Group-level phenotypes are idiosyncratic yet unpredictable over development in a clonal fish

Behavioral differences among groups can have profound impacts on individual fitness,¹ yet how group-level phenotypes arise and change during development is poorly understood. Group-level differences may originate from genetic differences or subtle variation among individuals,^2,3,4,5 but they can also emerge over time through social feedback mechanisms within and between groups, potentially producing dramatically different group-level behavior.^6,7,8 Disentangling these drivers of group phenotypes is challenging because groups are typically studied after they are already well established. Using the naturally clonal Amazon molly, we establish replicate groups of genetically identical individuals and continuously track group-level behavior from birth throughout the first 55 days of life. This design allows us to control for genetic and environmental sources of variance, thereby establishing a null baseline for group-level differences. Furthermore, by recording over an extended period beginning from day 1 of life, we test whether group phenotypes follow predictable trajectories driven by individual differences or instead diverge due to social feedback during development. We find that average behavioral changes across groups follow predictable patterns during this early developmental phase. In contrast, differences among groups are highly unstable over longer timescales: although groups differ from very early in life, it appears that the rank-order of group-level phenotypes undergoes a shuffling mid-development, around 4-5 weeks of age. These results suggest that distinct processes generate group-level variation across ontogeny, with early differences largely reflecting consistent individual variation; then as individuals mature and become more socially responsive, social feedback pushes groups along divergent, less predictable behavioral trajectories.