{"paper_id":"2a443188-e8fa-44a4-b6f8-15f260f9d221","body_text":"Neural Basis of Number Sense in Larval Zebrafish\nPeter Luu1,2, Anna Nadtochiy1,3, Mirko Zanon4,1, Noah Moreno1, Andrea Messina4, Maria Elena\nMiletto Petrazzini7, Jose Vicente Torres Perez7, Matthew Jones1,5, Kevin Keomanee-Dizon1,5,\nCaroline H. Brennan6, Giorgio Vallortigara4, Scott E. Fraser1, Thai V. Truong1\nCorresponding Authors: sfraser@provost.usc.edu, tvtruong@usc.edu\nAffiliation:\n1 Translational Imaging Center, Michelson Center for Convergent Bioscience, University of\nSouthern California, Los Angeles, CA, USA\n2 Department of Biological Sciences, Division of Molecular and Computational Biology,\nUniversity of Southern California, Los Angeles, CA, USA\n3 Department of Biological Sciences, Quantitative and Computational Biology, University of\nSouthern California, Los Angeles, CA, USA\n4 Centre for Mind/Brain Sciences, University of Trento, Rovereto, Italy\n5 Alfred Mann Department of Biomedical Engineering, University of Southern California, Los\nAngeles, CA, USA\n6 Dana and David Dornsife College of Letters, Arts and Sciences, University of Southern\nCalifornia, Los Angeles, CA, USA\n7 School of Biological and Behavioral Sciences, Queen Mary University of London, London,\nUnited Kingdom\nSignificance\n● Numerically responsive neurons can be detected as early as 3 days post-fertilization\n(dpf) in zebrafish\n○ Number-selective neurons preferring quantities greater than one increases after\n3 dpf\n● Neurons that are highly responsive to specific numbers of objects are mainly localized to\nthe forebrain and midbrain\n● Number stimulus can be decoded from Ca 2+ activity and show increased performance\nwith age\n● Administration of ethanol decreases the activity of number-selective neurons in the\nforebrain\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nAbstract\nNumber sense, the ability to discriminate the quantity of objects, is crucial for survival. To\nunderstand how neurons work together and develop to mediate number sense, we used\ntwo-photon fluorescence light sheet microscopy to capture the activity of individual neurons\nthroughout the brain of larval zebrafish (Danio rerio), while displaying a visual number stimulus\nto the animal. We identified number-selective neurons as early as 3 days post-fertilization (dpf)\nand found a proportional increase of neurons tuned to larger (>2) quantities after 3 dpf. To\ndetermine if these neurons are sufficient to encode the correct quantity, we used machine\nlearning to predict the stimulus from the neuronal activity and observed that the prediction\naccuracy improves with age. Given ethanol’s propensity to inhibit cognitive functions, we tested\nits effect on number sense and found a decrease of active number-selective neurons in the\nforebrain during ethanol exposure, suggesting cognitive impairment. The findings here are a\nsignificant step towards understanding neural circuits devoted to discrete magnitudes and our\nmethodology to track single neuron activity across the whole brain is broadly applicable to other\nfields in neuroscience.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nIntroduction\nUnderstanding quantity, either discrete (countable) or continuous, is a fundamental for survival,\ne.g. for avoiding predators, finding food, mating, and other group behaviors (Alder & Rose,\n2000; Cross & Jackson, 2017; Edwards et al., 2002; Templeton & Greene, 2007). This sense of\nquantity, often referred to as the Approximate Number System (ANS), allows both humans and\nanimals to intuitively estimate numerosity, or quantity of objects in a set, without precise\ncounting (Brannon & Merritt, 2011; Piazza, 2010; Piazza et al., 2004). The ANS develops during\nearly infancy, highlighting its importance as a foundational aspect of cognition (Xu & Spelke,\n2000). The ANS forms the basis of survival instincts to complex mathematical abilities,\nultimately shaping how we perceive and interact with the world around us (Bonny & Lourenco,\n2013; Feigenson et al., 2004; Szkudlarek & Brannon, 2017).\nCurrent studies are limited to identifying individual neurons or specific brain regions responsible\nfor number sense, but an understanding of how a network of number-selective neurons work\ntogether across the whole-brain remains elusive (Ditz & Nieder, 2015, 2016; Messina, et al.,\n2022a, 2022b; Pfeifer et al., 2018; Piazza et al., 2004; Viswanathan & Nieder, 2013). In\nprimates, Viswanathan & Nieder (2013) showed a visual sense of number mapped to the\nparietal and prefrontal cortices. Expanding on this, recent studies suggest visual number\nprocessing expands beyond these two regions and implicates the superior colliculus, a deep\nsubcortical area (Collins et al., 2017; Georgy et al., 2016). In birds, involvement of several pallial\nregions have been recently documented by early gene expression (Lorenzi et al., 2024). While\nthe use of electrodes can access individual neurons at multiple regions, it is difficult to\nunbiasedly capture all neurons especially without neuron damage after implantation (Eles et al.,\n2018; Ferguson et al., 2019; Goss-Varley et al., 2017). The ability to capture individual neuronal\nactivity at all regions would enable researchers to map the neural circuits implicated in ANS\nprocessing with unparalleled precision in encoding and representation.\nTo address this, we developed and optimized a 2-photon fluorescence light sheet microscopy\n(2P-LSFM) platform (Keomanee-Dizon et al., 2020; Messina, Potrich, Perrino, et al., 2022;\nTruong et al., 2011; Vito et al., 2020) and a customized data analysis pipeline. This allowed for\nthe noninvasive imaging of the functional activity of nearly all neurons across the whole brain in\nlarval zebrafish (Danio rerio) with single-neuron resolution. Larval zebrafish offer many\nadvantages as a model system for studying neural processes such as transparency, genetic\ntractability, and drug screening applications (Kawakami, 2005; Lubin et al., 2021; Trinh & Fraser,\n2013; Weber & Köster, 2013). By expressing a nuclear-localized calcium indicator\n(H2B-jGCaMP7f) (Dana et al., 2019a), we were able to monitor brain-wide neuronal activity in\nresponse to non-symbolic, visual numerical stimuli. While no studies have shown any ability to\ndiscriminate different numerosities at an early age (<7 days post-fertilization (dpf)) in zebrafish,\nit must develop before the behavior becomes apparent (Adam et al., 2024; Lorenzi et al., 2023;\nLucon-Xiccato et al., 2023; Sheardown et al., 2022a). Here, we aim to identify these\nnumber-neurons, across the brain, in real time as the animal processes a visual number\nstimulus, in an effort to improve our understanding of the neural basis of number sense.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nResult\nWe used our custom-developed 2P-LSFM to record the whole-brain neuronal activity of\nagarose-embedded zebrafish at age 3-, 5-, and 7-days post fertilization (dpf), while the animals\nwere presented with visual numerical stimuli based on dots (Figure 1a, Methods). Imaging was\nacquired at 1 Hz whole-brain volumetric rate, and the entire imaging experiment lasted for 90\nminutes, as the numerical stimuli sequenced from one to five dots. When the quantity of dots\nchanges, non-numerical geometric effects co-vary with the changes and confound the numerical\neffects. For example, two circles have a higher combined area than one circle of the same\ndiameter. To account for possible geometric effects, the number-based dot stimuli account for\nnumerical and non-numerical variables (i.e. for continuous physical variables that co-vary with\nnumerosity) (Zanon et al., 2022). The non-numerical variables were divided into spread and size\nof the individual dots. Angular diameter of the dots was kept to a minimum of 5° (angular\ndegree) above the visual acuity threshold of 2-3° (Haug et al., 2010). The spread of the dots\nincludes convex hull and distance between the dots, while the dot size includes constant radius,\ntotal dot area, and total dot perimeter (Figure 1b). The stimuli sequence includes all possible\ncombinations of spread and sizes using a new pattern for each stimulus (Supplementary Figure\n1). To account for any intrinsic neural oscillatory signal that may have a repetitive pattern in\nphase with the stimulus display, the inter-stimulus interval follows a pseudo-random stimulus\nsequence.\nWe applied and optimized several publicly available Python tools and software for managing,\nprocessing, and analyzing volumetric movie data and neuronal signals. We used VoDEx\n(Nadtochiy et al., 2023) to manage the 4D (volumetric movie) data and stimuli annotations.\nAdvanced Normalization Tools (Avants et al., 2009) was used to spatially align and correct for\nmotion artifacts in 4D datasets. Registration of multiple samples onto a representative brain\ntemplate was performed using ITK-SNAP (Yushkevich et al., 2006).\nTo segment for the signals from individual neurons, we applied the Python toolbox for\nlarge-scale Calcium Imaging Analysis (CaImAn) (Giovannucci et al., 2019). See Methods for full\ndetails. Figure 1c, d, e shows example images of the raw image data with whole-brain coverage\nat cellular resolution. The resulting processed segmentation result is represented in\nSupplementary Figure 2.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nFigure 1. Application of two-photon fluorescence light sheet microscopy to detect\nneuronal representation of number perception in larval zebrafish.\na Schematic of the two-photon light sheet fluorescence microscopy system. The sample is\nexcited by two orthogonal 920 nm scanning lasers to obtain uniform excitation of the labeled\nneurons. A physical eye-shaped mask blocks the laser illumination of the eye. Fluorescence is\ndetected by an objective above the sample. The visual-based number stimulus is displayed on\nthe right side of the larval fish. Bottom-left: brightfield image of the mounted sample, scale bar =\n200 μm. Center-bottom: stimulus projection image, scale bar = 10 mm. Bottom-right: Dorsal\nview of the sample relative to the stimulus direction and excitation laser. A = anterior, P =\nposterior, D = dorsal, V = ventral, L = left, R = right.\nb Examples of different continuous geometric parameters used to control for co-varying\nnon-numerical variables when changing quantities of objects. Convex hull and inter-distance\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\ncontrols spread of the dots, while radius, total area and perimeter controls for the dot size. See\nMethod: Stimuli Generation.\nc Example fluorescence maximum image projection of a 7 dpf zebrafish brain. Top, dorsal view,\nmaximum projection along the dorsal/ventral axis, white boxes area depicted in (d) and (e);\nmiddle, frontal view, maximum projection along the rostral/caudal axis; bottom, lateral view,\nmaximum projection along the left/right axis. Time average of 60 seconds. Scale bar = 100 μm.\nd,e Example magnified fluorescence image showing cellular resolution. (d) forebrain (e) tectum.\nLeft, dorsal view of a single plane; middle, frontal view of a single plane along the green dash\nlines; right, lateral view of a single plane along cyan dash line. Scale bar = 50 μm\nNeurons correlating to number stimuli are detected early in development\nTo identify neurons specifically responsive to changes in numerosity (Figure 2a, Supplementary\nFigure 3) from those responsive to geometric changes, we applied a two-way permutation\nANOVA. Neurons were filtered based on a significant main effect for changes in numerosity (p <\n0.01), without exhibiting significant main or interaction effects due to geometric changes. We\ndefine these filtered neurons as “number-selective neurons”. As an example, we present the\nsignificant main effect for changes in numerosity during the stimulus onset for a 7 dpf larva\n(Figure 2b). The analysis window of 3 seconds accounts for the typical ~2 second decay time\nconstant (Dana et al., 2019b; Yang et al., 2022). In general, a significant Ca2+ response to a\nnumerical stimulus was detected 0-3 seconds from the stimulus onset.\nOn average, we identified 1300±300, 800±100, 550±100 number-selective neurons, among\n14000±2700, 17000±1300, 17000±2000 detected active neurons in 3, 5, and 7 dpf larval\nzebrafish, respectively (n=5 for each age group) (Supplementary Table 1-5). Due to the varying\nexpression levels of H2B::GCaMP across individual fishes and varying signal-to-noise ratios\ndue to development of the skin (Li & Uitto, 2014) and head (Kimmel et al., 1995), quantifications\nof number-selective neurons are normalized by number preference, detected active neurons,\nand regions for each sample. Example Ca2+ signal trace and tuning curves of neurons tuned to\n1-5 objects for one fish are shown in Figure 2c. Neurons showing a peak Ca2+ response to a\nspecific numerosity is defined as tuned or having preference to that numerosity. Tuning curves\nshow a gradual decrease in Ca2+ response for numerosities further from the preferred\nnumerosity (Figure 2d; for population tuning see Supplementary Figure 4).\nAs the zebrafish larval age increases across 3, 5, and 7 dpf, we found that the proportion of\nneurons tuned to numerosities of two or more shows a trending increase with age (3dpf:4%;\n5dpf:12%; 7dpf:14%) (Figure 3a). When comparing the tuning preferences of the\nnumber-selective neurons (Figure 3b), we found a significant decrease of 1-tuned neurons in 3\ndpf (96 ± 1%) compared to 5 (88 ± 2%) and 7 (86 ± 5%) dpf groups (p < 0.05). For 3-tuned\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nneurons, we found a significant increase in 7 dpf (5 ± 1%) compared to the 3 dpf (0.2 ± 0.1%)\ngroup (p < 0.01).\nFigure 2. Number-selective neurons produce higher Ca 2+ activity for preferred\nnumerosities.\na Example raw fluorescence intensity trace of Ca2+ activity from a single neuron during\nnumerical stimuli. Neurons responsive to numerical stimuli show a Ca2+ spike after the stimulus\nonset.\nb Statistical significance of number selectivity over time during the visual numerical stimulus of\nnumber-selective neurons in one example 7 dpf larvae. Rows represent individual neurons (n =\n599). Stimulus starts at time = 0 and lasts for 1 second. Colormap indicates the p value of\nselectivity to the stimulus. Black solid line on top indicates a 3-second analysis window used to\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\ndetect number-selective neurons with a two-way permutation ANOVA to account for the rise and\ndecay time of the Ca2+ response.\nc Ca2+ activity trace of 5 example neurons with preference to 1-5 dots. Traces are centered\naround the stimulus onset (peristimulus) for preferred and non-preferred numerosities. Neuron\npreferences were selected by the highest average activity for a numerosity (for example traces\nto nonnumeric geometric effects see Supplementary Figure 3). Top labels indicate the number\nof dots presented, and each color represents the specific number tuning (1 = red, 2 = cyan, 3 =\ngreen, 4 = yellow, 5 = magenta). Baseline for ∆F/F is calculated by averaging the 3 time points\nprior to the stimulus onset (dotted vertical line). Black line indicates the average across 48 trials.\nd Tuning curves of each of the five neurons from c Each entry is the average of the 3-second\nanalysis window for each numerosity presentation (n = 3 * 48 numerosities). Error bar = SEM.\nFigure 3. Populations of neurons tuned to specific numerosities show redistribution of\nnumber preference during early development.\na Proportion of number-selective neurons as a percentage of all detected number-selective\nneurons across 3, 5, and 7 dpf. 1-tuned neurons = red, 2-tuned neurons = cyan, 3-tuned\nneurons = green, 4-tuned neurons = yellow, 5-tuned neurons = magenta. n = 5.\nb Percentages of number-selective neurons preferring specific numerosities for 3, 5, and 7 dpf.\nPercentage is expressed as a proportion to all detected number-selective neurons. Pairwise\ncomparisons were performed using a Mann-Whitney U-test for each numerosity, multiple\ncomparisons were adjusted using a Bonferroni correction (alpha = 0.016). Error bars = SEM, n =\n5, * = p < 0.05, ** = p < 0.01.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nNumber-selective neurons primarily localize to the forebrain and midbrain\nNumber-selective neurons were detected across the forebrain, midbrain, and hindbrain of 3, 5,\nand 7 dpf groups. To show the locations of the number-selective neurons, we registered all\nimaged samples onto the respective brain templates of each age group (see Methods), and\nmapped out the centroids of all identified neuronal nuclei (Figure 4a&b, Supplementary Figure\n5). We found a majority of number-selective neurons were localized to the forebrain and\nmidbrain (Figure 4c). For all three age groups, the forebrain (3dpf:28 ± 4%; 5dpf:45 ± 5%;\n7dpf:39 ± 6%) and midbrain (3dpf:65 ± 5%; 5dpf:49 ± 6%; 7dpf:45 ± 4%) had proportionally\nmore number-selective neurons than the hindbrain (3dpf:7 ± 2%; 5dpf:6 ± 2%; 7dpf:15 ± 5%) (p\n≤ 0.01). We did not identify any apparent mapping based on preferred numerosities\n(Supplementary Figure 6). In the 3 dpf group, we found significantly less neurons in the\nforebrain than the midbrain (p < 0.001).\nBuilding upon these findings, we further investigated the subregions of the forebrain (eminentia\nthalami, hypothalamus, pallium, pretectum, thalamus, subpallium) and found no significant\nchanges with age (Supplementary Figure 7, Supplementary Table 6).\nFigure 4. Number-selective neurons are primarily detected in the forebrain and midbrain.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\na The 3D map of the brain was divided into three major brain regions (forebrain, midbrain,\nhindbrain). Solid lines indicate delineation of major brain regions, dash lines indicate\noverlapping regions.\nb Locations of number-selective neurons in three different individual larval zebrafish at three\nstages of development, representing the results as point maps in orthographic projections. The\nwhite circles represent the centers of each identified number-selective neuron. Columns indicate\nage. Scale bar: 100 µm.\nc Comparison of number-selective neuron distribution across brain regions of three stages of\ndevelopment. Number-selective neurons per region is normalized by the total number of\nnumber-selective neurons detected in the whole brain. Comparisons were performed using a\ntwo-way ANOVA for age and brain region followed by Tukey’s HSD. Error bars = SEM, n = 5, * =\np < 0.05, ** = p < 0.01.\nNumber stimulus can be decoded from number-selective neurons\nTo determine if the Ca2+ activity of number-selective neurons across the brain is sufficient to\npredict the correct number of dots shown during a visual stimulus, we trained a support vector\nmachine (SVM) supervised classifier to estimate the visual stimulus from the recorded activities\n(Kirschhock & Nieder, 2022). The features were extracted averaging the Ca2+ activity of all\nidentified number-selective neurons for each preferred numerosity (Figure 5a, Methods). The\nCa2+ activity averages for number preferences 1-5 serve as the five input features and are used\nto predict six types of stimuli (1, 2, 3, 4, 5 dot, no dot). The classifier was trained on four out of\nfive individual zebrafish in each age group and tested on the remaining one, enabling\ngeneralized testing across conspecifics.\nTo assess the performance of the SVM classifier, we used a confusion matrix to evaluate the\nclassifier accuracy (Figure 5b, c, d, e). The confusion matrix shows the prediction instances\n(columns) during a visual numerical dot stimulus or “true label” (rows). Each entry of the matrix’s\nmain diagonal shows correct prediction of the true labels. By averaging the diagonal entries, the\noverall classifier accuracy can be obtained. At 3 dpf, the prediction accuracy was above chance\nlevel (16.7%) for the numerosities 1 (50%), 2 (30%), 3 (30%), and 5 (32%) (Figure 5b). At 5 dpf,\nthe general is similar but with an increase in accuracy: 1 (61%), 2 (43%), 3 (44%), and 5 (41%)\n(Figure 5c). At 7 dpf, for numerosities greater than 2 we see a general increase in accuracy: 1\n(56%), 2 (44%), 3 (52%), 4 (37%), and 5 (52%) (Figure 5c). The prediction improvement is\nconfirmed by the overall classifier accuracy increasing with age (42%, 48%, 55% for 3, 5, and 7\ndpf, respectively) (Figure 5f, Supplementary Table 7).\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nFigure 5. Prediction accuracy of the numerical stimuli from Ca 2+ activity using SVM\nclassifier show increased performance with age.\na Feature extraction of Ca2+ activity from numerically-tuned neurons. For each training and\ntesting instance, the average Ca2+ activity of each neuron population tuned to the 5 numerosities\nwas calculated and shown as a percentage of ∆F/F (5 input features). During each instance, a\nsingle visual stimulus (1, 2, 3, 4, 5, background/no dot) serves as the true label. Each sample\nlarval fish is comprised of 288 instances (48 repetition * 6 visual stimulus type). Cross validation\nwas performed using a leave-one-out cross validation where training was performed on the data\nfrom 4 of 5 larval fish and tested on the excluded sample then repeated on a different excluded\nsample.\nb, c, d, e Confusion matrix of the support vector machine (SVM) classifier of 3, 5, and 7 dpf\ngroups and a shuffled 7 dpf group. The percentage indicates the number of predictions out of\nthe total instances of each true label. Random chance = 16.7%.\nf Overall SVM classification accuracies for the three age groups. Dash line indicates chance\nlevel, calculated as the accuracy of shuffled data from the 7 dpf group.\nEthanol inhibits number-selective neurons in the forebrain\nAcute ethanol exposure affects learning and memory processing in zebrafish (Sartori et al.,\n2022). To understand how this affects number-selective neurons we examined the effect of\nethanol on number-selective neurons of 7 dpf larvae compared to the untreated larvae.\nLocation of detected number-selective neurons in the forebrain is noticeably less when treated\nwith 1.5% ethanol (Figure 6a). The percentage of number-selective neurons in the\nethanol-treated group (10%) showed a significant decrease in the forebrain when compared to\ndetected active neurons in the forebrain brain (20%) and the forebrain of untreated group (39%)\n(n=5 for each treatment group) (Figure 6b, Supplementary Figure 8). To measure the predictive\ncapacity of number-selective neurons of an ethanol-treated group, we trained a supervised\nclassifier to predict the visual stimulus using the Ca2+ activity (Figure 6c). The overall accuracy\nof the ethanol-treated group (41%) was decreased compared to the untreated group (55%)\n(Figure 6d). This effect is mainly driven by the decreased accuracy when predicting\nnumerosities of 3, 4, and 5. These results suggest ethanol may impair the function of\nnumber-selective neurons.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nFigure 6. Ethanol alters the activity of number-selective neurons in the forebrain.\na Location of number-selective neurons in three different brain regions in a single 7 dpf larval\nzebrafish treated with ethanol. Refer to the caption for Figure 4b for a detailed description.\nb Percentage of number-selective neurons in the forebrain during ethanol treatment.\nNumber-selective and active neurons in the forebrain are normalized to all number-selective or\nactive neurons (respectively) across the whole brain. Pairwise comparisons were performed\nusing a Mann-Whitney U-test with a Bonferroni correction for multiple comparisons (alpha =\n0.17). Error bars represents SEM, n = 5, ** denotes p < 0.01.\nc Confusion matrix of the SVM classifier of the numerical stimuli using Ca2+ activity during\nethanol treatment. Refer to the caption for Figure 5b, c, d, e for a detailed description.\nd Overall SVM classification accuracies for the 7 dpf EtOH treatment groups. Dash line\nindicates chance level, calculated as the accuracy of shuffled data from the 7 dpf untreated\ngroup.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nDiscussion\nIn this study, we investigated the neural basis of number sense in larval zebrafish, focusing on\nthe tuning of neurons to specific visual-based numerosities under different conditions. Our work\nhas, for the first time, discovered the existence of number-selective neurons in larval zebrafish.\nThe fast volumetric imaging rate of one volume per second and the high signal-to-noise ratio of\nlight sheet microscopy enabled the whole-brain recording and segmentation of approximately\n17,500 active neurons per larva. The use of two-photon excitation allowed for increased depth\nof coverage (deeper imaging) and reduced visual artifacts compared to one-photon excitation\n(Truong et al., 2011; Vito et al., 2022; Wolf et al., 2015).\nConsistent with studies using chick models (Kobylkov et al., 2022) and human infants (Izard et\nal., 2009; Xu & Spelke, 2000), we detected number-selective neurons during early\npost-embryonic (equivalent to post-natal) stages in zebrafish. Notably, the existence of these\nnumber-selective neurons at 3 dpf precedes any known numerically-driven behaviors such as\nhunting and shoaling (which start at 5 and ~24 dpf, respectively) (Adam et al., 2024; Borla et al.,\n2002; Lucon-Xiccato et al., 2023; Sheardown et al., 2022a), underscoring the fundamental role\nand necessity of early numerical cognition for survival.\nConsistent with studies using chick models (Kobylkov et al., 2022) and human infants (Izard et\nal., 2009; Xu & Spelke, 2000), we detected number-selective neurons during early\npost-embryonic (equivalent to post-natal) stages in zebrafish. Notably, we observed these\nneurons at 3 dpf, which is before the onset of known numerically-driven behaviors that typically\nbegin at 7 dpf (Adam et al., 2024; Lucon-Xiccato et al., 2023). This finding suggests that the\ndevelopment of these neurons precedes and potentially facilitates these behaviors, highlighting\nthe critical importance of early numerical cognition for survival.\nThe proportion of neurons tuned to numerosities of two or more shows a trending increase with\nage (Figure 3a). A significantly increased proportion of neurons preferring 3 objects was\ndetected after 3 dpf (Figure 3b). These results suggest number-selective neurons develop in an\nordinal fashion with age. An interesting question for future studies is whether this increase is\ndue to the generation of new neurons preferring higher numerosities or the re-tuning of existing\nneurons. This could be resolved by application and further refinement of our experimental\nplatform to observe number-selective neurons in the same zebrafish longitudinally over\ndevelopment time.\nThe increased proportion of neurons preferring larger numerosities (>2) developing after 3 dpf\nmay be caused by an improvement in visual acuity rather than changes to number-selective\nneurons. However, the zebrafish eye is emmetropic at 3 dpf (Easter & Nicola, 1996), and no\ndifferences in visual acuity were detected when comparing larvae at 4, 5, and 6 dpf (Haug et al.,\n2010). Furthermore, recognition of 5 dots does not require finer visual acuity than 2 dots when\nmaintaining equivalent inter-distances (Figure 1c). Because we detected neurons preferring 2\ndots in 3 dpf larvae, the increase of neurons preferring larger numerosities in older larvae is\nunlikely to be caused by improved visual acuity.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nWe identified number-selective neurons localized throughout the forebrain and midbrain (Figure\n4b). In the 3 dpf group, we found significantly less neurons in the forebrain compared to the\nmidbrain, whereas in the 5 and 7 dpf groups, the forebrain contained a similar proportion of\nthese neurons (Figure 4c), likely due to the forebrain being more developed in the older fish\n(Cheng et al., 2014). Further analysis of the forebrain subregions found no significant\nage-related changes in (Supplementary Figure 7, Supplementary Table 6). These results\nsuggest that while overall development affects neuron distribution, the specific sub regional\nchanges may not happen until the brain matures beyond 7 dpf.\nIn the mammalian brain, most number processing is to our knowledge localized to the prefrontal\nand parietal cortices (Nieder et al., 2002; Piazza et al., 2004). In non-mammals, such as\nzebrafish, the pallium generally fulfills the role of the prefrontal cortex (Medina et al., 2019).\nHowever, the non-mammalian brain lacks a structure that is directly analogous to the parietal\nlobe; instead, the optic tectum of the midbrain serves many cortical functions such as sensory\nprocessing and spatial perception (Förster et al., 2020; Gazzola et al., 2018; Muto et al., 2013).\nIn line with our result of finding number-selective neurons in the midbrain, emerging studies\nsuggest subcortical and optic tectal involvement in visuospatial processing on numerosities\n(Bengochea et al., 2023; Bengochea & Hassan, 2023; Collins et al., 2017; Lorenzi et al., 2021).\nThis suggests that the optic tectum participates in more complex functions than its long-studied\nroles in visual mapping and sensory integration.\nTo assess the predictive capabilities of the number-selective neurons across 3, 5, and 7 dpf\nzebrafish, we trained a supervised classifier using their underlying Ca2+ activity. The prediction\naccuracy of the classifier increases with age for 2-5 objects, indicating that higher\nnumber-selective neurons become more specific (generating more action potentials in response\nto specific numerical stimuli) as the larval zebrafish matures. Similar to the findings of number\nsense in crows (Kirschhock & Nieder, 2022), our results show that most misclassifications\noccurred near the correct choice, suggesting a numerical distance effect (i.e. discrimination\nerrors arise between closer numerosities) (Moyer & Landauer, 1967; Nieder, 2011). From 3 to 5\ndpf, the average prediction accuracy of 2, 3, and 5 objects increased from ~30% to ~43%, but\ninterestingly the accuracy of 4 objects remained near random chance (17%) until 7 dpf.\nA plausible explanation for the decreased accuracy of 4 objects for 5 dpf fish is that numerosity\nof 4 represents a transition point between Object Tracking System/Parallel Individuation System\n(OTS/PIS) and Approximate Number System (ANS) (Feigenson et al., 2004; Hyde, 2011;\nSheardown et al., 2022b). The OTS/PIS is thought to be responsible for tracking and\nrepresenting small quantities of objects with high precision, typically up to four items. Whereas\nthe ANS operates on an approximate level, allowing for rapid estimation of small (<4) and large\n(>4) numerical magnitudes beyond the capacity of the OTS/PIS. If the OTS develops ordinally\nthen neurons preferring 4 objects would develop last whereas neurons preferring 5 objects\nwould have emerged earlier with the ANS.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nWhen larval zebrafish were exposed to ethanol, the activity of number-selective neurons\ndecreased in the forebrain (Figure 6a&b). Given ethanol's well-known propensity to inhibit\ninformation processing in the frontal lobe of humans (Koelega, 1995; Tzambazis & Stough,\n2000), it is likely that the number-selective neurons in the forebrain, which are implicated in\ncomplex higher-order functions such as learning and memory (Dempsey et al., 2022; Rodríguez\net al., 2002), are selectively affected. When predicting the stimulus from the Ca2+ activity using\nan SVM classifier, the overall accuracy decreases for larger numerosities (>2) (Figure 6c).\nInterestingly, the predicting accuracy of 1 dot showed improvement. One possible explanation is\nethanol treatment is inhibiting the number-selective neurons part of the ANS system that prefer\n1, leaving OTS neurons remaining which have a more precise representation of numbers\n(Feigenson et al., 2004).\nTo date, this study is the first to comprehensively identify individual number-selective neurons\nacross the whole brain through the use of Ca2+ imaging and 2P-LSFM. The findings here are a\nsignificant step towards unraveling neural circuits devoted to the understanding of continuous\nand discrete magnitudes. Additionally, our methodology to register neurons across the whole\nbrain and follow their activity could be broadly applicable to other fields in neuroscience.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nMethods\nKey resources table\nReagent type\n(species) or resource\nDesignation Source or reference Identifiers\nGenetic reagent\n(Danio rerio)\nZebrafish:\nTg(elavl3:H2B::jGCa\nMP7f)\n(Dana et al., 2019b;\nYang et al., 2022),\ngift from David\nProber\nRRID:Addgene_1044\n88\nPython Library Analysis tools:\nANTs\n(Avants et al., 2009) https://github.com/AN\nTsX/ANTs\nPython library Analysis tools:\nNuMan\nhttps://github.com/Le\nmonJust/numan\nPython library Analysis tools:\nnuman_plus\nhttps://github.com/Mir\nkoZanon/numan_plus\nPython library Analysis tool:\nseaborn\n(Waskom, 2021) https://seaborn.pydat\na.org/index.html\nPython library Data management:\nVoDEx\n(Nadtochiy et al.,\n2023)\nhttps://github.com/Le\nmonJust/vodex\nPython library Cell Segmentation (Giovannucci et al.,\n2019)\nhttps://github.com/flat\nironinstitute/CaImAn\nPython library Stimuli presentation:\nPsychoPy\n(Peirce et al., 2019) https://psychopy.org/i\nndex.html\nSoftware/Python\nLibrary\nImage analysis\ntoolkit: ITK-SNAP\n(Yushkevich et al.,\n2006)\nhttp://www.itksnap.or\ng/\nSoftware Microscope GUI,\nµManager\n(Edelstein et al.,\n2010a)\nhttps://doi.org/10.100\n2/0471142727.mb142\n0s92\nSoftware Stimuli generation:\nGeNEsIS\n(Zanon et al., 2022) https://github.com/Mir\nkoZanon/GeNEsIS\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nAnimal care\nCasper zebrafish (Danio rerio) expressing a pan-neural, nuclear-localized fluorescence Ca2+\nreporter (elavl3:H2B::jGCaMP7f) was a gift from the lab of David Prober at California Institute of\nTechnology. Larval fish were raised accordingly to establish methods (Avdesh et al., 2012) with\nmodifications: 13:11 hr (light:dark) and fed dry food twice daily after 5 dpf. Experiments used\nzebrafish ranging from 3-7 days-post-fertilization (dpf). Sex is not defined at this stage of\ndevelopment. Larvae were raised in 50 mL petri dishes with approximately 50 larvae per dish.\nE3 medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2, 0.33 mM MgSO4). All animal\nprocedures conformed to the institutional guidelines set by the University of Southern California\nDepartment of Animal Research.\nCalcium imaging\nZebrafish larvae were embedded in 2% low-melting-point agarose (Invitrogen cat 16520100)\nand mounted in a custom sample holder. During image acquisition, the larvae were perfused\nwith oxygenated water using a peristaltic pump and heated to 28C. Image acquisition was\nperformed on a custom-built microscope (Keomanee-Dizon et al., 2020) that was further\nmodified by correcting the polarization and additional laser pulsing to increase the fluorescence\nsignal (Luu et al., 2024; Vito et al., 2020, 2022). The sample was stimulated using the\nChameleon Ultra II Ti:Sapphire laser (Coherent) at 920 nm with approximately 300 mW peak\npower and 180 mW average power (combined excitation laser at the sample after splitting).\nEmitted light was bandpass filtered at 525/45 nm and collected using 20x 1.0 NA water dipping\nobjective (Olympus). Images were acquired at a rate of 1 second per volume, in which a volume\nis composed of 60 z-slices at 500 x 510 pixels across 230 µm (900 x 500 x 230 µm3, equating\nto 3.83 µm section thickness). The total acquisition time per sample is approximately 90 minutes\nexcluding a 30-minute acclimation period. Software controls for image acquisition was\nperformed using µManager (Edelstein et al., 2010b)\nStimuli Generation\nDot patterns were generated using GeNEsIS (Zanon et al., 2022) and controlled for convex hull,\ninter-distance, total area, total perimeter, and radius in (Figure 1b) . Convex hull describes the\nsmallest convex polygon that encloses all of the elements, inter-distance is the average\ndistance between the elements. Total area equates average brightness and cumulative surface\narea for different numerosities. Total perimeter equates the cumulative circumference of all the\nelements for different numerosities. The parameters are summarized in Supplementary Table 8.\nNote: parameter values apply by use case (“1” dot stimuli does not have convex hull or\ninter-distance parameters, constant radius does not use radius variability). Angular diameter of\ndots were kept above 5° to maintain visibility (Haug et al., 2010) and below 18° to minimize an\nescape response (Temizer et al., 2015). Numerical elements were colored black on a red\nbackground to simulate objects' contrast in the natural environment and to prevent disassembly\nof the photoreceptor of the photoreceptor (Emran & Dowling, 2010).\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nVisual number-based display\nVisual stimuli were projected onto a diffuser placed 19mm away from the larvae. The diffuser is\nmade of cellulose acetate (Scotch Magic tape) that faces the right eye of the larval fish and is\nplaced orthogonally to its body axis. Illumination was generated using Qumi Q5 LED Projector\n(Vivitek) bandpass filtered at 660/45 nm (Thorlabs).\nEach numerical stimulus is presented 48 times following a pseudo-random order. The stimulus\nis displayed for 1 second followed by varying inter-stimulus intervals of a blank red background\nbetween 15-30 seconds. The display area is 22mm in diameter or 66° in angular diameter.\nStimulus control was performed using the PsychoPy toolkit (Peirce et al., 2019).\nEthanol administration\nTo determine the appropriate ethanol concentration, we adapted methods from previous studies\n(Dlugos & Rabin, 2003; Vossen et al., 2022) to assess larval zebrafish swimming behavior and\nmortality after treatment. In triplicates, five 7-day post-fertilization (dpf) zebrafish were immersed\nin 15 mm x 100 mm Petri dishes containing 1, 1.5, and 2% ethanol in E3 media for 1.5 hours\n(image acquisition duration). We then screened for hyperactivity by gently tapping on the Petri\ndish and chose the 1.5% ethanol concentration for this study. Before image acquisition, the\nzebrafish were treated with 1.5% ethanol 30 minutes prior and then continuously perfused with\noxygenated E3 media containing 1.5% ethanol during imaging.\nCell segmentation\nThe datasets were first motion corrected using Advanced Normalization Tools (Avants et al.,\n2009). Cell segmentation was performed using a python library designed for calcium imaging\n(CaImAn) (Giovannucci et al., 2019). CaImAn consists of a series of functions enabling the\nseparation of neurons using Ca2+ activity in time and space by applying non-negative matrix\nfactorization. Prior to cell segmentation, the data size was reduced by selecting only the time\npoints around the stimulus presentation (3 s before stimulus + 1 s stimulus + 5 s post-stimulus;\nsee Figure 2). The final volumetric time series was reduced from 5,472s to 2160s (9 seconds\ncrop X 5 numerosities X 48 repetitions). Large data handling and annotation were managed\nusing an inhouse python library that facilitated image processing (Nadtochiy et al., 2023).\nThe segmentation was performed in 2D where each time point consisted of 60 z-slices with the\nfollowing parameters: ‘decay_time’ = 5 (length of a typical transient in seconds); ‘gSig’ = 3x3\n(expected half size of neurons in pixels); min_SNR = 1.5 (signal to noise ratio to accept a\ncomponent); rval_thr = 0.85 (space correlation threshold to accept a component). The ‘K’\nparameter is the expected number of cells to be segmented that serves as a starting point for\noptimization. Since the number of cells expected in each z-slice varies, the 'K' parameter is\nestimated based on the standard deviation of Ca2+ flux in each z-slice over time. The standard\ndeviation image is thresholded (min_std + 0.08*(max_std - min_std)) resulting in an image with\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\npixels that represent cells with Ca2+ flux. The max number of the resulting pixels are then\ndivided by ‘gSig’ to approximate the number of cells per z-slice. Since a single cell’s Ca2+ signal\ncan span across 2-3 z-slices, we eliminate duplicates by merging cells with both a centroid\ndistance less than 1 pixel and with a Ca2+ activity correlation coefficient higher than 0.95.\nNumber neuron selection\nIdentification of numerically-tuned neurons involved additional preprocessing steps that\nremoved camera shot noise and established a baseline fluorescence. To remove false positive\nsegmented cells caused by the camera shot noise (identified as segmented cells that were\nfound outside of the brain), we calculated the coefficient of variation (CV), the ratio of the\nstandard deviation to the mean, for each timepoint in the peristimulus windows. We found a CV\nof 0.05 was sufficient to remove false positive cells related to shot noise. Baseline fluorescence\n(F0) was defined as the average of three time points before the visual stimulus around the\nstimulus presentation (peristimulus window) for each numerosity.\nTo differentiate neurons responsive to numerosity from those responsive to nonnumerical\ncovariables (size and spread), we utilized a two-way permutation ANOVA. This involved\nrandomizing the labels associated with the data and calculating the F-value. We repeated this\nprocess 10,000 times to construct a null distribution based on simulated F-values to which the\nactual F-value was compared to get the p-value. The criteria for identifying a number-selective\nneuron must have a significant main effect for the numerical stimulus (alpha = 0.01), a\nnon-significant main effect for the nonnumerical covariables, and no interaction effect.\nBrain spatial registration and region segmentation\nAll samples were first registered to a brain template of each respective age group with ITKsnap\n(Yushkevich et al., 2006) using the average Ca2+ signal in time. All identified neuron centers\nwere remapped to the final brain template to compare across different samples. To identify\nsubregions of the forebrain, we registered the brain templates to the mapZebrain atlas (Kunst et\nal., 2019) using affine transformation, then selected the available subregion Boolean masks.\nSupervised classification\nTo test the predictive properties of the number-selective neurons, we applied a support vector\nmachine based supervised classifier using a linear kernel. The classifier used the underlying\nCa2+ activity to predict the visual number-based stimulus. To extract the features, we calculated\nthe average Ca2+ activity of the neurons tuned to each of the five numerosities during a\n2-second window encompassing the 1-second visual stimulus and the following post-stimulus\nsecond. These five average activities served as input features for the SVM. The six classes\n(true labels) consisted of the five numerosities (1-5 objects from the visual stimulus) and the\naverage Ca2+ activity during the frames preceding the stimulus representing the no stimulus or\nbackground baseline.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nWe trained the SVM model on each experimental group (3 dpf, 5 dpf, 7 dpf, 7 dpf + etoh) to\nclassify trials based on the five extracted features. We applied a leave-one-out cross-validation\nscheme, where the model was trained on data from four fish within a group and tested on the\nremaining fish. This procedure was repeated five times, each time excluding a different fish. The\nfinal confusion matrices (figure 4b,c,d & 5c) were obtained by combining the test results from all\nfive repetitions.\nStatistical analysis\nStatistical analyses and graph preparation were conducted using:\ncustom python scripts, seaborn library (Waskom, 2021), Inkscape.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nFunding\nHuman Frontier Science Program\nGrant RGP0008/2017 to CB, SEF, GV\nERC European Union's Horizon 2020 research and innovation program\nGrant agreement 833504 – SPANUMBRA to GV, CB\nFARE–Ricerca in Italia: Framework per l'Attrazione ed il Rafforzamento delle Eccellenze per la\nricerca in Italia, III edizione, project \"NUMBRISH–The neurobiology of numerical cognition:\nsearching for a molecular genetic signature in the zebrafish brain\" Prot. R20YL9WN9N to GV.\nNational Institutes of Health 1U01NS122082-01, 1R34NS126800-01 to TVT.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nReferences\nAdam, E., Zanon, M., Messina, A., & Vallortigara, G. (2024). Looks like home: Numerosity, but\nnot spatial frequency guides preference in zebrafish larvae (Danio rerio). Animal\nCognition, 27(1), 53. https://doi.org/10.1007/s10071-024-01888-0\nAlder, T. B., & Rose, G. J. (2000). Integration and recovery processes contribute to the temporal\nselectivity of neurons in the midbrain of the northern leopard frog, Rana pipiens. Journal\nof Comparative Physiology. A, Sensory, Neural, and Behavioral Physiology, 186(10),\n923–937. https://doi.org/10.1007/s003590000144\nAvants, B. B., Tustison, N., Song, G., & others. (2009). Advanced normalization tools (ANTS).\nInsight j, 2(365), 1–35.\nAvdesh, A., Chen, M., Martin-Iverson, M. T., Mondal, A., Ong, D., Rainey-Smith, S., Taddei, K.,\nLardelli, M., Groth, D. M., Verdile, G., & Martins, R. N. (2012). Regular Care and\nMaintenance of a Zebrafish (Danio rerio) Laboratory: An Introduction. Journal of\nVisualized Experiments : JoVE, 69, 4196. https://doi.org/10.3791/4196\nBengochea, M., & Hassan, B. (2023). Numerosity as a visual property: Evidence from two highly\nevolutionary distant species. Frontiers in Physiology, 14.\nhttps://doi.org/10.3389/fphys.2023.1086213\nBengochea, M., Sitt, J. D., Izard, V., Preat, T., Cohen, L., & Hassan, B. A. (2023). Numerical\ndiscrimination in Drosophila melanogaster. Cell Reports, 42(7), 112772.\nhttps://doi.org/10.1016/j.celrep.2023.112772\nBonny, J. W., & Lourenco, S. F. (2013). The Approximate Number System and its Relation to\nEarly Math Achievement: Evidence from the Preschool Years. Journal of Experimental\nChild Psychology, 114(3), 375–388. https://doi.org/10.1016/j.jecp.2012.09.015\nBorla, M. A., Palecek, B., Budick, S., & O’Malley, D. M. (2002). Prey Capture by Larval\nZebrafish: Evidence for Fine Axial Motor Control. Brain Behavior and Evolution, 60(4),\n207–229. https://doi.org/10.1159/000066699\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nBrannon, E. M., & Merritt, D. J. (2011). Evolutionary foundations of the approximate number\nsystem. In Space, time and number in the brain: Searching for the foundations of\nmathematical thought (pp. 207–224). Elsevier Academic Press.\nhttps://doi.org/10.1016/B978-0-12-385948-8.00014-1\nCheng, R.-K., Jesuthasan, S. J., & Penney, T. B. (2014). Zebrafish forebrain and temporal\nconditioning. Philosophical Transactions of the Royal Society B: Biological Sciences,\n369(1637), 20120462. https://doi.org/10.1098/rstb.2012.0462\nCollins, E., Park, J., & Behrmann, M. (2017). Numerosity representation is encoded in human\nsubcortex. Proceedings of the National Academy of Sciences of the United States of\nAmerica, 114(14), E2806–E2815. https://doi.org/10.1073/pnas.1613982114\nCross, F. R., & Jackson, R. R. (2017). Representation of different exact numbers of prey by a\nspider-eating predator. Interface Focus, 7(3), 20160035.\nhttps://doi.org/10.1098/rsfs.2016.0035\nDana, H., Sun, Y., Mohar, B., Hulse, B. K., Kerlin, A. M., Hasseman, J. P., Tsegaye, G., Tsang,\nA., Wong, A., Patel, R., Macklin, J. J., Chen, Y., Konnerth, A., Jayaraman, V., Looger, L.\nL., Schreiter, E. R., Svoboda, K., & Kim, D. S. (2019a). High-performance calcium\nsensors for imaging activity in neuronal populations and microcompartments. Nature\nMethods, 16(7), 649–657. https://doi.org/10.1038/s41592-019-0435-6\nDana, H., Sun, Y., Mohar, B., Hulse, B. K., Kerlin, A. M., Hasseman, J. P., Tsegaye, G., Tsang,\nA., Wong, A., Patel, R., Macklin, J. J., Chen, Y., Konnerth, A., Jayaraman, V., Looger, L.\nL., Schreiter, E. R., Svoboda, K., & Kim, D. S. (2019b). High-performance calcium\nsensors for imaging activity in neuronal populations and microcompartments. Nature\nMethods, 16(7), 649–657. https://doi.org/10.1038/s41592-019-0435-6\nDempsey, W. P., Du, Z., Nadtochiy, A., Smith, C. D., Czajkowski, K., Andreev, A., Robson, D. N.,\nLi, J. M., Applebaum, S., Truong, T. V., Kesselman, C., Fraser, S. E., & Arnold, D. B.\n(2022). Regional synapse gain and loss accompany memory formation in larval\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nzebrafish. Proceedings of the National Academy of Sciences, 119(3), e2107661119.\nhttps://doi.org/10.1073/pnas.2107661119\nDitz, H. M., & Nieder, A. (2015). Neurons selective to the number of visual items in the corvid\nsongbird endbrain. Proceedings of the National Academy of Sciences of the United\nStates of America, 112(25), 7827–7832. https://doi.org/10.1073/pnas.1504245112\nDitz, H. M., & Nieder, A. (2016). Sensory and Working Memory Representations of Small and\nLarge Numerosities in the Crow Endbrain. The Journal of Neuroscience: The Official\nJournal of the Society for Neuroscience, 36(47), 12044–12052.\nhttps://doi.org/10.1523/JNEUROSCI.1521-16.2016\nDlugos, C. A., & Rabin, R. A. (2003). Ethanol effects on three strains of zebrafish: Model system\nfor genetic investigations. Pharmacology Biochemistry and Behavior, 74(2), 471–480.\nhttps://doi.org/10.1016/S0091-3057(02)01026-2\nEaster, S. S., & Nicola, G. N. (1996). The development of vision in the zebrafish (Danio rerio).\nDevelopmental Biology, 180(2), 646–663. https://doi.org/10.1006/dbio.1996.0335\nEdelstein, A., Amodaj, N., Hoover, K., Vale, R., & Stuurman, N. (2010a). Computer Control of\nMicroscopes Using µManager. Current Protocols in Molecular Biology, 92(1),\n14.20.1-14.20.17. https://doi.org/10.1002/0471142727.mb1420s92\nEdelstein, A., Amodaj, N., Hoover, K., Vale, R., & Stuurman, N. (2010b). Computer Control of\nMicroscopes Using µManager. Current Protocols in Molecular Biology, 92(1),\n14.20.1-14.20.17. https://doi.org/10.1002/0471142727.mb1420s92\nEdwards, C. J., Alder, T. B., & Rose, G. J. (2002). Auditory midbrain neurons that count.Nature\nNeuroscience, 5(10), 934–936. https://doi.org/10.1038/nn916\nEles, J. R., Vazquez, A. L., Kozai, T. D. Y., & Cui, X. T. (2018). In vivo imaging of neuronal\ncalcium during electrode implantation: Spatial and temporal mapping of damage and\nrecovery. Biomaterials, 174, 79–94. https://doi.org/10.1016/j.biomaterials.2018.04.043\nEmran, F., & Dowling, J. E. (2010). Larval zebrafish turn off their photoreceptors at night.\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nCommunicative & Integrative Biology, 3(5), 430–432.\nhttps://doi.org/10.4161/cib.3.5.12158\nFeigenson, L., Dehaene, S., & Spelke, E. (2004). Core systems of number. Trends in Cognitive\nSciences, 8(7), 307–314. https://doi.org/10.1016/j.tics.2004.05.002\nFerguson, M., Sharma, D., Ross, D., & Zhao, F. (2019). A Critical Review of Microelectrode\nArrays and Strategies for Improving the Neural Interface. Advanced Healthcare\nMaterials, 8(19), e1900558. https://doi.org/10.1002/adhm.201900558\nFörster, D., Helmbrecht, T. O., Mearns, D. S., Jordan, L., Mokayes, N., & Baier, H. (2020).\nRetinotectal circuitry of larval zebrafish is adapted to detection and pursuit of prey. eLife,\n9, e58596. https://doi.org/10.7554/eLife.58596\nGazzola, A., Vallortigara, G., & Pellitteri-Rosa, D. (2018). Continuous and discrete quantity\ndiscrimination in tortoises. Biology Letters, 14(12), 20180649.\nhttps://doi.org/10.1098/rsbl.2018.0649\nGeorgy, L., Celeghin, A., Marzi, C. A., Tamietto, M., & Ptito, A. (2016). The superior colliculus is\nsensitive to gestalt-like stimulus configuration in hemispherectomy patients. Cortex; a\nJournal Devoted to the Study of the Nervous System and Behavior, 81, 151–161.\nhttps://doi.org/10.1016/j.cortex.2016.04.018\nGiovannucci, A., Friedrich, J., Gunn, P., Kalfon, J., Brown, B. L., Koay, S. A., Taxidis, J., Najafi,\nF., Gauthier, J. L., Zhou, P., Khakh, B. S., Tank, D. W., Chklovskii, D. B., &\nPnevmatikakis, E. A. (2019). CaImAn an open source tool for scalable calcium imaging\ndata analysis. eLife, 8, e38173. https://doi.org/10.7554/eLife.38173\nGoss-Varley, M., Dona, K. R., McMahon, J. A., Shoffstall, A. J., Ereifej, E. S., Lindner, S. C., &\nCapadona, J. R. (2017). Microelectrode implantation in motor cortex causes fine motor\ndeficit: Implications on potential considerations to Brain Computer Interfacing and\nHuman Augmentation. Scientific Reports, 7(1), 15254.\nhttps://doi.org/10.1038/s41598-017-15623-y\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nHaug, M. F., Biehlmaier, O., Mueller, K. P., & Neuhauss, S. C. (2010). Visual acuity in larval\nzebrafish: Behavior and histology. Frontiers in Zoology, 7(1), 8.\nhttps://doi.org/10.1186/1742-9994-7-8\nHyde, D. (2011). Two Systems of Non-Symbolic Numerical Cognition. Frontiers in Human\nNeuroscience, 5. https://www.frontiersin.org/articles/10.3389/fnhum.2011.00150\nIzard, V., Sann, C., Spelke, E. S., & Streri, A. (2009). Newborn infants perceive abstract\nnumbers. Proceedings of the National Academy of Sciences of the United States of\nAmerica, 106(25), 10382–10385. https://doi.org/10.1073/pnas.0812142106\nKawakami, K. (2005). Transposon tools and methods in zebrafish. Developmental Dynamics:\nAn Official Publication of the American Association of Anatomists, 234(2), 244–254.\nhttps://doi.org/10.1002/dvdy.20516\nKeomanee-Dizon, K., Fraser, S. E., & Truong, T. V. (2020). A versatile, multi-laser\ntwin-microscope system for light-sheet imaging. Review of Scientific Instruments, 91(5),\n053703. https://doi.org/10.1063/1.5144487\nKimmel, C. B., Ballard, W. W., Kimmel, S. R., Ullmann, B., & Schilling, T. F. (1995). Stages of\nembryonic development of the zebrafish. Developmental Dynamics: An Official\nPublication of the American Association of Anatomists, 203(3), 253–310.\nhttps://doi.org/10.1002/aja.1002030302\nKirschhock, M. E., & Nieder, A. (2022). Number selective sensorimotor neurons in the crow\ntranslate perceived numerosity into number of actions. Nature Communications, 13(1),\n6913. https://doi.org/10.1038/s41467-022-34457-5\nKobylkov, D., Mayer, U., Zanon, M., & Vallortigara, G. (2022). Number neurons in the\nnidopallium of young domestic chicks. Proceedings of the National Academy of Sciences\nof the United States of America, 119(32), e2201039119.\nhttps://doi.org/10.1073/pnas.2201039119\nKoelega, H. S. (1995). Alcohol and vigilance performance: A review. Psychopharmacology,\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\n118(3), 233–249. https://doi.org/10.1007/BF02245951\nKunst, M., Laurell, E., Mokayes, N., Kramer, A., Kubo, F., Fernandes, A. M., Förster, D.,\nMaschio, M. D., & Baier, H. (2019). A Cellular-Resolution Atlas of the Larval Zebrafish\nBrain. Neuron, 103(1), 21-38.e5. https://doi.org/10.1016/j.neuron.2019.04.034\nLi, Q., & Uitto, J. (2014). Zebrafish as a Model System to Study Skin Biology and Pathology.\nJournal of Investigative Dermatology, 134(6), 1–6. https://doi.org/10.1038/jid.2014.182\nLorenzi, E., Perrino, M., Messina, A., Zanon, M., & Vallortigara, G. (2023). A Kaspar Hauser\nexperiment for innateness of numerical cognition (p. 2023.12.06.570352). bioRxiv.\nhttps://doi.org/10.1101/2023.12.06.570352\nLorenzi, E., Perrino, M., Messina, A., Zanon, M., & Vallortigara, G. (2024). Innate responses to\nnumerousness reveal neural activation in different brain regions in newly-hatched\nvisually naïve chicks. Heliyon, 10(14). https://doi.org/10.1016/j.heliyon.2024.e34162\nLorenzi, E., Perrino, M., & Vallortigara, G. (2021). Numerosities and Other Magnitudes in the\nBrains: A Comparative View. Frontiers in Psychology, 12.\nhttps://www.frontiersin.org/articles/10.3389/fpsyg.2021.641994\nLubin, A., Otterstrom, J., Hoade, Y., Bjedov, I., Stead, E., Whelan, M., Gestri, G., Paran, Y., &\nPayne, E. (2021). A versatile, automated and high-throughput drug screening platform\nfor zebrafish embryos. Biology Open, 10(9), bio058513.\nhttps://doi.org/10.1242/bio.058513\nLucon-Xiccato, T., Gatto, E., Fontana, C. M., & Bisazza, A. (2023). Quantity discrimination in\nnewly hatched zebrafish suggests hardwired numerical abilities. Communications\nBiology, 6(1), 1–8. https://doi.org/10.1038/s42003-023-04595-7\nLuu, P., Fraser, S. E., & Schneider, F. (2024). More than double the fun with two-photon\nexcitation microscopy. Communications Biology, 7(1), 364.\nhttps://doi.org/10.1038/s42003-024-06057-0\nMedina, L., Abellán, A., & Desfilis, E. (2019). Evolution of Pallial Areas and Networks Involved in\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nSociality: Comparison Between Mammals and Sauropsids. Frontiers in Physiology, 10,\n894. https://doi.org/10.3389/fphys.2019.00894\nMessina, A., Potrich, D., Perrino, M., Sheardown, E., Miletto Petrazzini, M. E., Luu, P.,\nNadtochiy, A., Truong, T. V., Sovrano, V. A., Fraser, S. E., Brennan, C. H., & Vallortigara,\nG. (2022). Quantity as a Fish Views It: Behavior and Neurobiology. Frontiers in\nNeuroanatomy, 16. https://www.frontiersin.org/articles/10.3389/fnana.2022.943504\nMessina, A., Potrich, D., Schiona, I., Sovrano, V. A., Fraser, S. E., Brennan, C. H., &\nVallortigara, G. (2022). Neurons in the Dorso-Central Division of Zebrafish Pallium\nRespond to Change in Visual Numerosity. Cerebral Cortex (New York, N.Y.: 1991),\n32(2), 418–428. https://doi.org/10.1093/cercor/bhab218\nMoyer, R. S., & Landauer, T. K. (1967). Time required for Judgements of Numerical Inequality.\nNature, 215(5109), 1519–1520. https://doi.org/10.1038/2151519a0\nMuto, A., Ohkura, M., Abe, G., Nakai, J., & Kawakami, K. (2013). Real-Time Visualization of\nNeuronal Activity during Perception. Current Biology, 23(4), 307–311.\nhttps://doi.org/10.1016/j.cub.2012.12.040\nNadtochiy, A., Luu, P., Fraser, S. E., & Truong, T. V. (2023). VoDEx: A Python library for time\nannotation and management of volumetric functional imaging data. Bioinformatics,\nbtad568. https://doi.org/10.1093/bioinformatics/btad568\nNieder, A. (2011). Chapter 8—The Neural Code for Number. In S. Dehaene & E. M. Brannon\n(Eds.), Space, Time and Number in the Brain (pp. 103–118). Academic Press.\nhttps://doi.org/10.1016/B978-0-12-385948-8.00008-6\nNieder, A., Freedman, D. J., & Miller, E. K. (2002). Representation of the quantity of visual items\nin the primate prefrontal cortex. Science (New York, N.Y.), 297(5587), 1708–1711.\nhttps://doi.org/10.1126/science.1072493\nPeirce, J., Gray, J. R., Simpson, S., MacAskill, M., Höchenberger, R., Sogo, H., Kastman, E., &\nLindeløv, J. K. (2019). PsychoPy2: Experiments in behavior made easy. Behavior\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nResearch Methods, 51(1), 195–203. https://doi.org/10.3758/s13428-018-01193-y\nPfeifer, J. H., Allen, N. B., Byrne, M. L., & Mills, K. L. (2018). Modeling Developmental Change:\nContemporary Approaches to Key Methodological Challenges in Developmental\nNeuroimaging. Developmental Cognitive Neuroscience, 33, 1–4.\nhttps://doi.org/10.1016/j.dcn.2018.10.001\nPiazza, M. (2010). Neurocognitive start-up tools for symbolic number representations. Trends in\nCognitive Sciences, 14(12), 542–551. https://doi.org/10.1016/j.tics.2010.09.008\nPiazza, M., Izard, V., Pinel, P., Le Bihan, D., & Dehaene, S. (2004). Tuning curves for\napproximate numerosity in the human intraparietal sulcus. Neuron, 44(3), 547–555.\nhttps://doi.org/10.1016/j.neuron.2004.10.014\nRodríguez, F., López, J. C., Vargas, J. P., Gómez, Y., Broglio, C., & Salas, C. (2002).\nConservation of Spatial Memory Function in the Pallial Forebrain of Reptiles and\nRay-Finned Fishes. The Journal of Neuroscience, 22(7), 2894–2903.\nhttps://doi.org/10.1523/JNEUROSCI.22-07-02894.2002\nSartori, B. M., Júnior, R. E. M., Paiva, I. M., Moraes, I. B., Murgas, L. D. S., & Brunialti Godard,\nA. L. (2022). Acute Ethanol Exposure Leads to Long-Term Effects on Memory, Behavior,\nand Transcriptional Regulation in the Zebrafish Brain (SSRN Scholarly Paper 4265391).\nhttps://doi.org/10.2139/ssrn.4265391\nSheardown, E., Torres-Perez, J. V., Anagianni, S., Fraser, S. E., Vallortigara, G., Butterworth, B.,\nMiletto-Petrazzini, M. E., & Brennan, C. H. (2022a). Characterizing ontogeny of quantity\ndiscrimination in zebrafish. Proceedings of the Royal Society B: Biological Sciences,\n289(1968), 20212544. https://doi.org/10.1098/rspb.2021.2544\nSheardown, E., Torres-Perez, J. V., Anagianni, S., Fraser, S. E., Vallortigara, G., Butterworth, B.,\nMiletto-Petrazzini, M. E., & Brennan, C. H. (2022b). Characterizing ontogeny of quantity\ndiscrimination in zebrafish. Proceedings of the Royal Society B: Biological Sciences,\n289(1968), 20212544. https://doi.org/10.1098/rspb.2021.2544\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nSzkudlarek, E., & Brannon, E. M. (2017). Does the approximate number system serve as a\nfoundation for symbolic mathematics? Language Learning and Development  : The\nOfficial Journal of the Society for Language Development, 13(2), 171–190.\nhttps://doi.org/10.1080/15475441.2016.1263573\nTemizer, I., Donovan, J. C., Baier, H., & Semmelhack, J. L. (2015). A Visual Pathway for\nLooming-Evoked Escape in Larval Zebrafish. Current Biology, 25(14), 1823–1834.\nhttps://doi.org/10.1016/j.cub.2015.06.002\nTempleton, C. N., & Greene, E. (2007). Nuthatches eavesdrop on variations in heterospecific\nchickadee mobbing alarm calls. Proceedings of the National Academy of Sciences of the\nUnited States of America, 104(13), 5479–5482.\nhttps://doi.org/10.1073/pnas.0605183104\nTrinh, L. A., & Fraser, S. E. (2013). Enhancer and gene traps for molecular imaging and genetic\nanalysis in zebrafish. Development, Growth & Differentiation, 55(4), 434–445.\nhttps://doi.org/10.1111/dgd.12055\nTruong, T. V., Supatto, W., Koos, D. S., Choi, J. M., & Fraser, S. E. (2011). Deep and fast live\nimaging with two-photon scanned light-sheet microscopy. Nature Methods, 8(9), Article\n9. https://doi.org/10.1038/nmeth.1652\nTzambazis, K., & Stough, C. (2000). Alcohol impairs speed of information processing and\nsimple and choice reaction time and differentially impairs higher-order cognitive abilities.\nAlcohol and Alcoholism (Oxford, Oxfordshire), 35(2), 197–201.\nhttps://doi.org/10.1093/alcalc/35.2.197\nViswanathan, P., & Nieder, A. (2013). Neuronal correlates of a visual “sense of number” in\nprimate parietal and prefrontal cortices. Proceedings of the National Academy of\nSciences, 110(27), 11187–11192. https://doi.org/10.1073/pnas.1308141110\nVito, G. de, Ricci, P., Turrini, L., Gavryusev, V., Müllenbroich, C., Tiso, N., Vanzi, F., Silvestri, L.,\n& Pavone, F. S. (2020). Effects of excitation light polarization on fluorescence emission\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nin two-photon light-sheet microscopy. Biomedical Optics Express, 11(8), 4651–4665.\nhttps://doi.org/10.1364/BOE.396388\nVito, G. de, Vito, G. de, Vito, G. de, Turrini, L., Turrini, L., Turrini, L., Müllenbroich, C.,\nMüllenbroich, C., Müllenbroich, C., Ricci, P., Sancataldo, G., Sancataldo, G.,\nMazzamuto, G., Mazzamuto, G., Tiso, N., Sacconi, L., Sacconi, L., Fanelli, D., Silvestri,\nL., … Pavone, F. S. (2022). Fast whole-brain imaging of seizures in zebrafish larvae by\ntwo-photon light-sheet microscopy. Biomedical Optics Express, 13(3), 1516–1536.\nhttps://doi.org/10.1364/BOE.434146\nVossen, L. E., Brunberg, R., Rådén, P., Winberg, S., & Roman, E. (2022). Sex-Specific Effects\nof Acute Ethanol Exposure on Locomotory Activity and Exploratory Behavior in Adult\nZebrafish (Danio rerio). Frontiers in Pharmacology, 13.\nhttps://doi.org/10.3389/fphar.2022.853936\nWaskom, M. L. (2021). seaborn: Statistical data visualization. Journal of Open Source Software,\n6(60), 3021. https://doi.org/10.21105/joss.03021\nWeber, T., & Köster, R. (2013). Genetic tools for multicolor imaging in zebrafish larvae.Methods\n(San Diego, Calif.), 62(3), 279–291. https://doi.org/10.1016/j.ymeth.2013.07.028\nWolf, S., Supatto, W., Debrégeas, G., Mahou, P., Kruglik, S. G., Sintes, J.-M., Beaurepaire, E., &\nCandelier, R. (2015). Whole-brain functional imaging with two-photon light-sheet\nmicroscopy. Nature Methods, 12(5), 379–380. https://doi.org/10.1038/nmeth.3371\nXu, F., & Spelke, E. S. (2000). Large number discrimination in 6-month-old infants. Cognition,\n74(1), B1–B11. https://doi.org/10.1016/s0010-0277(99)00066-9\nYang, E., Zwart, M. F., James, B., Rubinov, M., Wei, Z., Narayan, S., Vladimirov, N., Mensh, B.\nD., Fitzgerald, J. E., & Ahrens, M. B. (2022). A brainstem integrator for self-location\nmemory and positional homeostasis in zebrafish. Cell, 185(26), 5011-5027.e20.\nhttps://doi.org/10.1016/j.cell.2022.11.022\nYushkevich, P. A., Piven, J., Hazlett, H. C., Smith, R. G., Ho, S., Gee, J. C., & Gerig, G. (2006).\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint \n\nUser-guided 3D active contour segmentation of anatomical structures: Significantly\nimproved efficiency and reliability. NeuroImage, 31(3), 1116–1128.\nhttps://doi.org/10.1016/j.neuroimage.2006.01.015\nZanon, M., Potrich, D., Bortot, M., & Vallortigara, G. (2022). Towards a standardization of\nnon-symbolic numerical experiments: GeNEsIS, a flexible and user-friendly tool to\ngenerate controlled stimuli. Behavior Research Methods, 54(1), 146–157.\nhttps://doi.org/10.3758/s13428-021-01580-y\n.CC-BY 4.0 International licenseavailable under a \n(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made \nThe copyright holder for this preprintthis version posted September 1, 2024. ; https://doi.org/10.1101/2024.08.30.610552doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}