Characterizing ontogeny of quantity discrimination in zebrafish

Abstract

A sense of non-symbolic numerical magnitudes is widespread in the animal kingdom and has been documented in adult zebrafish. Here, we investigated the ontogeny of this ability using a group size preference (GSP) task in juvenile zebrafish. Fish showed GSP from 21 days post-fertilization and reliably chose the larger group when presented with discriminations of between 1 versus 3, 2 versus 5 and 2 versus 3 conspecifics but not 2 versus 4 conspecifics. When the ratio between the number of conspecifics in each group was maintained at 1 : 2, fish could discriminate between 1 versus 2 individuals and 3 versus 6, but again, not when given a choice between 2 versus 4 individuals. These findings are in agreement with studies in other species, suggesting the systems involved in quantity representation do not operate separately from other cognitive mechanisms. Rather they suggest quantity processing in fishes may be the result of an interplay between attentional, cognitive and memory-related mechanisms as in humans and other animals. Our results emphasize the potential of the use of zebrafish to explore the genetic and neural processes underlying the ontogeny and function of number cognition.

Quantity as a Fish Views It: Behavior and Neurobiology

“…The consistent failure to select the larger group between 2 and 4 individuals, whether controlled for space occupied or not, suggests that the fish did not only rely on the overall space as 4 fish occupied two times the space. Hence, the authors suggested that zebrafish may use both numerical and non-numerical information to represent quantities (Agrillo et al, 2016;Leibovich et al, 2017;Sheardown et al, 2022) and that attentional constraints and cognitive and working memory mechanisms may orchestrate numerical competence as hypothesized in both humans and other animals (Hyde, 2011). However, these results may also support the hypothesis of distinct quantification systems characterized by domain and task specificity operating largely independently from the others (Feigenson et al, 2004;Miletto Petrazzini et al, 2014).…”

“…Zebrafish are physiologically homologous to mammals and possess all major neurotransmitters, hormones, and receptors (Panula et al, 2006;Alsop and Vijayan, 2009). The high degree of protein and genetic homology with humans (Howe et al, 2013), coupled with refined gene-editing tools and behavioral paradigms, make this species a vertebrate system amenable to large-scale forward genetic analyses (Wolman et al, 2011;Kalueff et al, 2013;Sheardown et al, 2022). Transparency of embryos and larvae enables in vivo functional imaging of neural activity and establishes the zebrafish as a powerful optogenetic tool.…”

“…Altogether, these results suggest a limited role of maturation and experience on zebrafish's ability to estimate areas that parallel the existence of tectal neurons by selectively responding to item size already during the first week of life (see paragraph below for more details). This high accuracy seems to be in contrast with performance observed in juvenile zebrafish tested in a spontaneous choice test to investigate the ontogeny of numerical competence (Sheardown et al, 2022). In this study, the authors adapted the procedure previously used to investigate shoal choice discrimination in angelfish, in which the stimulus fish were located in individual compartments that limited movement and avoided fish hiding each other.…”

Decoy effect in shoaling decision making in zebrafish(Danio rerio)

“…However, we see evidence for the preference for the larger shoal both in the case of 4 versus 2 and 6 versus 3 dichotomous trials through trajectory density analysis, where the 9/14 more dense regions were near the larger shoal. Previous studies on shoaling preference report that zebrafish can discriminate between shoal sizes of 6 versus 3 in dichotomous choices, although many studies indicate a failure to discriminate 4 versus 2 shoal sizes 39,51 .However, both those studies also use a different rectangular three-tank shoal choice set up from ours, and one 39 used a perforated divider through which shoaling decisions were based on olfactory and not just visual cues from the display fish.…”

“…This difference could be because Potrich et al (2015) used only females in the display shoals, whereas we randomized the sex of the fish in the display shoal. Many other shoal discrimination studies have not tested for sex based differences in shoaling behaviour 39,54 . Most shoal choice studies also use a different apparatus from us - a rectangular three-tank set up, and one study 39 used a perforated divider through which shoaling decisions were based on olfactory and not just visual cues from the display fish.…”