The ability to discriminate between larger and smaller quantities has been demonstrated in several mammalian and avian species suggesting the possibility of evolutionary conservation of this characteristic. Preference for the larger of two groups has also been shown in fish species, although this ability has rarely been systematically studied in lower order vertebrates, and thus the mechanisms of such ability are not understood. Here, we exploit the tendency of angelfish to seek protection in an unfamiliar environment by joining a group of conspecifics, a behaviour called shoaling. Test fish were given a simultaneous choice between shoals varying both in terms of numerical ratios and absolute numbers of fish. Our results provide evidence for quantity discrimination in angelfish. In general, experimental subjects chose the larger of two shoals. Furthermore, in agreement with Weber's law, which holds that discrimination between two quantities depends on their ratio, the discrimination between shoals of different quantities of fish was more difficult when the shoal sizes became more similar. The limit of discrimination ratio was found to be below 2:1. Briefly, angelfish are able to discriminate between different quantities of conspecifics subject to a ratio limit, a finding that implies a fitness component in this behaviour similar to what has been demonstrated in higher order vertebrates.

The ability to quantify, i.e. to estimate quantity, may provide evolutionary advantages in some contexts and has been demonstrated in a variety of animal species. In a prior study, we showed that angelfish (Pterophyllum scalare) were able to discriminate between groups (shoals) in which a large number of conspecifics swam preferring to join the larger of the two. Our results implied that angelfish can compare relative shoal sizes likely on the basis of some quantitative attributes of the shoal. Here, also using a binary preference test, we examined whether angelfish are able to discriminate between shoals of small numbers of conspecifics, and if so whether their performance reveals a comparable underlying mechanism to that proposed for discrimination of small quantities in human and non-human animals, namely the possible precursor of the ability to count. Our results demonstrate that fish reliably chose 4 versus 1, 3 versus 1, 2 versus 1 and 3 versus 2 individuals, but were at chance performance level when having to choose between 4 versus 3, 5 versus 4 and 6 versus 5. Findings also reveal that the density of the fish in the stimulus shoals did not significantly affect the performance of experimental angelfish. These results are compatible with the hypothesis of the existence of an object-file mechanism to discriminate small quantities in vertebrates and provide evidence for spontaneous discrimination of up to three elements in angelfish, a similar limit to that found in human and non-human animals. The findings add to the growing body of data, suggesting that the mechanisms underlying discrimination between different quantities of items may be shared across different taxa and have an evolutionary ancient origin.

Previous studies investigating quantity discrimination have shown that angelfish are able to select the larger of two groups of conspecifics (shoals). The discrimination limits shown by angelfish were similar to those found for other vertebrates when large (≥4) and small quantities (<4) were presented. However, in these studies, no attempt was made to control for non-numerical features of the stimulus shoals and thus the question whether numerical or some quantitative attributes of the shoals were utilized for making the choices could not be answered. Here, we investigate whether angelfish can discriminate between shoals differing in numerical size using non-numerical attributes. We systematically manipulate density, inter-fish distance, and overall space occupied by the shoals, one factor at a time, and analyse the choices angelfish made between the contrasting stimulus shoals. The stimulus shoals consisted of contrasts between large (10 vs. 5) and small (3 vs. 2) number of conspecifics. We found density to be a sufficient condition for discrimination between large shoals as the test subjects preferred the more dense shoal. Manipulation of inter-fish distance indicated that this variable is not a necessary factor in discrimination at either shoal size contrast. Likewise, we found that the size of space occupied by the contrasted shoals also did not significantly influence discrimination. Sensitivity to the density of large shoals indicates that angelfish can discriminate shoal size using this non-numerical cue. Nevertheless, the factors we examined may represent only a subset of all possible non-numerical features upon which angelfish may base their discrimination. Thus, we suggest that further research is required to clarify whether and under what circumstances angelfish may use numerical or non-numerical features when discriminating between shoals of differing size.

The zebrafish may represent an excellent compromise between system complexity and practical simplicity for behavioral brain research. It may be particularly appropriate for large scale screening studies whose aim is to identify mutants with altered phenotypes or novel compounds with particular efficacy. For example, the zebrafish may have utility in the analysis of the biological mechanisms of learning and memory. Although learning and memory have been extensively studied and hundreds of underlying molecular mechanisms have been identified, this number may represent only the fraction of genes involved in these complex brain functions. Thus large scale mutagenesis screens may have utility. In order for such screens to succeed, appropriate screening paradigms must be developed. The first step in this research is the characterization of learning and memory capabilities of zebrafish and the development of automatable tasks. Here we show that zebrafish is capable of latent learning, i.e. can acquire memory of their environment after being allowed to explore it. For example, we found experimental zebrafish that experienced an open left tunnel or an open right tunnel of a maze during the unrewarded exploration phase of the test to show the appropriate side bias during a probe trial when they had to swim to a group of conspecifics (the reward). Given that exploration of the maze does not require the presence of the experimenter and the probe trial, during which the subjects are video-recorded and their memory is tested, is short, we argue that the paradigm has utility in high throughput screening.

Human infants and non-human animals can discriminate the larger of two sets of discrete items. This quantity discrimination may be based upon the number of items, or upon non-numerical variables of the sets that co-vary with number. We have demonstrated that angelfish select the larger of two shoals of conspecifics without using inter-fish distance or space occupied by the stimuli as cues. However, density appeared to influence the choice between large shoals. Here, we examine the role of another non-numerical cue, swimming activity of the stimulus fish, in quantity discrimination by angelfish. To control this variable, we varied the water temperature of the stimulus aquaria or restricted the space occupied by each fish in the stimulus shoals. We used the previously successfully discriminated contrasts consisting of large (10 vs. 5) and small (3 vs. 2) shoals. We also studied whether more active or less active shoals are preferred in case of equally sized shoals (10 vs. 10, 5 vs. 5, and 3 vs. 3). When differences in stimulus fish activity were minimized by temperature manipulation we found angelfish to prefer the larger shoal in the 3 vs. 2 comparison, but not in the 10 vs. 5 comparison. When activity was controlled by space restriction, angelfish preferred the larger shoal in both numerical contrasts. These results imply that the overall activity level of the contrasted shoals is not a necessary condition for small shoals discrimination in angelfish. On the other hand, the results obtained for the large shoals, together with results obtained in the control treatments (equal numerical contrasts and differing activity levels), suggest that activity is a sufficient condition for discrimination when large shoals are involved. Further experiments are needed to evaluate the influence of other continuous variables, and to assess whether the mechanisms underlying performance are comparable to those suggested for other animals.