Androgens and corticosteroids increase in response to mirror images and interacting conspecifics in males of the Siamese fighting fish Betta splendens

Hormone levels increase in response to video playback in the Siamese fighting fish

“…The duration of threat displays (average time with opercula open, 211 s vs 372 s) and of attacks (average frequency of bites, 66 vs 95) was also higher when dynamic feedback was available (compare Table 2 in this study with Table 3 in [11]). Accordingly, average post-fight plasma levels of KT (8.0 vs 10.6 ng/ml) and F (6.2 vs 13.0 ng/ml) were higher when both males could see each other (compare Figure 3 in this study with Figure 5 in [11]). Although this comparison supports the hypothesis that aggressive behaviour, and associated physiological responses, is stronger in interactive fights, results should be interpreted with caution as these were separate experiments and uncontrolled differences between studies may explain the results.…”

“…While controls in both studies had a similar frequency of air breathing, interactive fights caused a 2.5 to 3 - fold higher frequency of this behaviour as compared with one-way mirror fights, suggesting that they are metabolically more demanding (compare Table 1 in this study with Table 2 in [11]). The duration of threat displays (average time with opercula open, 211 s vs 372 s) and of attacks (average frequency of bites, 66 vs 95) was also higher when dynamic feedback was available (compare Table 2 in this study with Table 3 in [11]). Accordingly, average post-fight plasma levels of KT (8.0 vs 10.6 ng/ml) and F (6.2 vs 13.0 ng/ml) were higher when both males could see each other (compare Figure 3 in this study with Figure 5 in [11]).…”

“…Real-life fights are, by nature, interactive and the absence of dynamic feedback from the opponent, and the impossibility to display the full suite of aggressive behaviours (e.g., circling or chases during fights), may generate a poor response. Although this study was not designed to specifically test the importance of fight interaction for aggressive displays, results can be compared with previously published data from our lab where the aggressive response of males of the same strain, and under similar experimental conditions, were tested with matched for size opponents behind a transparent partition [11]. While controls in both studies had a similar frequency of air breathing, interactive fights caused a 2.5 to 3 - fold higher frequency of this behaviour as compared with one-way mirror fights, suggesting that they are metabolically more demanding (compare Table 1 in this study with Table 2 in [11]).…”

“…Although this study was not designed to specifically test the importance of fight interaction for aggressive displays, results can be compared with previously published data from our lab where the aggressive response of males of the same strain, and under similar experimental conditions, were tested with matched for size opponents behind a transparent partition [11]. While controls in both studies had a similar frequency of air breathing, interactive fights caused a 2.5 to 3 - fold higher frequency of this behaviour as compared with one-way mirror fights, suggesting that they are metabolically more demanding (compare Table 1 in this study with Table 2 in [11]). The duration of threat displays (average time with opercula open, 211 s vs 372 s) and of attacks (average frequency of bites, 66 vs 95) was also higher when dynamic feedback was available (compare Table 2 in this study with Table 3 in [11]).…”

“…Plasma levels of KT and F were measured with competitive enzyme-linked immunosorbent assay (ELISA) kits from Cayman Chemical following the manufacturer’s instructions. A lack of interference in the assay of other immunoreactive molecules for this species and for these ELISA kits had already been confirmed by serially diluting a plasma pool and comparing the slope with that of a standard curve [11]. All standards and samples were measured in duplicate with a dilution in the EIA buffer of 1:150 for KT and 1:20 for F. Experimental samples were measured in the same assay and the intra-assay coefficient of variation, calculated from the sample duplicates, was 2.86 % for KT and 3.20 % for F.…”