Fig. 3

Average change in shoaling densities (mean ± SE) of 39-day old fathead minnows as induced by a mechanical predator disturbance showcasing (a) maternal × paternal × offspring risk interactions and (b) biparental × caring parent × presence of parental care risk interactions. The dashed line is the zero referent and represents no change in shoaling density. In a, empty dots represent offspring low-risk whereas filled dots represent offspring high-risk environments. In b, offspring were always exposed to low-risk environments only; hence empty dots in b represent the absence of parental care whereas filled dots represent the presence of parental care. Within fish drawings, black fish indicate individuals that were from birth onwards exposed to conspecific alarm cues, white fish refer to the ones that instead received a water control treatment; for more detail see Fig. 1. Black circles and error bars represent treatments shown only once across figures, for every other color, same-colored markers refer to the same treatment across figures. Dashed error bars in b highlight the treatments that were cross-fostered. Different letters above bars indicate statistical differences at p < 0.05 according to post-hoc tests from linear mixed-effect models that also contained all remaining covariates (final model in Table 2). Raw values are plotted in Additional file 1: Fig. S1