Transient polymorphisms in parental care strategies drive divergence of sex roles

Parental care patterns vary widely across taxa and even within species. Classic theories link sex-role differences to anisogamy, operational or adult sex ratios, and sexual selection, yet these factors are interdependent and difficult to disentangle analytically. Previous analytical frameworks, notably those of Kokko and Jennions, predict egalitarian or continuum outcomes for biparental care under symmetry but rely on simplifying assumptions. This study asks whether individual-level processes and emergent variation can explain the emergence of sex-biased care, the role of sexual selection, and the apparent evolutionary lability of sex roles, even in constant environments. The purpose is to reassess predictions of sex-role theory using individual-based evolutionary simulations that naturally incorporate variation, coevolution, and stochasticity.

Debates have centered on: (1) anisogamy as a driver of higher female post-zygotic care (Trivers) versus critiques emphasizing future costs/benefits; (2) the role of operational versus adult sex ratios in shaping care; and (3) how sexual selection interacts with parental investment. Analytical work by Kokko and Jennions provided a general framework but required fixed parameters for sexual selection and monomorphic assumptions; later corrections (Fromhage and Jennions) altered some conclusions. Empirical and phylogenetic studies show frequent transitions among care patterns. Prior models incorporating biparental synergy suggest egalitarian outcomes. However, behavioral polymorphisms (“animal personalities”) and individual variation, known to affect evolution, are often neglected in analytical treatments of sex roles.

The authors implemented an individual-based evolutionary simulation with overlapping generations and daily time steps. Individuals (female/male) occupy one of four states: juvenile, pre-mating, mate search, and caring. Baseline daily mortality in all states is 0.001 (expected lifespan ~1000 days), unless varied. Juvenile maturation lasts 20 days. Pre-mating periods are sex-specific and can be zero (baseline) or positive; during pre-mating, individuals cannot mate. In the mate search state, individuals encounter opposite-sex partners; in random mating, encounters always result in mating; with mate choice, females probabilistically accept males based on male ornament and female preference. Upon mating, both partners enter the caring state and provide care for durations T_f (female) and T_m (male), heritable quantitative traits subject to rare small-effect mutations (Mendelian inheritance). If a parent dies during care, actual care duration is truncated at death. Offspring survival depends on total parental effort T_tot via S(T_tot) = T_tot/(T_tot + B^2) with B=20. Total effort is additive (T_tot = T_f + T_m) in the baseline; biparental synergy is introduced as an additional positive interaction term (T_tot = T_f + T_m + synergy_parameterT_fT_m), with synergy varied (e.g., 0, 0.05, 0.2, ≥2.0). Density dependence regulates population size by scaling offspring survival to S(T_tot)/(1 + γN) with γ=0.003, producing populations of ~2000 females and ~2000 males. The model tracks life history and reproduction without explicit fitness calculations; successful strategies propagate via their offspring. For joint evolution with sexual selection, male ornament size s and female preference p are heritable. The probability a female with preference p mates with a male of ornament s follows a logistic function: mating probability = (1 + χexp(αp*s))^−1, with χ=0.02 and α=2; male ornamentation and female choosiness incur survival or time costs, respectively. Analytical comparisons used selection gradient methods and adaptive dynamics to identify a branching point near T_f = T_m = 5, consistent with simulation dynamics.

• In symmetric, random mating scenarios (no sex differences), simulations never stabilized at egalitarian care or a continuum of equilibria. Instead, they consistently converged to one of two alternative stable equilibria with strongly sex-biased care (either female-biased or male-biased), each yielding total care ≈ B=20 (e.g., around T_f,T_m ≈ (17.5,2.5) or (2.5,17.5)). Initial symmetric conditions led to either equilibrium with ~50/50 probability. Analytical selection gradients erroneously suggested a line/curve of equilibria. - Mechanism: sexual conflict induces disruptive selection and transient within-sex polymorphisms. Populations first move toward low egalitarian care (≈5,5), an evolutionary branching point, then develop bimodal distributions within each sex (no-care and high-care strategies). Association of high-care with one sex and low-care with the other breaks the polymorphism, yielding sex-role divergence. - Parental synergy: With weak synergy (e.g., 0.05), simulations still produced alternating male- or female-biased care, with rapid switches and persistent polymorphisms; brief egalitarian phases were transient. With moderate synergy (≈0.20), egalitarian average care emerged but remained low (T_f ≈ T_m ≈ 5), yielding total effective care ≈15, less than the efficient total (20) seen without synergy; both sexes remained polymorphic. Only very high synergy (≥2.0) produced stable egalitarian care at levels satisfying the total-care target (T_f + T_m + synergyT_fT_m = B). - Evolutionary lability: Over long time scales, populations switch between the two asymmetric equilibria even in constant environments. Transitions are faster with smaller populations or weakened selection (e.g., longer pre-mating periods, higher mortality). Switches pass through the branching region (≈5,5) with transient polymorphisms before settling into the alternative equilibrium. - Joint evolution of mating and parental strategies: Two recurring outcomes: (1) male-biased care with near-random mating (low female preference and minimal male ornamentation), and (2) female-biased care with strong female preference and elaborate male ornamentation. In all examined transitions, changes in parental care preceded changes in sexual selection traits, indicating that parental role evolution drives sexual selection, not vice versa. - Pre-mating investment asymmetry: The sex with higher pre-mating investment often evolved higher post-zygotic care, especially when pre-mating mortality was high (5× mate-search mortality). When pre-mating mortality was zero, the sex with longer pre-mating time still tended to invest more, plausibly via sex-ratio effects (Fisher condition) and altered trade-offs between current and future reproduction. At intermediate mortality levels and small asymmetries, outcomes were mixed and sometimes reversed, indicating complex causation beyond simple Trivers-type arguments.

The simulations demonstrate that even under complete symmetry and random mating, sexual conflict over care leads to disruptive selection, transient within-sex polymorphisms, and ultimately stable sex-role divergence. This directly addresses and challenges analytical predictions of egalitarian or continuum outcomes, revealing that emergent polymorphisms can steer evolution toward strongly biased parental roles. Weak to moderate biparental synergy does not guarantee efficient or stable egalitarian care; instead it can maintain polymorphisms and drive rapid switching between sex roles, providing an intrinsic mechanism for the evolutionary lability observed in phylogenies without invoking environmental change. The joint-evolution results indicate causality from parental roles to sexual selection: mating preferences and ornaments emerge or disappear contingent on which sex provides more care. Asymmetries in pre-mating investment can bias outcomes toward greater care by the higher-investing sex, but intermediate regimes yield non-monotonic patterns, highlighting the complex interplay of life-history schedules, sex ratios, and selection gradients. Overall, recognizing transient polymorphisms as a central driver reconciles discrepancies between simulation and analytical predictions and offers an explanation for frequent evolutionary transitions in parental and mating roles.

This work shows that transient within-sex polymorphisms in parental strategies can intrinsically generate strong sex-role divergence and evolutionary lability, contradicting standard analytical predictions that assume monomorphism. Key contributions include: (1) identification of an evolutionary branching point at low egalitarian care that seeds polymorphism and drives sex-role specialization; (2) demonstration that weak to moderate biparental synergy does not necessarily stabilize or optimize biparental care; (3) evidence that parental roles causally precede and shape sexual selection dynamics; and (4) nuanced insights into how pre-mating investment asymmetries affect post-zygotic care. Future research should expand simulation frameworks to broader parameter regimes (e.g., different B values), incorporate additional life-history asymmetries, state-dependent decisions, spatial structure, environmental variation, and richer sexual selection components (good genes, direct benefits), and link model outputs to comparative and experimental data on polymorphisms and role switching.

Simulation outcomes depend on explored parameter spaces and stochasticity. Although tens of thousands of runs were conducted, untested regions may yield different results (e.g., results were derived for B > 10; extrapolation to B < 10 is uncertain). Computational demands limit exhaustive exploration, and some intermediate-regime patterns (e.g., the pre-mating investment effects at intermediate mortalities) lack a fully convincing mechanistic explanation. Analytical support (adaptive dynamics) required assumptions (sex-independent care trait) that cannot predict sex-role emergence. Persistent polymorphisms and switching behaviors may be sensitive to population size, density dependence, and mutation structure, which were specified but not varied exhaustively.