Extended Data Fig. 6: Initial choice behavior and position effects in Fig. 5d-f experiments.

a, The untrained spider P. labiata (n = 25) showed no initial orient preference between blackened/unblackened mantises, while trained spiders (n = 23) preferred blackened mantises. The relative positions of the praying mantis had no effect on the results. a’, The attention duration is not affected by the positions of the mantis. b, The first approach of untrained (n = 30) or black flower model trained Pieris rapae (n = 18) showed no preference between Redboy-RNAi and wild-type (WT) orchid mantises, while white flower model trained butterflies exhibit significant preference to WT orchid mantises (n = 16). This preference was irrelevant with the positions of the mantis. b’ The total number of approaches of trained or untrained butterflies show no relevant with the positions of the mantis. c, The number of chicks trying the flower model gradually decreased with the training, suggesting the effectiveness of the training. c’, Untrained chicks (n = 8) showed no initial attack preference, while white/black-flower-trained chicks (each n = 12) preferred Redboy-RNAi orchid mantises. This preference was irrelevant with the positions of the orchid mantis. c”, The total number of attacks of trained or untrained butterflies show no relevant with the positions of the orchid mantis. Box plots depict the median, upper and lower quartiles, and maximum and minimum values, with outliers represented by dots. *P < 0.05; **P < 0.01; ***P < 0.001. Source data are provided in Source Data Extended Data Fig. 5. a’, LMM statistical results are provided in Supplementary Table 4. b’ and c”, LMM statistical results are provided in Supplementary Table 5. a, b, and c’, GLMM statistical results are provided in Supplementary Table 6. c, Kruskal‒Wallis test or One-way ANOVA statistical results are provided in Supplementary Table 7.