Dilemmas and trade-offs in the diffusion of conventions

Lucas Gautheron · 2025

Read Dilemmas and trade-offs in the diffusion of conventions by Lucas Gautheron (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-17300.txt.

Introduction

Lucas Gautheron studies how shared conventions spread in a 2025 paper. Away from ideal settings, rival forces shape which convention wins. Gautheron names three trade-offs at the heart of that spread.

First, people weigh sameness with self, with sequence, and with setting when they pick a side. Second, local spread through contacts fights global pull from outside the net. Third, groups trade best outcomes against the cost of choosing when tastes clash. A statistical physics frame lets data reveal the game, the nets, and the clash paths at work. A physics sign case shows the method, and leadership gets a noted role.

Catalog record

CREATORLucas Gautheron
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE56 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2501-17300

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Dilemmas and trade-offs in the diffusion of conventions Lucas Gautheron Abstract Outside ideal settings, conventions are shaped by heterogeneous competing processes that can challenge the emergence of norms. In order to acknowledge this complexity, this paper develops a generalized account of conventions and identifies three trade-offs involved in their diffusion: (I) the trade-off between the imperatives of social, sequential, and contextual consistency that individuals balance when choosing between conventions; (II) the competition between local (bottom-up) and global (top-down) coordination, depending on whether individuals coordinate their behavior via interactions throughout a social network or external factors transcending the network; and (III) the balance between decision optimality (e.g., collective satisfaction) and decision costs when collectives with conflicting preferences choose a convention. A broadly applicable statistical physics framework for exploring these trade-offs is developed and applied to a sign convention in physics. The method can infer the structure of the underlying coordination game, the networks of social interactions involved, and the processes through which conflicts are resolved. This shows that the purpose of conventions may exceed coordination, and that individual preferences towards conventions are concurrently shaped by cultural factors and multiple social networks. Finally, this work emphasizes the role of leadership in the resolution of conflicts. 0 0 footnotetext: Evolution, Science and Society, University of Missouri, Columbia, United States 0 0 footnotetext: Interdisciplinary Centre for Science and Technology Studies (IZWT), University of Wuppertal, Germany Keywords : conventions; collective cognition; collective behavior; cultural evolution. 1 Introduction Since David Lewis ( Lewis2002 ) , conventions (including linguistic norms, technological or manufacturing standards, and many other social norms) are primarily conceived as solutions to coordination problems ( hawkins2019emergence ) . Yet, the attitude of individuals towards conventions involves a multitude of factors beyond social coordination, resulting in tensions that may disrupt the emergence of a universal norm. To acknowledge this complexity, this paper develops a generalized statistical physics account of conventions and identifies three trade-offs involved in their diffusion and the resolution of conflicts in the absence of consensus. The first trade-off is the balance between i) social consistency (driven by coordination with peers), ii) sequential consistency (driven by the cost of switching from one practice to another), and iii) contextual consistency (driven by a need for consistency with other choices or cultural traits) (§ 1.2 ). The second trade-off involves the balance between local (or bottom-up) versus global (or top-down) coordination, depending on whether individual preferences are formed endogenously through local interactions on a social network, or by factors transcending the network structure (or both, in possibly contradicting ways) (§ 1.3 ). Finally, the last trade-off is the balance between decision costs and the optimality of the outcome in the resolution of conflicts (§ 1.4 ). To illustrate these trade-offs, we apply a statistical physics framework to behavioral data about a sign convention in physics. This statistical framework allows us to retrieve information about individuals’ decision-making, the structure of the underlying coordination problem, and the multiple infrastructures (whether social or cultural) involved in the propagation of a convention. First, we show that scientists’ attitude is shaped by sequential consistency, as they tend to maintain a preferred choice in their solo-authored publications independently of the target research area (§ ). Then, we show that scientists’ preferences are correlated – albeit imperfectly – with those of their co-authors, which means that some level of social coordination is achieved (§ ). In order to explain how, the relative contribution of local coordination (via dyadic interactions with peers) and global coordination (i.e. via shared culture) is measured by applying the Ising model to the authors’ collaboration and citation network. This shows that both local and global processes contribute to coordinating scientists’ preferences. Third, we assess the plausibility of three mechanisms of preference-formation according to their ability to explain the observed magnitudes of local and global coordination, and find slightly more evidence for a model of cultural transmission involving the imitation of peers (§ ). Finally, we infer the process through which scientists resolve conflicts about which convention to use in collaborations (§ ). We find evidence that the last author’s preference most often prevails, thus highlighting the role of seniority and power in the resolution of conflicts. Taken together, these results indicate that decision-making processes related to conventions involve multiple and sometimes conflicting factors. 1.1 Background In relativistic physical theories (such as general relativity and quantum field theory), the “metric tensor” is a mathematical object that represents the metric properties of space-time. Broadly speaking, the metric tensor can take either of the two following forms: \begin{pmatrix}+1&0&0&0\\ 0&-1&0&0\\ 0&0&-1&0\\ 0&0&0&-1\\ \end{pmatrix}\text{ or }\begin{pmatrix}-1&0&0&0\\ 0&+1&0&0\\ 0&0&+1&0\\ 0&0&0&+1\\ \end{pmatrix} The first choice (+,-,-,-) is known as the mostly minus convention while the second choice, (-,+,+,+) is referred to as the mostly plus convention. These choices are regarded as physically equivalent and lead to identical predictions. However, depending on which convention is chosen, certain quantities arising in calculations will take either positive or negative values. Interestingly, there is no norm and both conventions are used. This creates a puzzle: if, as per Lewis, scientific conventions address an imperative of social coordination, why physicists persist in making opposite choices? The puzzle points to a gap in the existing theoretical toolkit: while formal models have provided rich insights into norms and conventions, they may have left out crucial aspects of reality by stripping away too much of its complexity ( Elsenbroich2013 ) , or by neglecting the interactions between phenomena studied in isolation 1 1 1 For instance, ( Delgado2002 ; pujol2005role ) demonstrated the sensitivity of conventions to the topological features of the underlying social networks (including their small-world, scale-free or clustering properties). . Similarly, while controlled experiments can uncover certain aspects of conventions in idealized settings ( Guala2010 ; Bruner2014 ; Centola2015 ; Hawkins2016 ; Formaux2021 ; hawkins2023partners ) , they may conceal the fact that complex heterogeneous processes and multiple social infrastructures can drive or prevent the emergence of conventions in naturalistic situations ( Boyce2024 ) . Fortunately, the advent of large online communities has opened up opportunities to investigate the diffusion of real norms and conventions in complex networks ( DanescuNiculescuMizil2013 ; kooti2012emergence ; Heaberlin2016 ; Rotabi2017 ) . Such data-driven approaches, however, have barely extended to the study of scientific conventions 2 2 2 with the exception of ( Rotabi2017 ) , who investigated LaTeX macros naming conventions in scientific papers. . Yet, “conventionalism” can be traced back to Poincaré and his account of the epistemic status of the axioms of geometry ( ben2006conventionalism ; Wilholt2012-WILCPD ) . In fact, conventions are ubiquitous in science ( Wu2023 ) , including statistical practices such as statistical significance thresholds ( wilholt2009bias ) , measurement strategies ( Smaldino2022 ) , and unit systems. By exploring a scientific convention – namely, the metric signature –, this work highlights the interactions between multiple phenomena involved in the diffusion of conventions that prior works have addressed separately or ignored, and provides cues for understanding how conventions can fail to develop into universal norms, in naturalistic settings. This task, as we will now show, calls for an expanded account of conventions. 1.2 The trade-off between social, sequential, and contextual consistency While Lewis’ account of conventions is focused

Read the complete text