Chater and Christiansen present a compelling alternative to the “heroic individual” or “deliberate design” accounts of cultural evolution (Claidière, Scott-Phillips, & Sperber, Reference Claidière, Scott-Phillips and Sperber2014; Henrich & McElreath, Reference Henrich and McElreath2003). Their concept of social tinkering provides a precise characterization of the myopic and ad hoc process by which human institutions, languages, and norms are formed. However, we argue that for social tinkering to occur, there must first be a stable yet dynamic structural substrate upon which these innovations can act. This substrate is not merely a sum of individuals and relationships but rather collective social niche construction through self-organized social network dynamics (Sueur, Reference Sueur2023; Sueur & Deneubourg, Reference Sueur and Deneubourg2025). The integration of collective social niche construction and adaptive social networks enables the extension of Chater and Christiansen’s framework, facilitating the explanation of the biological and evolutionary continuity that precedes explicit human social rules.

The target article proposes that social tinkering is characterized by the gradual accumulation of innovations to social rules that coordinate behavior. We propose that before the emergence of such rules – whether explicit or implicit – regularities manifest through self-organization. In animal societies, social structures (networks of associations, grooming, and agonistic interactions) are not static backdrops (Borgeaud et al., Reference Borgeaud, Sosa, Sueur and Bshary2017; Sosa et al., Reference Sosa, Jacoby, Lihoreau and Sueur2021); instead, they are collectively computed by the individuals within them (Brush, Krakauer, & Flack, Reference Brush, Krakauer and Flack2018). These structures emerge from local interactions in which individuals adjust their associations based on feedback, such as social tolerance, prestige, or homophily (Sueur et al., Reference Sueur, Romano, Sosa and Puga-Gonzalez2019). This process represents the most elementary stage of social tinkering: the modification of social network topology itself.

This tinkering with network topology creates a social niche. In a manner analogous to how organisms engage in ecological niche construction by modifying their physical environments (Laland, Matthews, & Feldman, Reference Laland, Matthews and Feldman2016; Odling-Smee, Laland, & Feldman, Reference Odling-Smee, Laland and Feldman1996), social animals engage in collective social niche construction by modifying their social environments (Sueur & Deneubourg, Reference Sueur and Deneubourg2025). This constructed niche – the network – subsequently functions as a filter and support for behavioral variation. As observed in many species, the propagation of behaviors, information, and disease follows network topology (Allen et al., Reference Allen, Weinrich, Hoppitt and Rendell2013; Battesti et al., Reference Battesti, Pasquaretta, Moreno, Teseo, Joly, Klensch, Petit, Sueur and Mery2015; MacIntosh et al., Reference MacIntosh, Jacobs, Garcia, Shimizu, Mouri, Huffman and Hernandez2012). Thus, the network is not merely a consequence of behavior; it is a selective environment that biases which innovations persist and become available for subsequent modification. In this view, the spontaneous order attributed by Chater and Christiansen to human systems is already present in the social network organization of non-human animal societies (Sueur, Reference Sueur2023).

A central claim advanced by Chater and Christiansen is that social tinkering involves directed innovation rather than random variation, yet without long-term foresight – a property they term myopia. We offer a mechanistic explanation for this directionality without foresight by examining the relationship between collective decision-making and social networks (Almaatouq et al., Reference Almaatouq, Noriega-Campero, Alotaibi, Krafft, Moussaid and Pentland2020; Maeda et al., Reference Maeda, Sueur, Hirata and Yamamoto2021; Sueur, Deneubourg, & Petit, Reference Sueur, Deneubourg and Petit2012). In many biological systems, conflicts of interest among individuals can enhance the collective computation or emergence of social networks (Brush et al., Reference Brush, Krakauer and Flack2018). Theoretical models derived from first principles demonstrate that the aggregation of imperfect, often conflicting information from individuals can produce group-level outcomes that are more accurate and adaptive than those achievable by any single member (Couzin et al., Reference Couzin, Ioannou, Demirel, Gross, Torney, Hartnett, Conradt, Levin and Leonard2011). This suggests that innovations in social tinkering do not require individual cognitive insight or explicit negotiation; rather, they are filtered through the feedback loops of a self-organized system that favors coordination.

Furthermore, we suggest that the entrenchment of innovations is accomplished through network constraints. Positive feedback loops in social systems, analogous to those observed in the recruitment patterns of social insects (Bonabeau et al., Reference Bonabeau, Theraulaz, Deneubourg, Aron and Camazine1997) or the formation of co-cultures between different species (Sueur & Huffman, Reference Sueur and Huffman2024), can propel systems toward critical thresholds, resulting in phase transitions to new organizational levels – such as the transition from disorganized groups to highly efficient, scale-free network structures observed in primates (Sueur et al., Reference Sueur, King, Pelé, Petit, Gilbert, Kirkilionis and Nicolis2013). These structural phase transitions provide the self-organized scaffolding necessary for momentary innovations to persist as precedents, enabling social tinkering to generate cumulative cultural complexity without requiring centralized design or long-term foresight.

The social tinkering perspective has significant implications for the Cultural Brain Hypothesis (Muthukrishna et al., Reference Muthukrishna, Doebeli, Chudek and Henrich2018). While Chater and Christiansen focus on how social tinkering shapes human cognition, we argue that adaptive social networks themselves constitute a primary target and driver of evolution. Research indicates that social capital – defined as an individual’s position and investment in a social network – functions as an independent dimension of fitness, affecting health, longevity, and reproductive success (Sueur et al., Reference Sueur, Quque, Naud, Bergouignan and Criscuolo2021). When social tinkering improves network efficiency in information flow or stress-buffering, it creates selective pressure for cognitive traits that better favor network adaptation. This establishes a positive feedback loop linking cultural evolution at the network level to genetic evolution, in a manner consistent with multilevel selection acting simultaneously on individuals and social structures.

Chater and Christiansen propose a distinction between social and non-social forms of tinkering, suggesting they require different kinds of learning. The collective social niche perspective, however, suggests a more unified view. In both cases, the environment – whether a physical tool or a social network – is modified, and those modifications alter the learning constraints for subsequent generations. The child-as-scientist metaphor may therefore be more accurately reframed as a child-as-niche-dweller, wherein the child learns to navigate a pre-constructed social topology already containing the precedents for innovation and coordination. In non-human animals (Sueur, Pelé, & Huffman, Reference Sueur, Pelé and Huffman2025), from the stone-handling traditions of Japanese macaques to medicinal plant use transmitted across generations, tinkering is embedded in the social interactions that render these behaviors learnable and culturally transmissible (Huffman, Leca, & Nahallage, Reference Huffman, Leca, Nahallage, Nakagawa, Nakamichi and Sugiura2010a; Huffman et al., Reference Huffman, Spiezio, Sgaravatti and Leca2010b).

We endorse Chater and Christiansen’s departure from design-based models of culture, yet situating social tinkering within collective social niche construction reveals that cultural complexity has deeper roots than the target article suggests. Cumulative culture does not begin with the first ad hoc rule; it is preceded by the first self-organized regularity in a social network. Social tinkering is thus the sophisticated, derived manifestation of an ancient biological process: the collective construction of the social world.