The ability to represent approximate quantities appears to be phylogenetically widespread, but the selective pressures and proximate mechanisms favouring this ability remain unknown. We analysed quantity discrimination data from 672 subjects across 33 bird and mammal species, using a novel Bayesian model that combined phylogenetic regression with a model of number psychophysics and random effect components. This allowed us to combine data from 49 studies and calculate the Weber fraction (a measure of quantity representation precision) for each species. We then examined which cognitive, socioecological and biological factors were related to variance in Weber fraction. We found contributions of phylogeny to quantity discrimination performance across taxa. Of the neural, socioecological and general cognitive factors we tested, cortical neuron density and domain-general cognition were the strongest predictors of Weber fraction, controlling for phylogeny. Our study is a new demonstration of evolutionary constraints on cognition, as well as of a relation between species-specific neuron density and a particular cognitive ability. This article is part of the theme issue ‘Systems neuroscience through the lens of evolutionary theory’.

Much research has focused on the development and evolution of cognition in the realm of numerical knowledge in human and nonhuman animals but often fails to take into account ecological realities that, over time, may influence and constrain cognitive abilities in real-life decision-making. Cognitive abilities such as enumerating and timing are central to many psychological and ecological models of behavior, yet our knowledge of how these are affected by environmental fluctuations remains incomplete. Our research bridges the gap between basic cognitive research and ecological decision-making. We used coyotes (Canis latrans) as a model animal system to study decision-making about smaller, more proximal food rewards and larger, more distant food rewards; we tested animals across their four reproductive cycle phases to examine effects of ecological factors such as breeding status and environmental risk on quantitative performance. Results show that coyotes, similar to other species, spatially discount food rewards while foraging. The degree to which coyotes were sensitive to the risk of obtaining the larger food reward, however, depended on the season in which they completed the foraging task, the presence of unfamiliar humans (i.e., risk), and the presence of conspecifics. Importantly, our results support that seasonal variations drive many differences in nonhuman animal behavior and cognition (e.g., hibernation, breeding, food resource availability). Further, it may be useful in the future to extend this work to humans because seasons may influence human cognition as well, and this remains unexplored in the realms of enumeration, timing, and spatial thinking.

Article Figures and data Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Abstract Absence is a notion that is usually captured by language-related concepts like zero or negation. Whether nonlinguistic creatures encode similar thoughts is an open question, as everyday behavior marked by absence (of food, of social partners) can be explained solely by expecting presence somewhere else. We investigated 8-day-old chicks’ looking behavior in response to events violating expectations about the presence or absence of an object. We found different behavioral responses to violations of presence and absence, suggesting distinct underlying mechanisms. Importantly, chicks displayed an avian signature of novelty detection to violations of absence, namely a sex-dependent left-eye-bias. Follow-up experiments excluded accounts that would explain this bias by perceptual mismatch or by representing the object at different locations. These results suggest that the ability to spontaneously form representations about the absence of objects likely belongs to the initial cognitive repertoire of vertebrate species. Editor's evaluation The research detailed in this manuscript investigates whether young chicks represent the absence of objects. This work is important to multiple fields of inquiry, such as ethology and neuroscience, and is the first time that this ability has been demonstrated to be exhibited spontaneously, as opposed to after many trials of experience. https://doi.org/10.7554/eLife.67208.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Imagine looking at a domino that has four dots on one end and no dots on the other. The ways we can represent specific items (e.g., four dots) have been intensively investigated for decades. First, one can think of these dots as individual objects. Investigations targeting object cognition revealed that roughly four objects can be tracked and maintained in mind simultaneously, even when they are moving or are occasionally occluded (Kahneman et al., 1992; Scholl and Pylyshyn, 1999). Another way to look at the dots is to encode them as a set of objects. Such encoding is performed by the approximate number system, which provides imprecise representations of sets to pre- and nonlinguistic creatures as well. However, in contrast to the object tracking system, information in this number system is only an approximation of the size of the set and it is sensitive to proportional rather than to absolute differences (Dehaene, 1997). Both systems emerge early in the individual development (Piazza, 2010) and are shared by several species (e.g., mosquito fish, domestic chicks, rhesus monkeys, and great apes; Haun et al., 2011; Brannon and Merritt, 2011; Vallortigara, 2012; Brannon and Roitman, 2003). A third way to think about the four dots on the domino is as a symbolic (and precise) representation of the number ‘4’ (Dehaene, 1997; Carey, 2009). Interpreting such symbols clearly requires processing number concepts and being familiar with the specific notation system (e.g., the conventions of dominos or the Arabic numerals) (Carey, 2009). Now, let us focus on the other end of the domino. How will the blank square turn into zero in our mind? Such a representation may be outside the scope of the abovementioned cognitive systems: ‘no object’ is not tracked by the visual system, ‘no dots’ is not proportional to anything, and ‘empty space’ can denote a number only in special circumstances. Indeed, understanding the absence of something as ‘nothing’ is frequently related to complex and human-specific concepts, such as zero or linguistic negation. In this work, we focus on the representation of absence that should, however, rely on a more basic capacity and possibly be part of the initial cognitive repertoire of different species and we will target such abilities in 8-day-old domestic chicks. Clear evidence regarding when human children start representing the absence of objects comes from language development research. Negation conveying absence (e.g., ‘all gone’) emerges among the first linguistic expressions between 1 and 2 years of life (Bloom, 1970; Pea, 1980; Choi, 1988). Some years later, preschoolers can flexibly use sentential negation to express the absence of something in a numerical context (Bialystok and Codd, 2000), and can recruit complex numerical concepts, such as zero (Wellman and Miller, 1986; Merritt and Brannon, 2013). While by the age of 5 children seem to successfully operate with counterintuitive concepts like zero and nothing, the cognitive foundations of this human capacity are frequently suggested to be grounded in linguistic abilities. How would pre- and nonlinguistic creatures see the zero end of the domino? Nonhuman animals were found to accommodate stimuli defined by the lack of a stimulant in two main types of tasks: numerical and perceptual decision tasks. Studies involving numerical tasks indicate that monkeys can integrate empty sets with other sets relying on the approximate number system (Biro and Matsuzawa, 2001; Merritt et al., 2009; Howard et al., 2018). For instance, comparisons including empty sets are also subject to distance effects, characteristic to the approximate number system (the smaller the distance between two numerosities, the more errors the subjects make) Merritt et al., 2009. Empty sets, similar to other numerosities, are represented in the number specific areas of the monkey brain (Macaca fuscata: Okuyama et al., 2015; Macaca mulatta: Ramirez-Cardenas et al., 2016), which provides further evidence for the involvement of this system. Furthermore, an interesting finding suggests that Ai, the chimpanzee learned to use a symbol for zero (Biro and Matsuzawa, 2001). However, Ai’s performance likely reflected a rather limited conceptual understanding of zero, as she did not show transfer effects when switching from cardinality judgments to ordering tasks. Although these findings are very impressive, it is still unclear how the approximate number system could represent exactly no objects, given that it is specialized for approximating numerosity. Evidence pointing to the possibility that this system may not be appropriate for such encoding comes from studies with preschoolers (Merritt and Brannon, 2013) and monkeys Merritt et al., 2009 who tend to fail to discriminate an empty set from one item. Thus, the question emerges how the representation of exactly no objects might be encoded. ‘Nothing’ is an amount of less than one, but crucially, it can also be thought of as one side of the binary information of presence and absence. The role of the approximate number system in the first, continuous conceptualization of ‘nothing’ seems unequivocal, however, the binary coding of ‘nothing’ is a more peculiar subject of investigation. Things can be present or absent, yet how these intuitively simple opposing categories are formed and encoded is largely unexplored. Bermúdez, 2003 proposed that contrary concepts, like absence/presence might be available even for nonhuman animals. Such contrary concepts encompass alternatives that are mutually exclusive (e.g., nothing can be present and absent at the same time) and may support specific inferences. The availability of contrary concepts in pre- and nonlinguistic animals has not been targeted by researchers, nevertheless, extensive research cumulating for over a century suggests that various species are able to exploit the presence and absence of stimulus (Pearce, 2011). However, clear evidence that absence is explicitly represented is scarce. Importantly, not representing a stimulus is not equivalent with representing its absence, in a way that this would be distinguishable from a nonspecific default activation of a system (de Lafuente and Romo, 2005; Merten and Nieder, 2012). A recent study has targeted this issue, by investigating prefrontal neural activations in monkeys while performing abstract detection decisions regarding the presence and absence of stimuli (Merten and Nieder, 2012). Notably, in this study the stimulus presentation phase was separated from a later phase preceding decision. While presence-specific neurons were found to be active when the animal perceived the stimulus and also later when making a decision, absence-specific neurons showed activation only when the subject decided about absence. This finding, besides providing evidence for forming some representations of absence in monkeys, points to the possibility that different processes are involved in encoding the presence and absence of a stimulus. Asymmetries between performance relying on representing the presence and absence of stimuli were documented in behavioral tasks as well. Pigeons (Hearst, 1984; as well as human adults, Newman et al., 1980) display feature-positive biases in learning tasks. Pigeons learned relatively easily the relation between the presence of a stimulus and food, but they had difficulties with discovering a similar relation between the absence of a stimulus and food. In line with such asymmetries, human infants automatically detect and keep in mind the presence of objects after occlusion, while they seem to fail to do so with the absence of objects (Wynn and Chiang, 2016; Kaufman et al., 2003). While in these tasks the representation of an object being present can be supported by the object tracking system (Kahneman et al., 1992), it is unclear how a specific object that is absent could be encoded by the same system (note that simply discarding the object file results in no representation whatsoever and it is not equivalent to representing, for instance, ‘the lion is absent’). In fact, the representation of absence might be beyond the scope of this system, as it operates with spatiotemporal information of the items, which absent objects do not have. Thus, up to date it is unclear under which circumstances individuals form representations of ‘no object’, whether such representations can be used for further processing as readily as the presence of a stimulus, and most importantly, whether they can be encoded spontaneously. Absence is trivial in experience, but peculiar in information processing. While some nonhuman species show success in dealing with the absence of stimuli in experimental tasks involving training or hundreds of trials (Merten and Nieder, 2012), it is not yet known whether nonlinguistic creatures can spontaneously rely on such information and what inferences they can draw from it. A possible way to investigate the emergence and the nature of the representation of absence is to target developmentally precocious animals. We addressed these questions by studying naive domestic chicks — creatures that start to search for food soon after hatching, and could make good use of information regarding the presence or absence of potential food sources and social partners. In four experiments, 8-day-old chicks were placed inside a confining cylinder that had a small circular opening to provide the opportunity of putting through the head and attend the events in the testing arena. The age was determined by the specific paradigm we used in the present work (see the details of imprinting and familiarization with the apparatus in the Materials and methods) and the dependent measures we targeted (looking time and lateralization index). This is also the age when chicks were found to show strong lateralized responses to familiar and unfamiliar objects (Dharmaretnam and Andrew, 1994). Chicks were presented with events in which the target object they were imprinted to (for the details see Materials and methods) was either hidden behind a screen or was removed from the arena. Afterwards, the screen was dropped and it revealed an expected outcome (congruent with the previous event, e.g., the object appeared from behind the screen after it was hidden behind the screen) or an unexpected outcome (contradicting the previous events, e.g., the object appeared from behind the screen after it was removed from the arena). If in this latter case chicks represented the object as being absent, and they see an incongruent outcome (the object ‘magically’ appears), they should show a different behavior compared to when they see the scene-congruent outcome (the object was present and appears expectedly). We measured how long the chicks looked at these outcomes and which eye they used to inspect the scene. Regarding our first measurement, based on former research with human infants (Baillargeon et al., 1985) and a study involving adult rooks (Bird and Emery, 2010), we expected longer looking for unexpected outcomes. For instance, Bird and Emery, 2010 have found that rooks looked longer to unexpected events that violate the laws of physics (e.g., objects remaining in the air without any support) compared to expected events (e.g., objects in a support relation with other objects), indicating a violation of their expectation or surprise, a method commonly used in infancy research to study a wide range of competencies. For our second measurement, we coded which eye the chicks used to inspect the expected and unexpected outcomes. Earlier research suggests that eye usage is modulated by the novelty of the object attended, and also by sex (Rogers and Anson, 1979; Dharmaretnam and Andrew, 1994; Vallortigara and Andrew, 1991). A preferential use of the left eye (mainly feeding the right brain structures) is associated with response to novelty in birds with laterally placed eyes such as domestic chicks (Rogers et al., 2013). Note that the selective involvement of structures in the right hemisphere when attending to novel stimuli is widely documented among vertebrates (review in Rogers et al., 2013), being likely a general feature inherited by early chordates (MacNeilage et al., 2009). In animals with laterally placed eyes and lack of callosum, such as birds, fish, reptiles, and amphibians the brain asymmetry can be easily documented without any invasive procedure by simply measuring preferences in eye use (Vallortigara, 2000; Vallortigara and Versace, 2017; Vallortigara and Rogers, 2020). In the present study, we expected a left-eye bias to the novel unexpected outcomes compared to the expected ones. In addition, interestingly, lateralized sex differences have been repeatedly observed in response to novel objects. For instance, Vallortigara and Andrew, 1991 documented stronger left-eye-mediated choices of unfamiliar stimuli for males compared to females, and different preferences for unfamiliar and familiar objects between sexes. Lateralized sex differences have been found also by Dharmaretnam and Andrew, 1994, where unfamiliar stimuli evoked left-eye bias in females, but not in males. Vallortigara and Andrew, 1991 observed that left-eye (and binocular) males preferred unfamiliar objects, while left-eye (and binocular) females preferred familiar objects, whereas both males and females tested with the right eye did not exhibit significant preferences for familiar or novel stimuli. Hence, a potential modulation of sex in eye use must be considered for the exploration of unexpected vs. expected scenes in our study as well. Study 1 In Study 1, we investigated whether the chicks encoded the presence (Experiment 1) and the absence (Experiment 2) of the object behind the screen. In Experiment 1 (Encoding Presence), chicks (n = 27, 13 females, 14 males) watched as the object moved behind the screen till full occlusion, and then they either saw it moving out of the scene (Expected Disappearance condition) or did not see it moving out (Unexpected Disappearance condition). Both scenes ended identically, by the screen falling and revealing an outcome with no object being present (Figure 1A). In Experiment 2 (Encoding Absence), the test trials started with the screen in a lowered position. A different group of chicks (n = 28, 15 females, 13 males) observed an object moving to the area behind the lowered screen (Expected Appearance condition) or moving out of the scene (Unexpected Appearance condition) before the screen was raised (Figure 1B). Both scenes ended by the screen falling and revealing an outcome with the object being present. We used a repeated measures design; thus, each chick was presented with both the Expected and the Unexpected conditions. We coded the chicks’ overall Looking Times and Lateralization Index (the difference between left- and right-eye usage proportional to the total eye usage) in the outcome phases. Figure 1 Download asset Open asset Schematic illustration of the events in Experiments 1 and 2. (A) Experiment 1 – Encoding Presence. The upper panels depict the events in the Expected Disappearance condition, where the target object was removed from the arena before the screen was lowered revealing the empty space behind it. The lower panels depict the events in the Unexpected Disappearance condition, where the target object was placed behind the screen visibly to the chick but then it was secretly removed from the arena. When the screen was lowered, it revealed the empty space behind. (B) Experiment 2 – Encoding Absence. The upper panels depict the events in the Expected Appearance condition, where the target object moved behind the screen and when the screen was lowered, it revealed the presence of the object. The lower panels depict the Unexpected Appearance condition, in which the target object was visibly removed from the arena, and then the vertical position of the screen was restored. Afterwards, the target object was secretly reintroduced into the arena, and when the screen was lowered, it revealed the target object. Results Chicks’ overall Looking Times were differently modulated as a function of the outcomes violating or confirming the chicks’ expectations about the presence and the absence of the object in Experiments 1 and 2 (Figure 2A). We run a repeated measures analysis of variance (ANOVA) on the square root transformed data set. A 2 × 2 × 2 repeated measures ANOVA with Experiment (Experiment 1: Encoding Presence vs. Experiment 2: Encoding Absence), Outcome (Expected vs. Unexpected), and Sex (Female vs. Male) as factors yielded a significant interaction between Experiment and Outcome (F1, 51 = 4.244, p = 0.045, ηp2 = 0.077) with no other effects (IBM SPSS Statistics 20). In Experiment 1, chicks’ looking behavior seemed to be in line with the prediction of longer looking times for unexpected outcomes (untransformed means: Unexpected Disappearance: M = 15.421; SD = 7.34; Expected Disappearance: M = 12.671, SD = 6.566; Scheffé test p = 0.078, η2 = 0.037); however, this was not the case for Experiment 2 (Unexpected Appearance: M = 12.906; SD = 8.409; Expected Appearance: M = 14.556, SD = 9.351; Scheffé test, p = 0.273, η2 = 0.008). Subjects looked longer to the Unexpected outcomes in Experiment 1 compared to Experiment 2 (Scheffé test, p = 0.028, d = 0.318). Human infants show similar looking patterns to such scenes (i.e., longer looks to violations of presence but not to that of absence; Wynn and Chiang, 2016; Kaufman et al., 2003), and it was suggested that they are more sensitive to violations of presence compared to violations regarding the absence of objects. Figure 2 Download asset Open asset Results of Experiments 1 and 2. (A) Mean Looking Times elicited by Expected and Unexpected outcomes in the two experiments. The asterisk indicates a significant interaction between Outcome (Expected/Unexpected) and Experiment (F1,51 = 4.244, p = 0.045) and a significantly higher looking time observed for the Unexpected outcome in Experiment 1 (Encoding Absence) than in Experiment 2 (Encoding Presence) (Scheffé test, p = 0.028). Error bars represent standard error of the mean. (B) Lateralization Index as a function of Experiment and types of outcome. Asterisks indicate a significant interaction between Outcome (Expected/Unexpected) and Experiment (F1,51 = 4.652, p = 0.036) and a significantly higher left-eye bias observed for the Unexpected outcome in Experiment 2 (Encoding Absence) than in Experiment 1 (Encoding Presence) (Scheffé test, p = 0.023), suggesting that the Lateralization Index is sensitive to violations of expectation regarding the absence of objects. For box plots, the horizontal line represents the median, yellow diamonds depict the mean values, box height depicts first and third quartiles, and vertical lines represent the 95th percentile. Dots represent the outliers of the data set. A 2 × 2 × 2 repeated measures ANOVA performed on the Lateralization Index revealed a significant Experiment by Outcome interaction (F1, 51 = 4.652, p = 0.036, ηp2 = 0.084), with no other significant effects. Pairwise comparisons revealed that in the Unexpected Outcome condition subjects displayed a greater left-eye bias in Experiment 2 (Encoding Absence, M = 0.193, SD = 0.517) compared to Experiment 1 (Encoding Presence, M = −0.138, SD = 0.459; Scheffé test, p = 0.023, d = 0.677). The other pairwise comparisons did not reach significance, though the pattern of eye usage was congruent with our predictions in Experiment 2 (greater left-eye bias for the unexpected outcomes: Unexpected Appearance, M = 0.193, SD = 0.518; Expected Appearance M = −0.054, SD = 0.518; Scheffé test, p = 0.091, η2 = 0.05), but not in Experiment 1. These data suggest that the Lateralization Index may be sensitive to violations of expectation regarding the absence of objects, but not regarding the presence of objects (Figure 2B). A positive Lateralization Index reflects a left-eye bias, which was previously linked to detecting novel outcomes (Rogers and Anson, 1979; Dharmaretnam and Andrew, 1994), however in our case it seems to be specific to violations of expectations about the object’s absence. Thus, while there is currently no report suggesting that human infants (Wynn and Chiang, 2016; Kaufman et al., 2003) or other animals would spontaneously encode absence, the Lateralization Index in the present study points to 8-day-old chicks’ ability to encode and form expectations about ‘no objects’ at a particular location. The left-eye bias, found in response to the unexpected appearance of the object, likely reflected the chicks’ exploration, and attempt of identification of the unexpectedly emerging ‘new’ item. Importantly, this item was new only if chicks had encoded that the identically looking familiar item left, and thus it was absent from the scene. They likely investigated the ‘new’ object more carefully with their left eye because they expected the old object to be absent. Interestingly, in contrast to the left-eye bias, overall looking time did not seem to be sensitive to the unexpected appearance of the object, and therefore to absence. We assumed that if chicks are sensitive to both presence and absence violations, they should show surprise (and produce longer looking times) in the Unexpected outcomes compared to the Expected outcomes in both experiments. Instead, we found an Experiment by Outcome interaction. Only Experiment 1 (Encoding Presence) revealed a looking time that patterned with our prediction, but not Experiment 2 (Encoding Absence). The present work was, however, to our knowledge, the first attempt to measure young chicks’ looking time in violation of expectation paradigms, thus it may be difficult to interpret this pattern. It may be the case that in our setting overall looking times did not reveal clear differences due to statistical variability. To sum up, in Study 1 we found asymmetric patterns of looking times and lateralization in response to outcomes congruent or violating chicks’ expectations of presence and absence. Furthermore, we found that the left-eye bias seems to reflect chicks’ ability to form an expectation about the absence of an object. In Study 2, we targeted the cognitive processes that could support the representation of absence. Study 2 In two additional experiments, we focused on the left-eye bias effect observed in Experiment 2. We aimed on the one hand to strengthen our findings regarding chicks’ absence encoding, and on the other hand to test the possible representational processes underlying such a behavior. According to one possibility, chicks in Experiment 2 (Encoding Absence) might have not formed an actual representation that there was no object behind the screen, instead their reaction to the unexpected appearance could have been perceptually driven and derived from detecting the mismatch between the memory of the empty space behind the screen (i.e., an iconic, picture-like representation of the empty floor, and the wall of the testing arena) and the perceived outcome (i.e., the scene with the object). This possibility, however, would predict no left-eye bias in a situation where there was no possibility for perceptually encoding the empty space. In contrast, left-eye bias without relying on the percept of an empty space would point to a more complex mental representation of absence that is inferred from the sequence of events (object going in/going out) (Experiment 3. Absence vs. Perceptual Comparison). A second alternative explanation might be that the chicks did not encode the absence of the object, but tracked the location of the target object even when it left the scene, and encoded its presence somewhere outside the arena, and the left-eye bias reflected their ‘surprise’ of seeing this object at an unexpected location (i.e., behind the screen, inside the arena). According to this alternative, an outcome which would feature a different object behind the occluder should not elicit a left-eye bias (Experiment 4. Absence vs. Tracking). In the second study, Experiments 3 and 4 test these alternative explanations, respectively, and aimed at replicating the findings from Experiment 2. Experiment 3 (Absence vs. Perceptual Comparison) followed the procedure of Experiment 2. However, unlike in Experiment 2, each trial started with the screen in upright position preventing the chicks (n = 31, 15 females, 16 males) to see the space occluded by the screen at the beginning of the trial (Figure 3A). This manipulation aimed at testing chicks’ ability to update their expectation about what is (not) behind the screen without giving them the opportunity to perform perceptual comparison between an initial empty scene and the outcome. Without clear perceptual evidence, we expect chicks to arrive at encoding absence by first representing the object being behind the screen, and then inferring the outcome (i.e., absence) after the object being removed from behind the screen. Based on the observation of these events chicks may compute the absence objects and show similar behavior (i.e., left-eye bias in response to the unexpected appearance of the object) as we observed in Experiment 2. Figure 3 Download asset Open asset Schematic illustration of the events in Experiments 3 and 4. (A) Experiment 3 – Absence vs. Perceptual Comparison. The upper panels depict the event in the Expected Appearance condition, where the target object moved behind the screen and when the screen was lowered, it revealed the object behind. The lower panels depict the events in the Unexpected Appearance condition, in which the object first moved behind the screen and then it was visibly removed from the arena. Afterwards, the object was secretly placed behind the screen and when the screen was lowered, it revealed the object. Note that, in both conditions, the screen’s initial position was vertical. (B) Experiment 4 – Tracking vs. Absence. The upper panels depict the events in the Expected Appearance condition, which was identical to the same condition of Experiment 3. The lower panels depict the Unexpected Appearance condition where the (green) target object first moved behind the screen and then it was visibly removed from the arena. Afterwards, the red object was secretly reintroduced behind the screen. In the outcome phase, the screen was lowered and the red object appeared. Experiment 4 (Absence vs. Tracking) (n = 22, 9 females, 13 males) was the same as Experiment 3, except that in the Unexpected Appearance condition the target object that moved behind the screen and then left the scene was different from the object that appeared when the screen was lowered in the outcome phase (Figure 3B). Finding a second object at the previously empty location should not be surprising if chicks simply track the location of the first target object. However, finding an object at a location that is represented as empty would lead to surprise even if this object is different from the one that has left. Thus a left-eye bias in the Unexpected Appearance condition of Experiment 4 would provide evidence for chicks’ capacity to form expectation about absence of objects at a specific location that must be therefore empty. In contrast, the lack of such response would rather point to object tracking processes underlying chicks’ left-eye bias in Experiment 2 that resulted in encoding the presence of the first object at a different location (outside the scene). Results We analyzed the Lateralization Index of Experiments 3 and 4 with a 2 × 2 × 2 repeated measures ANOVA with Experiment, Outcome, and Sex as factors. There was a main effect of Outcome (F1, 49 = 4.29, p = 0.044, ηp2 = 0.081), revealing more positive values (more usage of the left eye) in response to unexpected outcomes (M = 0.087, SD = 0.5) compared to expected outcomes (M = −0.045, SD = 0.459), similar to Experiment 2. Additionally, a significant interaction was observed between Outcome and Sex (F1, 49 = 4.804, p = 0.033, ηp2 = 0.089), indicating females’ sensitivity to events violating their expectations regarding absence of ent

Article Figures and data Abstract Editor's evaluation Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Abstract Absence is a notion that is usually captured by language-related concepts like zero or negation. Whether nonlinguistic creatures encode similar thoughts is an open question, as everyday behavior marked by absence (of food, of social partners) can be explained solely by expecting presence somewhere else. We investigated 8-day-old chicks’ looking behavior in response to events violating expectations about the presence or absence of an object. We found different behavioral responses to violations of presence and absence, suggesting distinct underlying mechanisms. Importantly, chicks displayed an avian signature of novelty detection to violations of absence, namely a sex-dependent left-eye-bias. Follow-up experiments excluded accounts that would explain this bias by perceptual mismatch or by representing the object at different locations. These results suggest that the ability to spontaneously form representations about the absence of objects likely belongs to the initial cognitive repertoire of vertebrate species. Editor's evaluation The research detailed in this manuscript investigates whether young chicks represent the absence of objects. This work is important to multiple fields of inquiry, such as ethology and neuroscience, and is the first time that this ability has been demonstrated to be exhibited spontaneously, as opposed to after many trials of experience. https://doi.org/10.7554/eLife.67208.sa0 Decision letter Reviews on Sciety eLife's review process Introduction Imagine looking at a domino that has four dots on one end and no dots on the other. The ways we can represent specific items (e.g., four dots) have been intensively investigated for decades. First, one can think of these dots as individual objects. Investigations targeting object cognition revealed that roughly four objects can be tracked and maintained in mind simultaneously, even when they are moving or are occasionally occluded (Kahneman et al., 1992; Scholl and Pylyshyn, 1999). Another way to look at the dots is to encode them as a set of objects. Such encoding is performed by the approximate number system, which provides imprecise representations of sets to pre- and nonlinguistic creatures as well. However, in contrast to the object tracking system, information in this number system is only an approximation of the size of the set and it is sensitive to proportional rather than to absolute differences (Dehaene, 1997). Both systems emerge early in the individual development (Piazza, 2010) and are shared by several species (e.g., mosquito fish, domestic chicks, rhesus monkeys, and great apes; Haun et al., 2011; Brannon and Merritt, 2011; Vallortigara, 2012; Brannon and Roitman, 2003). A third way to think about the four dots on the domino is as a symbolic (and precise) representation of the number ‘4’ (Dehaene, 1997; Carey, 2009). Interpreting such symbols clearly requires processing number concepts and being familiar with the specific notation system (e.g., the conventions of dominos or the Arabic numerals) (Carey, 2009). Now, let us focus on the other end of the domino. How will the blank square turn into zero in our mind? Such a representation may be outside the scope of the abovementioned cognitive systems: ‘no object’ is not tracked by the visual system, ‘no dots’ is not proportional to anything, and ‘empty space’ can denote a number only in special circumstances. Indeed, understanding the absence of something as ‘nothing’ is frequently related to complex and human-specific concepts, such as zero or linguistic negation. In this work, we focus on the representation of absence that should, however, rely on a more basic capacity and possibly be part of the initial cognitive repertoire of different species and we will target such abilities in 8-day-old domestic chicks. Clear evidence regarding when human children start representing the absence of objects comes from language development research. Negation conveying absence (e.g., ‘all gone’) emerges among the first linguistic expressions between 1 and 2 years of life (Bloom, 1970; Pea, 1980; Choi, 1988). Some years later, preschoolers can flexibly use sentential negation to express the absence of something in a numerical context (Bialystok and Codd, 2000), and can recruit complex numerical concepts, such as zero (Wellman and Miller, 1986; Merritt and Brannon, 2013). While by the age of 5 children seem to successfully operate with counterintuitive concepts like zero and nothing, the cognitive foundations of this human capacity are frequently suggested to be grounded in linguistic abilities. How would pre- and nonlinguistic creatures see the zero end of the domino? Nonhuman animals were found to accommodate stimuli defined by the lack of a stimulant in two main types of tasks: numerical and perceptual decision tasks. Studies involving numerical tasks indicate that monkeys can integrate empty sets with other sets relying on the approximate number system (Biro and Matsuzawa, 2001; Merritt et al., 2009; Howard et al., 2018). For instance, comparisons including empty sets are also subject to distance effects, characteristic to the approximate number system (the smaller the distance between two numerosities, the more errors the subjects make) Merritt et al., 2009. Empty sets, similar to other numerosities, are represented in the number specific areas of the monkey brain (Macaca fuscata: Okuyama et al., 2015; Macaca mulatta: Ramirez-Cardenas et al., 2016), which provides further evidence for the involvement of this system. Furthermore, an interesting finding suggests that Ai, the chimpanzee learned to use a symbol for zero (Biro and Matsuzawa, 2001). However, Ai’s performance likely reflected a rather limited conceptual understanding of zero, as she did not show transfer effects when switching from cardinality judgments to ordering tasks. Although these findings are very impressive, it is still unclear how the approximate number system could represent exactly no objects, given that it is specialized for approximating numerosity. Evidence pointing to the possibility that this system may not be appropriate for such encoding comes from studies with preschoolers (Merritt and Brannon, 2013) and monkeys Merritt et al., 2009 who tend to fail to discriminate an empty set from one item. Thus, the question emerges how the representation of exactly no objects might be encoded. ‘Nothing’ is an amount of less than one, but crucially, it can also be thought of as one side of the binary information of presence and absence. The role of the approximate number system in the first, continuous conceptualization of ‘nothing’ seems unequivocal, however, the binary coding of ‘nothing’ is a more peculiar subject of investigation. Things can be present or absent, yet how these intuitively simple opposing categories are formed and encoded is largely unexplored. Bermúdez, 2003 proposed that contrary concepts, like absence/presence might be available even for nonhuman animals. Such contrary concepts encompass alternatives that are mutually exclusive (e.g., nothing can be present and absent at the same time) and may support specific inferences. The availability of contrary concepts in pre- and nonlinguistic animals has not been targeted by researchers, nevertheless, extensive research cumulating for over a century suggests that various species are able to exploit the presence and absence of stimulus (Pearce, 2011). However, clear evidence that absence is explicitly represented is scarce. Importantly, not representing a stimulus is not equivalent with representing its absence, in a way that this would be distinguishable from a nonspecific default activation of a system (de Lafuente and Romo, 2005; Merten and Nieder, 2012). A recent study has targeted this issue, by investigating prefrontal neural activations in monkeys while performing abstract detection decisions regarding the presence and absence of stimuli (Merten and Nieder, 2012). Notably, in this study the stimulus presentation phase was separated from a later phase preceding decision. While presence-specific neurons were found to be active when the animal perceived the stimulus and also later when making a decision, absence-specific neurons showed activation only when the subject decided about absence. This finding, besides providing evidence for forming some representations of absence in monkeys, points to the possibility that different processes are involved in encoding the presence and absence of a stimulus. Asymmetries between performance relying on representing the presence and absence of stimuli were documented in behavioral tasks as well. Pigeons (Hearst, 1984; as well as human adults, Newman et al., 1980) display feature-positive biases in learning tasks. Pigeons learned relatively easily the relation between the presence of a stimulus and food, but they had difficulties with discovering a similar relation between the absence of a stimulus and food. In line with such asymmetries, human infants automatically detect and keep in mind the presence of objects after occlusion, while they seem to fail to do so with the absence of objects (Wynn and Chiang, 2016; Kaufman et al., 2003). While in these tasks the representation of an object being present can be supported by the object tracking system (Kahneman et al., 1992), it is unclear how a specific object that is absent could be encoded by the same system (note that simply discarding the object file results in no representation whatsoever and it is not equivalent to representing, for instance, ‘the lion is absent’). In fact, the representation of absence might be beyond the scope of this system, as it operates with spatiotemporal information of the items, which absent objects do not have. Thus, up to date it is unclear under which circumstances individuals form representations of ‘no object’, whether such representations can be used for further processing as readily as the presence of a stimulus, and most importantly, whether they can be encoded spontaneously. Absence is trivial in experience, but peculiar in information processing. While some nonhuman species show success in dealing with the absence of stimuli in experimental tasks involving training or hundreds of trials (Merten and Nieder, 2012), it is not yet known whether nonlinguistic creatures can spontaneously rely on such information and what inferences they can draw from it. A possible way to investigate the emergence and the nature of the representation of absence is to target developmentally precocious animals. We addressed these questions by studying naive domestic chicks — creatures that start to search for food soon after hatching, and could make good use of information regarding the presence or absence of potential food sources and social partners. In four experiments, 8-day-old chicks were placed inside a confining cylinder that had a small circular opening to provide the opportunity of putting through the head and attend the events in the testing arena. The age was determined by the specific paradigm we used in the present work (see the details of imprinting and familiarization with the apparatus in the Materials and methods) and the dependent measures we targeted (looking time and lateralization index). This is also the age when chicks were found to show strong lateralized responses to familiar and unfamiliar objects (Dharmaretnam and Andrew, 1994). Chicks were presented with events in which the target object they were imprinted to (for the details see Materials and methods) was either hidden behind a screen or was removed from the arena. Afterwards, the screen was dropped and it revealed an expected outcome (congruent with the previous event, e.g., the object appeared from behind the screen after it was hidden behind the screen) or an unexpected outcome (contradicting the previous events, e.g., the object appeared from behind the screen after it was removed from the arena). If in this latter case chicks represented the object as being absent, and they see an incongruent outcome (the object ‘magically’ appears), they should show a different behavior compared to when they see the scene-congruent outcome (the object was present and appears expectedly). We measured how long the chicks looked at these outcomes and which eye they used to inspect the scene. Regarding our first measurement, based on former research with human infants (Baillargeon et al., 1985) and a study involving adult rooks (Bird and Emery, 2010), we expected longer looking for unexpected outcomes. For instance, Bird and Emery, 2010 have found that rooks looked longer to unexpected events that violate the laws of physics (e.g., objects remaining in the air without any support) compared to expected events (e.g., objects in a support relation with other objects), indicating a violation of their expectation or surprise, a method commonly used in infancy research to study a wide range of competencies. For our second measurement, we coded which eye the chicks used to inspect the expected and unexpected outcomes. Earlier research suggests that eye usage is modulated by the novelty of the object attended, and also by sex (Rogers and Anson, 1979; Dharmaretnam and Andrew, 1994; Vallortigara and Andrew, 1991). A preferential use of the left eye (mainly feeding the right brain structures) is associated with response to novelty in birds with laterally placed eyes such as domestic chicks (Rogers et al., 2013). Note that the selective involvement of structures in the right hemisphere when attending to novel stimuli is widely documented among vertebrates (review in Rogers et al., 2013), being likely a general feature inherited by early chordates (MacNeilage et al., 2009). In animals with laterally placed eyes and lack of callosum, such as birds, fish, reptiles, and amphibians the brain asymmetry can be easily documented without any invasive procedure by simply measuring preferences in eye use (Vallortigara, 2000; Vallortigara and Versace, 2017; Vallortigara and Rogers, 2020). In the present study, we expected a left-eye bias to the novel unexpected outcomes compared to the expected ones. In addition, interestingly, lateralized sex differences have been repeatedly observed in response to novel objects. For instance, Vallortigara and Andrew, 1991 documented stronger left-eye-mediated choices of unfamiliar stimuli for males compared to females, and different preferences for unfamiliar and familiar objects between sexes. Lateralized sex differences have been found also by Dharmaretnam and Andrew, 1994, where unfamiliar stimuli evoked left-eye bias in females, but not in males. Vallortigara and Andrew, 1991 observed that left-eye (and binocular) males preferred unfamiliar objects, while left-eye (and binocular) females preferred familiar objects, whereas both males and females tested with the right eye did not exhibit significant preferences for familiar or novel stimuli. Hence, a potential modulation of sex in eye use must be considered for the exploration of unexpected vs. expected scenes in our study as well. Study 1 In Study 1, we investigated whether the chicks encoded the presence (Experiment 1) and the absence (Experiment 2) of the object behind the screen. In Experiment 1 (Encoding Presence), chicks (n = 27, 13 females, 14 males) watched as the object moved behind the screen till full occlusion, and then they either saw it moving out of the scene (Expected Disappearance condition) or did not see it moving out (Unexpected Disappearance condition). Both scenes ended identically, by the screen falling and revealing an outcome with no object being present (Figure 1A). In Experiment 2 (Encoding Absence), the test trials started with the screen in a lowered position. A different group of chicks (n = 28, 15 females, 13 males) observed an object moving to the area behind the lowered screen (Expected Appearance condition) or moving out of the scene (Unexpected Appearance condition) before the screen was raised (Figure 1B). Both scenes ended by the screen falling and revealing an outcome with the object being present. We used a repeated measures design; thus, each chick was presented with both the Expected and the Unexpected conditions. We coded the chicks’ overall Looking Times and Lateralization Index (the difference between left- and right-eye usage proportional to the total eye usage) in the outcome phases. Figure 1 Download asset Open asset Schematic illustration of the events in Experiments 1 and 2. (A) Experiment 1 – Encoding Presence. The upper panels depict the events in the Expected Disappearance condition, where the target object was removed from the arena before the screen was lowered revealing the empty space behind it. The lower panels depict the events in the Unexpected Disappearance condition, where the target object was placed behind the screen visibly to the chick but then it was secretly removed from the arena. When the screen was lowered, it revealed the empty space behind. (B) Experiment 2 – Encoding Absence. The upper panels depict the events in the Expected Appearance condition, where the target object moved behind the screen and when the screen was lowered, it revealed the presence of the object. The lower panels depict the Unexpected Appearance condition, in which the target object was visibly removed from the arena, and then the vertical position of the screen was restored. Afterwards, the target object was secretly reintroduced into the arena, and when the screen was lowered, it revealed the target object. Results Chicks’ overall Looking Times were differently modulated as a function of the outcomes violating or confirming the chicks’ expectations about the presence and the absence of the object in Experiments 1 and 2 (Figure 2A). We run a repeated measures analysis of variance (ANOVA) on the square root transformed data set. A 2 × 2 × 2 repeated measures ANOVA with Experiment (Experiment 1: Encoding Presence vs. Experiment 2: Encoding Absence), Outcome (Expected vs. Unexpected), and Sex (Female vs. Male) as factors yielded a significant interaction between Experiment and Outcome (F1, 51 = 4.244, p = 0.045, ηp2 = 0.077) with no other effects (IBM SPSS Statistics 20). In Experiment 1, chicks’ looking behavior seemed to be in line with the prediction of longer looking times for unexpected outcomes (untransformed means: Unexpected Disappearance: M = 15.421; SD = 7.34; Expected Disappearance: M = 12.671, SD = 6.566; Scheffé test p = 0.078, η2 = 0.037); however, this was not the case for Experiment 2 (Unexpected Appearance: M = 12.906; SD = 8.409; Expected Appearance: M = 14.556, SD = 9.351; Scheffé test, p = 0.273, η2 = 0.008). Subjects looked longer to the Unexpected outcomes in Experiment 1 compared to Experiment 2 (Scheffé test, p = 0.028, d = 0.318). Human infants show similar looking patterns to such scenes (i.e., longer looks to violations of presence but not to that of absence; Wynn and Chiang, 2016; Kaufman et al., 2003), and it was suggested that they are more sensitive to violations of presence compared to violations regarding the absence of objects. Figure 2 Download asset Open asset Results of Experiments 1 and 2. (A) Mean Looking Times elicited by Expected and Unexpected outcomes in the two experiments. The asterisk indicates a significant interaction between Outcome (Expected/Unexpected) and Experiment (F1,51 = 4.244, p = 0.045) and a significantly higher looking time observed for the Unexpected outcome in Experiment 1 (Encoding Absence) than in Experiment 2 (Encoding Presence) (Scheffé test, p = 0.028). Error bars represent standard error of the mean. (B) Lateralization Index as a function of Experiment and types of outcome. Asterisks indicate a significant interaction between Outcome (Expected/Unexpected) and Experiment (F1,51 = 4.652, p = 0.036) and a significantly higher left-eye bias observed for the Unexpected outcome in Experiment 2 (Encoding Absence) than in Experiment 1 (Encoding Presence) (Scheffé test, p = 0.023), suggesting that the Lateralization Index is sensitive to violations of expectation regarding the absence of objects. For box plots, the horizontal line represents the median, yellow diamonds depict the mean values, box height depicts first and third quartiles, and vertical lines represent the 95th percentile. Dots represent the outliers of the data set. A 2 × 2 × 2 repeated measures ANOVA performed on the Lateralization Index revealed a significant Experiment by Outcome interaction (F1, 51 = 4.652, p = 0.036, ηp2 = 0.084), with no other significant effects. Pairwise comparisons revealed that in the Unexpected Outcome condition subjects displayed a greater left-eye bias in Experiment 2 (Encoding Absence, M = 0.193, SD = 0.517) compared to Experiment 1 (Encoding Presence, M = −0.138, SD = 0.459; Scheffé test, p = 0.023, d = 0.677). The other pairwise comparisons did not reach significance, though the pattern of eye usage was congruent with our predictions in Experiment 2 (greater left-eye bias for the unexpected outcomes: Unexpected Appearance, M = 0.193, SD = 0.518; Expected Appearance M = −0.054, SD = 0.518; Scheffé test, p = 0.091, η2 = 0.05), but not in Experiment 1. These data suggest that the Lateralization Index may be sensitive to violations of expectation regarding the absence of objects, but not regarding the presence of objects (Figure 2B). A positive Lateralization Index reflects a left-eye bias, which was previously linked to detecting novel outcomes (Rogers and Anson, 1979; Dharmaretnam and Andrew, 1994), however in our case it seems to be specific to violations of expectations about the object’s absence. Thus, while there is currently no report suggesting that human infants (Wynn and Chiang, 2016; Kaufman et al., 2003) or other animals would spontaneously encode absence, the Lateralization Index in the present study points to 8-day-old chicks’ ability to encode and form expectations about ‘no objects’ at a particular location. The left-eye bias, found in response to the unexpected appearance of the object, likely reflected the chicks’ exploration, and attempt of identification of the unexpectedly emerging ‘new’ item. Importantly, this item was new only if chicks had encoded that the identically looking familiar item left, and thus it was absent from the scene. They likely investigated the ‘new’ object more carefully with their left eye because they expected the old object to be absent. Interestingly, in contrast to the left-eye bias, overall looking time did not seem to be sensitive to the unexpected appearance of the object, and therefore to absence. We assumed that if chicks are sensitive to both presence and absence violations, they should show surprise (and produce longer looking times) in the Unexpected outcomes compared to the Expected outcomes in both experiments. Instead, we found an Experiment by Outcome interaction. Only Experiment 1 (Encoding Presence) revealed a looking time that patterned with our prediction, but not Experiment 2 (Encoding Absence). The present work was, however, to our knowledge, the first attempt to measure young chicks’ looking time in violation of expectation paradigms, thus it may be difficult to interpret this pattern. It may be the case that in our setting overall looking times did not reveal clear differences due to statistical variability. To sum up, in Study 1 we found asymmetric patterns of looking times and lateralization in response to outcomes congruent or violating chicks’ expectations of presence and absence. Furthermore, we found that the left-eye bias seems to reflect chicks’ ability to form an expectation about the absence of an object. In Study 2, we targeted the cognitive processes that could support the representation of absence. Study 2 In two additional experiments, we focused on the left-eye bias effect observed in Experiment 2. We aimed on the one hand to strengthen our findings regarding chicks’ absence encoding, and on the other hand to test the possible representational processes underlying such a behavior. According to one possibility, chicks in Experiment 2 (Encoding Absence) might have not formed an actual representation that there was no object behind the screen, instead their reaction to the unexpected appearance could have been perceptually driven and derived from detecting the mismatch between the memory of the empty space behind the screen (i.e., an iconic, picture-like representation of the empty floor, and the wall of the testing arena) and the perceived outcome (i.e., the scene with the object). This possibility, however, would predict no left-eye bias in a situation where there was no possibility for perceptually encoding the empty space. In contrast, left-eye bias without relying on the percept of an empty space would point to a more complex mental representation of absence that is inferred from the sequence of events (object going in/going out) (Experiment 3. Absence vs. Perceptual Comparison). A second alternative explanation might be that the chicks did not encode the absence of the object, but tracked the location of the target object even when it left the scene, and encoded its presence somewhere outside the arena, and the left-eye bias reflected their ‘surprise’ of seeing this object at an unexpected location (i.e., behind the screen, inside the arena). According to this alternative, an outcome which would feature a different object behind the occluder should not elicit a left-eye bias (Experiment 4. Absence vs. Tracking). In the second study, Experiments 3 and 4 test these alternative explanations, respectively, and aimed at replicating the findings from Experiment 2. Experiment 3 (Absence vs. Perceptual Comparison) followed the procedure of Experiment 2. However, unlike in Experiment 2, each trial started with the screen in upright position preventing the chicks (n = 31, 15 females, 16 males) to see the space occluded by the screen at the beginning of the trial (Figure 3A). This manipulation aimed at testing chicks’ ability to update their expectation about what is (not) behind the screen without giving them the opportunity to perform perceptual comparison between an initial empty scene and the outcome. Without clear perceptual evidence, we expect chicks to arrive at encoding absence by first representing the object being behind the screen, and then inferring the outcome (i.e., absence) after the object being removed from behind the screen. Based on the observation of these events chicks may compute the absence objects and show similar behavior (i.e., left-eye bias in response to the unexpected appearance of the object) as we observed in Experiment 2. Figure 3 Download asset Open asset Schematic illustration of the events in Experiments 3 and 4. (A) Experiment 3 – Absence vs. Perceptual Comparison. The upper panels depict the event in the Expected Appearance condition, where the target object moved behind the screen and when the screen was lowered, it revealed the object behind. The lower panels depict the events in the Unexpected Appearance condition, in which the object first moved behind the screen and then it was visibly removed from the arena. Afterwards, the object was secretly placed behind the screen and when the screen was lowered, it revealed the object. Note that, in both conditions, the screen’s initial position was vertical. (B) Experiment 4 – Tracking vs. Absence. The upper panels depict the events in the Expected Appearance condition, which was identical to the same condition of Experiment 3. The lower panels depict the Unexpected Appearance condition where the (green) target object first moved behind the screen and then it was visibly removed from the arena. Afterwards, the red object was secretly reintroduced behind the screen. In the outcome phase, the screen was lowered and the red object appeared. Experiment 4 (Absence vs. Tracking) (n = 22, 9 females, 13 males) was the same as Experiment 3, except that in the Unexpected Appearance condition the target object that moved behind the screen and then left the scene was different from the object that appeared when the screen was lowered in the outcome phase (Figure 3B). Finding a second object at the previously empty location should not be surprising if chicks simply track the location of the first target object. However, finding an object at a location that is represented as empty would lead to surprise even if this object is different from the one that has left. Thus a left-eye bias in the Unexpected Appearance condition of Experiment 4 would provide evidence for chicks’ capacity to form expectation about absence of objects at a specific location that must be therefore empty. In contrast, the lack of such response would rather point to object tracking processes underlying chicks’ left-eye bias in Experiment 2 that resulted in encoding the presence of the first object at a different location (outside the scene). Results We analyzed the Lateralization Index of Experiments 3 and 4 with a 2 × 2 × 2 repeated measures ANOVA with Experiment, Outcome, and Sex as factors. There was a main effect of Outcome (F1, 49 = 4.29, p = 0.044, ηp2 = 0.081), revealing more positive values (more usage of the left eye) in response to unexpected outcomes (M = 0.087, SD = 0.5) compared to expected outcomes (M = −0.045, SD = 0.459), similar to Experiment 2. Additionally, a significant interaction was observed between Outcome and Sex (F1, 49 = 4.804, p = 0.033, ηp2 = 0.089), indicating females’ sensitivity to events violating their expectations regarding absence of ent

Number line estimation (NLE) is an educational task in which children estimate the location of a value (e.g., 25) on a blank line that represents a numerical range (e.g., 0-100). NLE performance is a strong predictor of success in mathematics, and error patterns on this task help provide a glimpse into how children may represent number internally. However, a missing and fundamental element of this puzzle is the identification of neural correlates of NLE in children. That is, understanding possible neural signatures related to NLE performance will provide valuable insight into the cognitive processes that underlie children's development of NLE ability. Using functional near-infrared spectroscopy (fNIRS), we provide the first investigation of concurrent behavioral and cortical signatures of NLE performance in children. Specifically, our results highlight significant fronto-parietal changes in cortical activation in response to increases in NLE scale (e.g., 0-100 vs. 0-100,000). Furthermore, our results demonstrate that NLE performance feedback (auditory, visual, or audiovisual), as well as children's grade (2nd vs. 3rd) influence cortical responding during an NLE task.

Environments are unique in terms of structural composition and evoked human experience. Previous studies suggest that natural compared to built environments may increase positive emotions. Humans in natural environments also demonstrate greater performance on attention-based tasks. Few studies have investigated cortical mechanisms underlying these phenomena or probed these differences from a neural perspective. Using a temporally sensitive electrophysiological approach, we employ an event-related, implicit passive viewing task to demonstrate that in humans, a greater late positive potential (LPP) occurs with exposure to built than natural environments, resulting in a faster return of activation to pre-stimulus baseline levels when viewing natural environments. Our research thus provides new evidence suggesting natural environments are perceived differently from built environments, converging with previous behavioral findings and theoretical assumptions from environmental psychology.

This theoretical discussion provides insight into an intersect of the mathematics education, cognitive psychology, and special education fields. To examine this intersect, the authors focus on how students identified with a learning disability develop actions on material when constructing and coordinating units. This theoretical frame considers results from several case studies in special education and cognitive learning fields, focusing on young students’ number development, set in their subitizing activity and units construction/coordination. These results provide context and illustrate critical importance to their actions in light of neural differences and differences in their rate of development for future number and operation construction.

Research within psychology and other disciplines has shown that exposure to natural environments holds extensive physiological and psychological benefits. Adding to the health and cognitive benefits of natural environments, evidence suggests that exposure to nature also promotes healthy human decision-making. Unhealthy decision-making (e.g., smoking, non-medical prescription opioid misuse) and disorders associated with lack of impulse control [e.g., tobacco use, opioid use disorder (OUD)], contribute to millions of preventable deaths annually (i.e., 6 million people die each year of tobacco-related illness worldwide, deaths from opioids from 2002 to 2017 have more than quadrupled in the United States alone). Impulsive and unhealthy decision-making also contributes to many pressing environmental issues such as climate change. We recently demonstrated a causal link between visual exposure to nature (e.g., forests) and improved self-control (i.e., decreased impulsivity) in a laboratory setting, as well as the extent to which nearby nature and green space exposure improves self-control and health decisions in daily life outside of the experimental laboratory. Determining the benefits of nearby nature for self-controlled decision-making holds theoretical and applied implications for the design of our surrounding environments. In this article, we synergize the overarching results of recent research endeavors in three domains including the effects of nature exposure on (1) general health-related decision-making, (2) health and decision-making relevant for application to addiction related processes (e.g., OUD), and (3) environmentally relevant decision-making. We also discuss key future directions and conclusions.

Prominent theories suggest that time and number are processed by a single neural locus or a common magnitude system (e.g., Meck and Church, 1983; Walsh, 2003). However, a growing body of literature has identified numerous inconsistencies between temporal and numerical processing, casting doubt on the presence of such a singular system. Findings of distinct temporal and numerical biases in the presence of emotional content (Baker et al., 2013; Young and Cordes, 2013) are particularly relevant to this debate. Specifically, emotional stimuli lead to temporal overestimation, yet identical stimuli result in numerical underestimation. In the current study, we tested adults' temporal and numerical processing under cognitive load, a task that compromises attention. Under the premise of a common magnitude system, one would predict cognitive load to have an identical impact on temporal and numerical judgments. Inconsistent with the common magnitude account, results revealed baseline performance on the temporal and numerical task was not correlated and importantly, cognitive load resulted in distinct and opposing quantity biases: numerical underestimation and marginal temporal overestimation. Together, our data call into question the common magnitude account, while also providing support for the role of attentional processes involved in numerical underestimation.

Mathematics education researchers are tasked with solving practical research problems involving complex constructs in complex settings. The effective integration of quantitative and qualitative data allows researchers to draw more nuanced conclusions about these complex phenomena. This article describes the use of a convergent parallel mixed methods design to integrate two seemingly conflicting data sources that measured six second-grade students' development of computational fluency. The mixed methods analysis of students' computational fluency assessments and interviews showed that there was variation in students' assessment scores, strategy use, and engagement of number sense. Within these variations, the quantitative and qualitative data converged or diverged at various measurement points, and the results highlight the importance of merging the two data sets to capture a richer picture of students' computational fluency. Implications for using mixed methods in understanding how mathematics learning occurs in classrooms are discussed.

Current research shows that digital games can significantly enhance children's learning. The purpose of this study was to examine how design features in 12 digital math games influenced children's learning. The participants in this study were 193 children in Grades 2 through 6 (ages 8–12). During clinical interviews, children in the study completed pre-tests, interacted with digital math games, responded to questions about the digital math games, and completed post-tests. We recorded the interactions using two video perspectives that recorded children's gameplay and responses to interviewers. We employed mixed methods to analyze the data and identify salient patterns in children's experiences with the digital math games. The analysis revealed significant gains for 9 of the 12 digital games and most children were aware of the design features in the games. There were eight prominent categories of design features in the video data that supported learning and mathematics connections. Six categories focused on how the design features supported learning in the digital games. These categories included: accuracy feedback, unlimited/multiple attempts, information tutorials and hints, focused constraint, progressive levels, and game efficiency. Two categories were more specific to embodied cognition and action with the mathematics, and focused on how design features promoted mathematics connections. These categories included: linked representations and linked physical actions. The digital games in this study that did not include linked representations and opportunities for linked physical actions as design features did not produce significant gains. These results suggest the key role of mathematics-specific design features in the design of digital math games.

The detrimental health effects of exposure to air pollution are well established. Fostering behavioral change concerning air quality may be challenging because the detrimental health effects of exposure to air pollution are delayed. Delay discounting, a measure of impulsive choice, encapsulates this process of choosing between the immediate conveniences of behaviors that increase pollution and the delayed consequences of prolonged exposure to poor air quality. In Experiment 1, participants completed a series of delay-discounting tasks for air quality and money. We found that participants discounted delayed air quality more than money. In Experiment 2, we investigated whether the common finding that large amounts of money are discounted less steeply than small amounts of money generalized to larger and smaller improvements in air quality. Participants discounted larger improvements in air quality less steeply than smaller improvements, indicating that the discounting of air quality shares a similar process as the discounting of money. Our results indicate that the discounting of delayed money is strongly related to the discounting of delayed air quality and that similar mechanisms may be involved in the discounting of these qualitatively different outcomes. These data are also the first to demonstrate the malleability of delay discounting of air quality, and provide important public health implications for decreasing delay discounting of air quality.

This paper focuses on understanding the role that affordances played in children’s learning performance and efficiency during clinical interviews of their interactions with mathematics apps on touch-screen devices. One hundred children, ages 3 to 8, each used six different virtual manipulative mathematics apps during 30–40-min interviews. The study used a convergent mixed methods design, in which quantitative and qualitative data were collected concurrently to answer the research questions (Creswell and Plano Clark 2011). Videos were used to capture each child’s interactions with the virtual manipulative mathematics apps, document learning performance and efficiency, and record children’s interactions with the affordances within the apps. Quantitized video data answered the research question on differences in children’s learning performance and efficiency between pre- and post-assessments. A Wilcoxon matched pairs signed-rank test was used to explore these data. Qualitative video data was used to identify affordance access by children when using each app, identifying 95 potential helping and hindering affordances among the 18 apps. The results showed that there were changes in children’s learning performance and efficiency when children accessed a helping or a hindering affordance. Helping affordances were more likely to be accessed by children who progressed between the pre- and post-assessments, and the same affordances had helping and hindering effects for different children. These results have important implications for the design of virtual manipulative mathematics learning apps.