Mating System And Behavioral Game In Ground Tits In Southeastern Tibet

Among socially monogamous birds, both males and females would seek to mate with extra-pair mates to maximize fitness benefits. In cooperatively breeding species, however, a high-level extra-pair paternity would reduce relatedness among group members, which may limit the evolution of helping behavior. Therefore, there is an evolutionary trade-off between maximizing fitness and stabilizing sociality.The ground tit Pseudopodoces humilis is a passerine endemic to the Tibetan Plateau. With complicated social and genetic structure, the birds provide a good chance for us to test the evolutionary hypotheses about formation and maintaince of social system and to explore the behavioral game. The field work was conducted from 2013 to 2014 in Qamdo (30° 33'13" N,97° 06'31" E,4300 m), in southeast Tibet. Sex and paternity of individuals were analyzed by molecular techniques. Major results are summarized as follows:1. Social system. There were 52% of onsets (n= 45) with helpers, including one plural breeding nest, and the remaining 48%(n= 41) without them. A cooperatively breeding nest contained 1.79 helpers (range:1-6) on average. The majority of helpers are males (68) and the minority are females (9). Male helpers were closely relatives of the dominant breeders (95.5% of 66), and a few were unrelated individuals (4.5%). The pedigree data of 2014 showed that 22.9% of male helpers (n-35) were offspring that delayed dispersal. Of the 9 female helpers,7 were related and 2 unrelated with the breeders. For a given nest, helpers may be males only (36 of 45), females only (3), and both sexes (6). Most of helpers (83.1% of 77) only helped in 2013, with 11.7% helped in 2013 and 2014, and the rest of 5.2% bred independently in 2014.2. Genetic mating system. Extra-pair paternity was detected from 51.2% of the 43 bi-parental nests, with 23.5% of the 162 offspring involved; extra-pair maternity was found in 25.6% of the 43 nests, with 10.5% of the 162 offspring involved. Among the 41 helped nests,31.7% contained offspring (9.9% of 161) with extra-group extra-pair paternity; male helpers shared paternity in 36.6% of the 41 nests and produced 18.6% offspring (30 of 161). Extra-pair maternity was detected from 22.0% of the 41 helped nests (6.2% of 161 offspring). There were 3 nests (2.5% of 161 offspring) in which female helpers shared reproduction. Of the 41 helped nests, one nest included one offspring whose both parents were one neighboring pair.3. Social system evolution based on kin selection. We found that females tended to choose extra-pair mates that had a higher level of phenotypic expression than their social mates (P=0.067), and these extra-pair mates were significantly more related with them than with their social mates (P< 0.001). Cooperatively breeding nests had significantly lower extra-group extra-pair paternity than bi-parental nests (P= 0.038), and there was no difference in proportion of extra-pair maternity between the two social systems. Male offspring of delayed dispersal did not sire offspring. The relatedness between helpers which had sired offspring and primary males was significantly higher than that between helpers without offspring and primary males (P = 0.027). Whether helpers had access to group paternity depended on their relatedness with the primary pairs (males, P= 0.032; females, P= 0.018). However, how much reproduction was partitioned by helpers relied on the relatedness between helpers and the primary females (P-0.017). Primary males (2.73±0.23 offspring, n= 40 nests) shared significantly higher paternity than did male helpers(1.36±0.16, n=22) (P< 0.001). These results are consistent with kin selection theory, which predicts that selection would favor to cooperative breeding if the benefit of cooperation outweighs the cost of mating with extra-group individuals.4. Cooperative breeding evolution from a perspective of game theory. A game arose in the ground tit social system:while individuals adopted cooperative strategy to prevent the loss of parentage, they should search extra-pair mating to maximize fitness so that cuckolding and cuckolded individuals got into a "prisoners dilemma". In the ground tits, defectors (only cuckolding individuals) gained a greater benefit in terms of the numbers of offspring (males,4.44; females,3.89), and the one who obtained extra-pair parentage would benefit more (P< 0.001). In contrast, cooperators (only cuckolded individuals) suffered greater fitness loss (males,2.30; females,2.92). When both players took a cooperative strategy, they would not gain the greatest benefit (males,3.15; females,3.39). If the partner defected and the one who defected as well, they would gain a maximum pay-off respectively (males,4.15; females,4.31). In the "prisoners dilemma" game, individuals of both sexes adopted different reproductive strategies that directly affected individual fitness. While cuckoldry benefited the defector with additional reproductive pay-off but punished the loyal one by loss of parentage. Betrayal became the optimum strategy for themselves and led a high-level of extra-pair parentage in Qamdo. While the "Tit for Tat" strategy permitted the reciprocation of benefits for both players, and TFT was the best evolutionarily stable strategy for "Nash equilibrium".5. Helper effect. The presence of helpers significantly decreased incidence of nestling loss within cooperative groups (P= 0.04), while (1) helpers did not increase the success of breeding (P= 0.24):the number of fledging in bi-parental pairs (3.83± 0.23) and cooperative groups (3.85±0.20) were not different significantly (P= 0.94); (2) helping did not promote female breeders to adjust clutch size, which was independent of either help status (P= 0.23) or helper numbers (P-0.12); (3) helper effect did not improve hatching rate (P= 0.86); (4) helping did not contribute to the survival of breeders. The quality of individuals would lead them into a dilemma between "independent breeding" and "help". Therefore, in the Qamdo population of ground tits, we concluded that the quality of helpers was significantly lower than that of primary males (P< 0.001), which may be a cause of "help-at-the-nest", as the low-quality helpers could not always bring positive effect, which was also a reason why we had not detect the significant reproductive contribution of helpers to the cooperative groups.