Binary or nonbinary? An evolutionary learning approach to gender identity
Hung Truong · 2024
Read Binary or nonbinary? An evolutionary learning approach to gender identity by Hung Truong (2024) 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-2412-10959.txt.
Introduction
Hung Truong asks whether gender identity is best understood as binary or nonbinary, and he treats the question with game theory and a learning model. People are assumed to choose an identity while also hoping to match well with others, an idea drawn from the classic battle of the sexes game. In that setup, a world with only binary identities and a world with only nonbinary ones can each persist once established, because no single person gains by switching alone.
Truong then uses a genetic learning method to see which outcome a changing population drifts toward. Starting from a binary world, people who take a nonbinary identity turn out to match more flexibly, grow in number, and become the majority. The author offers the framework as a simple tool for studying how diversity and exclusion can arise in different economic settings.
Catalog record
| CREATOR | Hung Truong |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-10959
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Binary or nonbinary? An evolutionary learning approach to gender identity Hung Truong Abstract Is gender identity binary or nonbinary? My analysis shows that while both are possible, the latter is a more attracting equilibrium under an adaptive learning perspective. I frame the gender identity problem as a modified battle of the sexes game, where individuals define their gender identity under pairwise matching motives. From a baseline game-theoretical standpoint, I demonstrate that the binary-only world and the nonbinary-only world are both Nash equilibria in the stage game and are locally stable in the infinitely repeated game. Thus, any state of gender identity could theoretically persist. I then adopt a genetic learning algorithm as an equilibrium selection criterion to investigate evolutionary dynamics further and provide a rationale for the transition from binary to nonbinary gender identity. Specifically, in a binary-origin world, divergence occurs as individuals identifying as nonbinary gender evolve to become the majority due to their higher flexibility in matching outcomes. My framework captures how adaptive learning drives identity evolution, offering a parsimonious tool to analyze how diversity and exclusivity emerge in varying economic environments. keywords gender identity, social behavior, social norms, genetic algorithms, evolutionary game theory, simulations. 1 Introduction Understanding the evolution of social identities, including gender identity, is critical to explaining how diversity and exclusivity emerge in various economic and social contexts. While traditional economic models often treat identity as fixed or exogenous, a growing body of interdisciplinary research emphasizes that identity is malleable and influenced by social and economic interactions ( akerlof2000economics ; shayo2020social ) . In particular, gender identity—whether binary or nonbinary—raises fundamental questions about how individuals define themselves and interact within their environments. This paper contributes to this literature by providing a novel framework to analyze the dynamics of gender identity evolution and its implications for social coordination and diversity. Social identity theory underscores the interplay between individual preferences and group affiliation, suggesting that identity formation is shaped by the incentives and trade-offs inherent in social and economic structures. In a recent survey of the literature, shayo2020social discusses works on social identity and economic policy, and demonstrates how individuals’ identification with different social groups can drive preferences over redistributive policies. Similarly, the framework developed in this paper explores how individuals’ gender identity evolves in response to pairwise matching motives, reflecting both individual utility maximization and broader social dynamics. The conceptual foundation of this paper lies in game theory, specifically a modified version of the classic battle of the sexes game. The game theoretical setup captures the tension between coordination and individual preferences in defining gender identity. At a baseline game-theoretical standpoint, both binary and nonbinary gender identity worlds are Nash equilibria of the stage game, and are locally stable then the game is infinitely repeated. These findings align with the theoretical prediction that multiple equilibria can coexist, contingent on the social and economic environment ( young2020individual ) . To move beyond stage equilibrium analysis, this paper introduces a genetic learning algorithm ( holland1992genetic ) as an equilibrium selection criterion, drawing inspiration from evolutionary game theory and adaptive learning models ( sandholm2010population ) . The genetic algorithm models how individuals adapt and evolve their gender identities over time based on the payoffs from pairwise matching. The results reveal a striking transition dynamic: starting from a binary-origin world, the population evolves toward a predominance of nonbinary gender identity. This transition is driven by the higher flexibility and adaptability of nonbinary identities, which enhance matching outcomes and social coordination. The broader implications of these findings are twofold. First, they provide a rationale for the observed increase in nonbinary gender identification in contemporary societies, highlighting the role of adaptive learning and social dynamics. Second, the framework offers a parsimonious tool for analyzing the emergence of diversity and exclusivity across different economic and social environments. By situating the analysis within the broader context of social identity theory and evolutionary dynamics, this study bridges the gap between theoretical modeling and real-world phenomena. Related literature This analysis builds on and contributes to a rich tradition of research on social identity, including foundational works by akerlof2000economics ; chen2011potential ; benjamin2010social ; charness2020social ; shayo2020social . It also draws methodological inspiration from macroeconomic equilibria and adaptive learning models ( arifovic2000evolutionary ; arifovic1997transition ; arifovic2023history ) , providing a novel lens to understand identity evolution in dynamic and diverse environments. Research on conformity to ingroup norms demonstrates that individuals adapt their behaviors to align with those of their ingroup, influenced by social salience, primed group membership, and shared threats ( abrams1990knowing ; van2013self ; lakin2003using ; benjamin2010social ; hoff2006discrimination ) . This aligns with the framework in my study, highlighting how social identity dynamics shape adaptive learning and identity evolution. Notably, chen2011potential present insights on how salient group identity fosters efficient coordination through learning, which corroborate my framework by illustrating the role of identity-driven dynamics in shifting equilibria, reinforcing the evolutionary transition from binary to nonbinary gender identity. This paper is structured as follows. Section 2 details the environment framework and the use genetic learning algorithm as what drive the learning process of individuals. Section 3 introduces the theoretical model and outlines the game-theoretical properties of the stage game. Section 4 presents the results from the genetic algorithm simulations and discusses the implications of the transition dynamics from binary to nonbinary gender identity. Finally, Section 5 concludes by situating the findings within the broader literature on social identity and offering directions for future research. 2 Framework 2.1 The “world” The computerized environment in this analysis consists of N heterogeneous individuals. Each individual is represented by a chromosome encoded as a binary string of length 3\times l genes/bits. For instance, with l=10 , a chromosome would be represented as: \underbrace{010110111}_{\text{type }}\underbrace{0100010110}_{\alpha}\underbrace{1101010100}_{\beta} The first l bits represent the “type” of an individual, which is inherently stochastic and subject to no evolutionary processes beyond randomness. In other words, the type of an individual is naturally determined by random initialization. The remaining 2\times l genes form the biological interpreter, which translates the type into a corresponding gender identity. These 2\times l genes undergo evolutionary processes that reflect natural selection, where gene strings associated with higher fitness are more likely to be selected for reproduction. I refer to these components as the type-chromosome, \alpha -chromosome, and \beta -chromosome strings. Each of these strings serves a distinct role in determining how an individual’s type is translated into gender identity. Figure 1 summarizes the processes that govern the interactives within the population. \underbrace{\text{ born }\rightarrow\text{ gender identification }\rightarrow\text{ matching }\rightarrow\text{ fitness computation }}_{\text{intra-generation }} Figure 1: Intra-generation activities First, individuals are born, and each identifies their gender. Matching then occurs, allowing individuals to obtain payoffs based on the matching outcomes. Specifically, as previously mentioned, an individual is born with a chromosome that consists of a stochastic type and a biological interpreter, which may be inherited from prior generations. For gender identification, the binary strings are decoded, normalized, and processed by the biological interpreter to produce a gender identity. Matches are then established according to specific rules, which I will outline in the following sections. The procedures for decoding gender identity are detailed below. The strings are decoded as follows. First, the three chromosome