Is heightened-shoaling a good candidate for positive emotional behavior in zebrafish?

Abstract

Zebrafish, a highly-social species of freshwater fish, are widely studied across many fields of laboratory science including developmental biology, neuroscience, and genomics. Nevertheless, as standard housing for zebrafish typically consists of small and simplistic environments, less is known about their social behavioral repertoire in more naturalistic settings. In particular, spontaneously occurring, socio-positive affiliative behaviors (e.g., social coordination and cohesion) that may be indicative of positive emotional experiences have rarely been reported or studied deliberately in zebrafish. Housing adult zebrafish (10 fish/tank) in large semi-natural tanks (110 L; n = 6) with sloping gravel substrate, rocks, and artificial plants, we observed a previously undescribed behavior: Distinct periods of spontaneous, synchronized, compact aggregations, what we call “heightened-shoaling”. This project aimed to quantify the characteristics of this distinctive behavior and compare parameters of heightened-shoaling to baseline periods (normal behavior) and pre-feed periods (feed-anticipatory behavior). First, across 4 days, we selected video-clips (100 s each) from within (i) instances of heightened-shoaling (n = 9), (ii) baseline periods (n = 18), and (iii) pre-feed periods (n = 18). For each of these video clips, we scan sampled every 10 s to determine fish orientations and location within the tank and agonistic behavior. Next, we used an all-occurrence sampling method to record the timing and duration of all episodes of heightened-shoaling behavior when tank-lights were on (8:00 h to 18:00 h) across 10 days. From the scan-sampling data, we found that compared to baseline periods, heightened-shoaling was characterized by increased shoal cohesion (p < 0.0001), increased adherence to the horizontal plane (p < 0.0001), decreased agonism (p < 0.0001), and no diving behavior (lower positions within the water column signal negative effect in zebrafish, p > 0.1). From the all-occurrence sampling data, we found 31 episodes of heightened-shoaling with instances observed in all six tanks and only a single case in which heightened-shoaling occurred in two tanks at the same time. The median episode duration was 7.6 min (Range 1.3–28.6). As the first systematic description of heightened-shoaling behavior, this research contributes to our knowledge of the range of zebrafish social dynamics living in naturalistic environments. Moreover, as a spontaneously occurring, protracted, affiliative behavior, heightened-shoaling appears to be a good candidate for future research into positive emotional behavior in zebrafish.