complementary ways: the former (along with social learning) provides data about the world (e.g., discovery of electricity), and the latter brings something new into the world (e.g., invention of the flashlight). The distinction enables us to demarcate transition points in the evolution of complex cognition and in trajectories of actual technological lineages (Gabora & Steel Reference Gabora and Steel2017; Reference Gabora and Steelunder review; Gabora et al. Reference Gabora, Leijnen, Veloz, Lipo, Carlson, Hőlscher and Shipley2011; Veloz et al. Reference Veloz, Temkin, Gabora, Miyake, Peebles and Cooper2012).

The authors curiously state that “working memory is not a cognitive mechanism that is used to generate content,” but if so then where is content generated? Although incubation, intuition, and subconscious processing play a role in creative cognition (e.g., Bowers et al. Reference Bowers, Farvolden and Mermigis1995), the notion that generative capacities do not require working memory contradicts decades of research on the psychology of creativity. The authors also refer to “trial-and-error strategies that are not random but reasoned,” but if the learning is “reasoned” then by definition it is not “trial and error.”

It would be interesting to test the authors' hypothesis that “opaque” artifacts require more social learning for their transmission. Their notion of “opacity” is reminiscent of Bateson's (Reference Bateson1979) notion of affordances, except that affordances arise dynamically in the interaction between observer and observed. We believe this distinction is important; those who contribute most to culture may be those who see possibilities that others miss.