Female zebrafish (Danio rerio) demonstrate stronger preference for established shoals over newly-formed shoals in the three-tank open-swim preference test

Abstract

Zebrafish (Danio rerio) share a considerable amount of biological similarity with mammals, including identical or homologous gene expression pathways, neurotransmitters, hormones, and cellular receptors. Zebrafish also display complex social behaviors like shoaling and schooling, making them an attractive model for investigating normal social behavior as well as exploring impaired social function conditions such as autism spectrum disorders. Newly-formed and established shoals exhibit distinct behavior patterns and inter-member interactions that can convey the group's social stability. We used a three-chamber open-swim preference test to determine whether individual zebrafish show a preference for an established shoal over a newly-formed shoal. Results indicated that both sexes maintained greater proximity to arena zones nearest to the established shoal stimulus. In addition, we report the novel application of Shannon entropy to discover sex differences in systematicity of responses not revealed by unit-based measurements; male subjects spent more time investigating between the two shoals than female subjects. This novel technique using established versus newly-formed shoals can be used in future studies testing transgenics and pharmacological treatments that mimic autism spectrum disorder and other disorders that affect social interaction.

Fig 1. Experimental apparatus.

Illustrative example with the subject in the center tank to test preference between an established shoal on the right and a newly-formed shoal on the left. The apparatus consists of three identical 18.9L (5.5 gal.) glass aquaria (each 20.3 cm wide × 25.4 cm high × 40.6 cm long; total width = 121.8 cm) filled with water to a depth of 21 cm.

Fig 2. Heatmaps collapsed across trials with the same configuration of stimuli.

Warm colors indicate highest intensities of localization while cooler colors indicate lowest intensities of localization.

Fig 3. Percent cumulative duration in each quadrant zone across both sexes.

Quadrants are labeled as upper/lower sections of the tank and novel/established stimulus shoal proximity. Error bars are +/- SEM. While exact p-values are reported for post-hoc comparisons, p-values smaller than 0.001 are indicated as p < 0.001.

Fig 4

Percent duration in motion (A) and motionless (B) by zone across both sexes. Quadrants of the subject tank were labeled by upper/lower sections of the tank and novel/established stimulus shoal proximity. Error bars are +/- SEM. While exact p-values are reported for post-hoc comparisons, p-values smaller than 0.001 are indicated as p < 0.001.

Fig 5

Average velocity (A) and velocity SD (B) during movement. Quadrants of the subject tank were labeled by upper/lower sections of the tank and novel/established stimulus shoal proximity. Error bars are +/- SEM. While exact p-values are reported for post-hoc comparisons, p-values smaller than 0.001 are indicated as p < 0.001.

Fig 6. Sex differences in inter-quadrant and cross-tank transitions.

Quadrants of the subject tank were labeled by upper/lower sections of the tank and novel/established stimulus shoal proximity. Error bars are +/- SEM. While exact p-values are reported for post-hoc comparisons, p-values smaller than 0.001 are indicated as p < 0.001.

Fig 7. Sex differences in behavioral entropy.

H₃ measures entropy across three vertical zones while H₄ measures entropy across four quadrants. Error bars are +/- SEM. While exact p-values are reported for post-hoc comparisons, p-values smaller than 0.001 are indicated as p < 0.001.