Behavioral and endocrine responses to noninteractive live and video conspecifics in males of the Siamese fighting fish

Abstract

The physiological mechanisms underlying variation in aggression in fish remain poorly understood. One possibly confounding variable is the lack of standardization in the type of stimuli used to elicit aggression. The presentation of controlled stimuli in videos, a.k.a. video playback, can provide better control of the fight components. However, this technique has produced conflicting results in animal behavior studies and needs to be carefully validated. For this, a similar response to the video and an equivalent live stimulus needs to be demonstrated. Further, different physiological responses may be triggered by live and video stimuli, and it is important to demonstrate that video images elicit appropriate physiological reactions. Here, the behavioral and endocrine responses of male Siamese fighting fish Betta splendens to a matched-for-size conspecific fighting behind a one-way mirror, presented live or through video playback, were compared. The video playback and live stimulus elicited a strong and similar aggressive response by the focal fish, with a fight structure that started with stereotypical threat displays and progressed to overt attacks. Postfight plasma levels of the androgen 11-ketotestosterone were elevated as compared to controls, regardless of the type of stimuli. Cortisol also increased in response to the video images, as previously described for live fights in this species. These results show that the interactive component of a fight and its resolution are not needed to trigger an endocrine response to aggression in this species. The study also demonstrates for the first time in a fish a robust endocrine response to video stimuli and supports the use of this technique for researching aggressive behavior in B. splendens.

Keywords: Betta splendens; aggression; androgens; corticosteroids; one-way mirror; video playback.

Figure 1.

Experimental setup used to test (A) the response to a live conspecific fighting its mirror image or (B) to a video playback of an equivalent stimulus; (C) footage for the video playback was obtained by filming behind a one-way mirror a fish fighting its image. An opaque smart screen that becomes transparent when activated was placed between the one-way mirror and both the focal and the live stimulus tank and turned on after the acclimation period. Trials were recorded with top and side cameras for behavioral analysis.

Figure 2.

(A) Duration of open opercula, (B) frequency of bites, and (C) frequency of air breathing in the first (0–15 min) and second (15–30 min) halves of the aggression trials. Focal fish were presented with a conspecific fighting a one-way mirror, live (one-way mirror), or as video images (video playback). Mean ± SE are shown. * represent significant differences (P < 0.027) between the first and second halves of the trial. The live and video stimuli triggered similar responses in the focal fish (P > 0.407).

Figure 3.

Posttest plasma levels of 11-ketotestosterone (KT) and cortisol (F) in fish presented for 30 min with a conspecific fighting a one-way mirror, live (one-way mirror), or as video images (video playback). Control fish were presented with either an empty tank or with a video playback of the empty tank. Mean ± SE are shown. Different letters represent significant differences (P < 0.05) between groups.

Figure 4.

Representation of the first 2 components of a principal component analysis with all the measured endocrine and behavioral variables: (A) individual data—large symbols represent the centroid of the groups; (B) variable loadings—the direction of arrows represents the correlation of variables and size their relative contribution. Numbers in the axis’s legend represent the percentage of variation explained by each component.