Male cooperation improves their own and kin-group productivity in a group-foraging spider

Abstract

Cooperation should only evolve if the direct and/or indirect benefits exceed the costs. Hence, cooperators are expected to generate selective benefits for themselves and the kin-group while defectors will impose costs. The subsocial spider, Australomisidia ergandros, shows consistent cooperation and defection tactics while foraging. Cooperative individuals are consistently likely to share prey with other group members whereas defector spiders rarely share the prey they acquired. Here, we assess costs and benefits of cooperation, and the causal determinants behind cooperative and defective phenotypes. We constructed experimental kin-colonies of A. ergandros composed of pure cooperative or defector foragers and show that pure cooperative groups had higher hunting success as they acquired prey more quickly with greater joint participation than pure defector groups. Importantly, defectors suffered higher mortality than cooperators and lost considerable weight. A social network approach using subadult spiders revealed that foraging tactic is sex dependent with males cooperating more frequently than females. Our results provide a rare empirical demonstration of sex-specific male cooperation that confer individual and kin-group benefits.

Figure 1

Formation of experimental groups from selected nests (N_nests = 10) shown for one nest as an example. We formed two to three initial groups per nest for ‘phase 1’, the assessment of individual feeding strategy. We then re-grouped individuals according to their feeding strategy (cooperator or defector, marked green and brown respectively) in the initial groups. Grey spiders did not show pronounced cooperator or defector behaviours. In the sorted groups, we tested for group composition effects on social foraging behaviour and individual fitness payoffs.

Figure 2

Group composition effects on social foraging behaviour and individual fitness payoffs. A, B: The effects of group composition (cooperative group ‘c’ or defector group ‘d’) on attack latency (A) and the extent of prey sharing (B); only successful attacks are considered (c: N = 27, d: N = 21). The boxplots show median, upper and lower quartiles and interquartile range (1.5 times). C: The effect on individual weight gain (c: N = 38, d: N = 31) presented as mean (circle) ± SE (whiskers). * indicates significant difference between group compositions (see also Table 1).

Figure 3

A social network showing the differences in cooperative tendencies between males and females in A. ergandros. The graph shows the foraging interactions in one exemplary group of ten males (green nodes) and ten females (brown nodes) recorded over ten repeated feeding trials. A line between two nodes indicates that one individual shared its prey with the other, the respective arrow specifies who acquired food for whom. The number of outgoing arrows per node reflects the food acquisition and sharing tendency of the particular individual in terms of spread over different group members. The node size reflects the frequency with which the individual acquired and shared prey over the duration of the feeding experiments.