A meta-analysis of sex differences in animal personality: no evidence for the greater male variability hypothesis.

The notion that men are more variable than women has become embedded into scientific thinking. For mental traits like personality, greater male variability has been partly attributed to biology, underpinned by claims that there is generally greater variation among males than females in non-human animals due to stronger sexual selection on males. However, evidence for greater male variability is limited to morphological traits, and there is little information regarding sex differences in personality-like behaviours for non-human animals. Here, we meta-analysed sex differences in means and variances for over 2100 effects (204 studies) from 220 species (covering five broad taxonomic groups) across five personality traits: boldness, aggression, activity, sociality and exploration. We also tested if sexual size dimorphism, a proxy for sex-specific sexual selection, explains variation in the magnitude of sex differences in personality. We found no significant differences in personality between the sexes. In addition, sexual size dimorphism did not explain variation in the magnitude of the observed sex differences in the mean or variance in personality for any taxonomic group. In sum, we find no evidence for widespread sex differences in variability in non-human animal personality.

A framework for studying behavioral evolution by reconstructing ancestral repertoires.

Although different animal species often exhibit extensive variation in many behaviors, typically scientists examine one or a small number of behaviors in any single study. Here, we propose a new framework to simultaneously study the evolution of many behaviors. We measured the behavioral repertoire of individuals from six species of fruit flies using unsupervised techniques and identified all stereotyped movements exhibited by each species. We then fit a Generalized Linear Mixed Model to estimate the intra- and inter-species behavioral covariances, and, by using the known phylogenetic relationships among species, we estimated the (unobserved) behaviors exhibited by ancestral species. We found that much of intra-specific behavioral variation has a similar covariance structure to previously described long-time scale variation in an individual's behavior, suggesting that much of the measured variation between individuals of a single species in our assay reflects differences in the status of neural networks, rather than genetic or developmental differences between individuals. We then propose a method to identify groups of behaviors that appear to have evolved in a correlated manner, illustrating how sets of behaviors, rather than individual behaviors, likely evolved. Our approach provides a new framework for identifying co-evolving behaviors and may provide new opportunities to study the mechanistic basis of behavioral evolution.

Multi-View Mouse Social Behaviour Recognition With Deep Graphic Model.

Home-cage social behaviour analysis of mice is an invaluable tool to assess therapeutic efficacy of neurodegenerative diseases. Despite tremendous efforts made within the research community, single-camera video recordings are mainly used for such analysis. Because of the potential to create rich descriptions for mouse social behaviors, the use of multi-view video recordings for rodent observations is increasingly receiving much attention. However, identifying social behaviours from various views is still challenging due to the lack of correspondence across data sources. To address this problem, we here propose a novel multi-view latent-attention and dynamic discriminative model that jointly learns view-specific and view-shared sub-structures, where the former captures unique dynamics of each view whilst the latter encodes the interaction between the views. Furthermore, a novel multi-view latent-attention variational autoencoder model is introduced in learning the acquired features, enabling us to learn discriminative features in each view. Experimental results on the standard CRMI13 and our multi-view Parkinson's Disease Mouse Behaviour (PDMB) datasets demonstrate that our proposed model outperforms the other state of the arts technologies, has lower computational cost than the other graphical models and effectively deals with the imbalanced data problem.

Protocol for behavioral tests using chemogenetically manipulated mice.

Behavioral analyses using mice chemogenetically manipulated by designer receptors exclusively activated by designer drugs (DREADDs) are powerful tools to elucidate neural functions. Here, we describe the detailed protocols for stereotaxic surgery, adeno-associated virus (AAV)-mediated introduction to Gq-DREADDs in mice, and for behavioral testing and analyses related to anxiety, risk assessment, and burying behaviors. A series of these tests are useful in evaluating animal anxiety and their defensive response patterns to potential threats. For complete details on the use and execution of this protocol, please refer to Horii-Hayashi et al. (2021).

Protocol for a Binary Choice Feeding Assay Using Adult, Axenic Drosophila.

This protocol is designed to prepare adult axenic Drosophila before monitoring their behavior in a two-choice feeding assay, where flies are confronted with an axenic versus a dead or alive bacteria-contaminated feeding solution. Several aspects of the procedure, including raising and aging flies in axenic conditions, starving adult flies, and composing feeding solutions, are detailed. The bacterium used in this protocol, Erwinia carotovora carotovora-15 2141 (Ecc-15 2141 ), is commonly used to decipher the mechanisms controlling host-pathogen interactions in the Drosophila model. For complete details on the use and execution of this protocol, please refer to Charroux et al. (2020).

Automated measurement of long-term bower behaviors in Lake Malawi cichlids using depth sensing and action recognition.

In the wild, behaviors are often expressed over long time periods in complex and dynamic environments, and many behaviors include direct interaction with the environment itself. However, measuring behavior in naturalistic settings is difficult, and this has limited progress in understanding the mechanisms underlying many naturally evolved behaviors that are critical for survival and reproduction. Here we describe an automated system for measuring long-term bower construction behaviors in Lake Malawi cichlid fishes, in which males use their mouths to sculpt sand into large species-specific structures for courtship and mating. We integrate two orthogonal methods, depth sensing and action recognition, to simultaneously track the developing bower structure and the thousands of individual sand manipulation behaviors performed throughout construction. By registering these two data streams, we show that behaviors can be topographically mapped onto a dynamic 3D sand surface through time. The system runs reliably in multiple species, across many aquariums simultaneously, and for up to weeks at a time. Using this system, we show strong differences in construction behavior and bower form that reflect species differences in nature, and we gain new insights into spatial, temporal, social dimensions of bower construction, feeding, and quivering behaviors. Taken together, our work highlights how low-cost tools can automatically quantify behavior in naturalistic and social environments over long timescales in the lab.

Pain behaviors before and after treatment of oral disease in cats using video assessment: a prospective, blinded, randomized clinical trial.

BACKGROUND: Specific behaviors associated with pain in cats with oral disease have not been consistently studied. The aim of this exploratory study was to identify pain-induced behaviors in cats before and after treatment of oral disease using video assessment. Twenty-four cats (6 ± 3.3 years old; 4.9 ± 1.7 kg) were included in a prospective, blinded, randomized clinical trial. Cats were equally divided into minimal (G1: minimal dental treatment) or severe (G2: multiple dental extractions) oral disease groups. After acclimation at day 0, they underwent oral examination, radiographs, scaling, and dental extractions under general anesthesia (anesthetic protocol: acepromazine, hydromorphone, propofol, isoflurane, meloxicam, and local anesthetic blocks; day 1), and were discharged at day 6. Cats were filmed remotely for 10 min using a wide-angle glass lens camera before surgery (baseline) and throughout the study at different time points (36 h of video recording). The videos consisted of four parts namely general, playing, feeding and post-feeding behaviors. A board-certified behaviorist evaluated the duration/frequency of different behaviors based on an ethogram, which were analyzed using linear mixed models and a generalized linear model, respectively (p < 0.05). RESULTS: In comparison with baseline, duration of "not pawing the face" was significantly shorter at day 3 in G2. These cats spent significantly longer time "standing" and "laying" at days 3 and 6, respectively; G1 spent significantly less time "walking" and "standing" at days 3 and 4, respectively and significantly longer time "immobile" at day 3. Duration of "no/slow tail movement" was significantly longer in G2 than G1 at day 5. Duration of "pawing the ribbon" (playing) was significantly shorter in G2 than G1 at day 1. Feeding and post-feeding behaviors with soft food were not significantly different between groups or over time. Frequency of "difficulty grasping dry food" was significantly higher in G2 than G1 up to day 6. Frequency of post-feeding "head shaking" was significantly higher in both groups at day 6 when compared with baseline. CONCLUSIONS: This study identified pain-induced behaviors in cats undergoing treatment of oral disease. These behaviors may be used to differentiate painful versus pain-free cats in clinical practice.

Establishment of brain ischemia model in tree shrew.

BACKGROUND: Tree shrew, as a kind of small and inexpensive animal between insectivores and primates with the general anatomy being similar to human, could be considered as developed animal model for brain ischemia (BI) study. However, there is no neural behavior scores criterion from tree shrew with BI up to now. METHODS: To produce BI model of tree shrew, a novel systematic neurobehavioral assessment scale, named as neural behavior scores (NBS) including aggressive behavior, seeking behavior, gait, startle reflex, high jump and warped-tail phenomenon was firstly established and used in this study. Moreover, magnetic resonance imaging (MRI) was performed on the first day after the operation to detect the imaging changes caused by ischemia. Then TTC, HE staining and immunofluorescent staining for GFAP and NeuN, were performed 24 h after surgery respectively. RESULTS: NBS in BI group were significantly higher than that of sham operation group at 1d, 3d, 5d and 7d after ischemia. Meanwhile, compared with the sham operation group, the T2 images demonstrated significant higher signal and local brain swelling after cerebral ischemia in tree shrews. The staining of TTC and HE showed apparent infarction and necrosis of the cerebral region, and most of neurons exhibited a shrink. CONCLUSION: We have successfully established the BI model of tree shrew, confirmed by NBS (a new developed method), MRI, HE staining, TTC staining and immunofluorescence staining. It is the first time to report a novel neurobehavioral assessment scale for BI in tree shrew.

Classifying grey seal behaviour in relation to environmental variability and commercial fishing activity - a multivariate hidden Markov model.

Classifying movement behaviour of marine predators in relation to anthropogenic activity and environmental conditions is important to guide marine conservation. We studied the relationship between grey seal (Halichoerus grypus) behaviour and environmental variability in the southwestern Baltic Sea where seal-fishery conflicts are increasing. We used multiple environmental covariates and proximity to active fishing nets within a multivariate hidden Markov model (HMM) to quantify changes in movement behaviour of grey seals while at sea. Dive depth, dive duration, surface duration, horizontal displacement, and turning angle were used to identify travelling, resting and foraging states. The likelihood of seals foraging increased in deeper, colder, more saline waters, which are sites with increased primary productivity and possibly prey densities. Proximity to active fishing net also had a pronounced effect on state occupancy. The probability of seals foraging was highest <5 km from active fishing nets (51%) and decreased as distance to nets increased. However, seals used sites <5 km from active fishing nets only 3% of their time at sea highlighting an important temporal dimension in seal-fishery interactions. By coupling high-resolution oceanographic, fisheries, and grey seal movement data, our study provides a scientific basis for designing management strategies that satisfy ecological and socioeconomic demands on marine ecosystems.

Three-Dimensional Pose Estimation for Laboratory Mouse From Monocular Images.

Video-based activity and behavior analysis of mice has garnered wide attention in biomedical research. Animal facilities hold large numbers of mice housed in "home-cages" densely stored within ventilated racks. Automated analysis of mice activity in their home-cages can provide a new set of sensitive measures for detecting abnormalities and time-resolved deviation from the baseline behavior. Large-scale monitoring in animal facilities requires minimal footprint hardware that integrates seamlessly with the ventilated racks. The compactness of hardware imposes the use of fisheye lenses positioned in close proximity to the cage. In this paper, we propose a systematic approach to accurately estimate the 3D pose of the mouse from single-monocular fisheye-distorted images. Our approach employs a novel adaptation of a structured forest algorithm. We benchmark our algorithm against existing methods. We demonstrate the utility of the pose estimates in predicting mouse behavior in a continuous video.

Behavioral effects evoked by the beta globin-derived nonapeptide LVV-H6.

LVV-hemorphin-6 (LVV-h6) is bioactive peptide and is a product of the degradation of hemoglobin. Since LVV-h6 effects are possibly mediated by opioid or AT4/IRAP receptors, we hypothesized that LVV-h6 would modify behavior. We evaluated whether LVV-h6 affects: i) anxiety-like behavior and locomotion; ii) depression-like behavior; iii) cardiovascular and neuroendocrine reactivity to emotional stress. Male Wistar rats ( ± 300 g) received LVV-h6 (153 nmol/kg i.p.) or vehicle (NaCl 0.9% i.p.). We used: i) open field (OF) test for locomotion; ii) elevated plus maze (EPM) for anxiety-like behavior; iii) forced swimming test (FST) for depression-like behavior and iv) air jet for cardiovascular and neuroendocrine reactivity to stress. Diazepam (2 mg/kg i.p.) and imipramine (15 mg/kg i.p.) were used as positive control for EPM and FST, respectively. To evaluate the LVV-h6 mechanisms, we used: the antagonist of oxytocin (OT) receptors (atosiban - ATS 1 and 0.1 mg/kg i.p.); the inhibitor of tyrosine hydroxylase (Alpha-methyl-p-tyrosine - AMPT 200 mg/kg i.p.) to investigate the involvement of catecholaminergic paths; and the antagonist of opioid receptors (naltrexone - NTX 0.3 mg/kg s.c.). We found that LVV-h6: i) evoked anxiolytic-like effect; ii) evoked antidepressant-like effect in the FST; and iii) did not change the locomotion, neuroendocrine and cardiovascular responses to stress. The LVV-h6 anxiolytic-like effect was not reverted by ATS and AMPT. However, the antidepressant effects were reverted only by NTX. Hence, our findings demonstrate that LVV-h6 modulates anxiety-like behavior by routes that are not oxytocinergic, catecholaminergic or opioid. The antidepressant-like effects of LVV-h6 rely on opioid pathways.

Dairy cattle behavior classifications based on decision tree learning using 3-axis neck-mounted accelerometers.

Demand has been increasing recently for an automated monitoring system of animal behavior as a tool for the management of livestock animals. This study investigated the association between the behavior of dairy cattle and the acceleration data collected using three-axis neck-mounted accelerometers, as well as the feasibility of improving the precision of behavior classifications through machine learning. In total 38 Holstein dairy cows were used, and kept in four different farms. A logger was mounted to each collar to obtain acceleration data for calculating the activity level and variations. At the same time the behavior of the cattle was observed visually. Characteristic acceleration waves were recorded for eating, rumination, and lying, respectively; and the activity level and variations were significantly different among these behaviors (p < 0.01). Decision tree learning was performed on the data set from Farm A and validated its precision; which proved to be 99.2% in cross-validation, and 100% in test data sets from Farms B to D. This study showed that highly precise classifications for eating, rumination, and lying is possible by using decision tree learning to calculate the activity level and variations of cattle based on the data obtained by three-axis accelerometers mounted to a collar.

Resource-Efficient Pet Dog Sound Events Classification Using LSTM-FCN Based on Time-Series Data.

The use of IoT (Internet of Things) technology for the management of pet dogs left alone at home is increasing. This includes tasks such as automatic feeding, operation of play equipment, and location detection. Classification of the vocalizations of pet dogs using information from a sound sensor is an important method to analyze the behavior or emotions of dogs that are left alone. These sounds should be acquired by attaching the IoT sound sensor to the dog, and then classifying the sound events (e.g., barking, growling, howling, and whining). However, sound sensors tend to transmit large amounts of data and consume considerable amounts of power, which presents issues in the case of resource-constrained IoT sensor devices. In this paper, we propose a way to classify pet dog sound events and improve resource efficiency without significant degradation of accuracy. To achieve this, we only acquire the intensity data of sounds by using a relatively resource-efficient noise sensor. This presents issues as well, since it is difficult to achieve sufficient classification accuracy using only intensity data due to the loss of information from the sound events. To address this problem and avoid significant degradation of classification accuracy, we apply long short-term memory-fully convolutional network (LSTM-FCN), which is a deep learning method, to analyze time-series data, and exploit bicubic interpolation. Based on experimental results, the proposed method based on noise sensors (i.e., Shapelet and LSTM-FCN for time-series) was found to improve energy efficiency by 10 times without significant degradation of accuracy compared to typical methods based on sound sensors (i.e., mel-frequency cepstrum coefficient (MFCC), spectrogram, and mel-spectrum for feature extraction, and support vector machine (SVM) and k-nearest neighbor (K-NN) for classification).

An automated approach to the quantitation of vocalizations and vocal learning in the songbird.

Studies of learning mechanisms critically depend on the ability to accurately assess learning outcomes. This assessment can be impeded by the often complex, multidimensional nature of behavior. We present a novel, automated approach to evaluating imitative learning. Conceptually, our approach estimates how much of the content present in a reference behavior is absent from the learned behavior. We validate our approach through examination of songbird vocalizations, complex learned behaviors the study of which has provided many insights into sensory-motor learning in general and vocal learning in particular. Historically, learning has been holistically assessed by human inspection or through comparison of specific song features selected by experimenters (e.g. fundamental frequency, spectral entropy). In contrast, our approach uses statistical models to broadly capture the structure of each song, and then estimates the divergence between the two models. We show that our measure of song learning (the Kullback-Leibler divergence between two distributions corresponding to specific song data, or, Song DKL) is well correlated with human evaluation of song learning. We then expand the analysis beyond learning and show that Song DKL also detects the typical song deterioration that occurs following deafening. Finally, we illustrate how this measure can be extended to quantify differences in other complex behaviors such as human speech and handwriting. This approach potentially provides a framework for assessing learning across a broad range of behaviors like song that can be described as a set of discrete and repeated motor actions.

Reevaluating cases of trait-dependent diversification in primates.

OBJECTIVES: Previous investigations have identified several ecological traits that may have shaped differences in species richness among primate clades by influencing speciation and extinction probabilities. In this study, I reevaluate these cases in light of concerns about high false-positive rates of the methods used to detect trait-dependent diversification. MATERIALS AND METHODS: Data for six traits were taken from the literature: activity pattern, sociality, mating system, terrestriality, frugivory, and mutualistic interactions with angiosperms (i.e., seed dispersal and pollination). These traits were treated as binary variables and analyzed in a phylogenetic framework using likelihood-based and nonparametric methods for estimating trait-specific diversification rates. RESULTS: Of the traits analyzed here, only activity pattern was found by all methods to be correlated with diversification: diurnal primates diversify at higher rates than nocturnal primates. Solitary foragers have a lower rate of diversification than group-living primates, but this link is driven by the association between solitariness and nocturnality. DISCUSSION: These results indicate that ecological traits have had a more limited influence on primate diversification than suggested by previous studies. The false positives detected here can be linked with heterogeneity in diversification caused by sources other than the focal traits. One of these sources is the very high rate of diversification in Cercopithecoidea found in previous investigations. Another is the correlation between activity patterns and diversification. The underlying causal mechanism for the contrast in diversification dynamics between diurnal and nocturnal primates is unclear, but this correlation highlights the important role of activity pattern in shaping primate evolutionary history.

Classifying elephant behaviour through seismic vibrations.

Seismic waves - vibrations within and along the Earth's surface - are ubiquitous sources of information. During propagation, physical factors can obscure information transfer via vibrations and influence propagation range [1]. Here, we explore how terrain type and background seismic noise influence the propagation of seismic vibrations generated by African elephants. In Kenya, we recorded the ground-based vibrations of different wild elephant behaviours, such as locomotion and infrasonic vocalisations [2], as well as natural and anthropogenic seismic noise. We employed techniques from seismology to transform the geophone recordings into source functions - the time-varying seismic signature generated at the source. We used computer modelling to constrain the propagation ranges of elephant seismic vibrations for different terrains and noise levels. Behaviours that generate a high force on a sandy terrain with low noise propagate the furthest, over the kilometre scale. Our modelling also predicts that specific elephant behaviours can be distinguished and monitored over a range of propagation distances and noise levels. We conclude that seismic cues have considerable potential for both behavioural classification and remote monitoring of wildlife. In particular, classifying the seismic signatures of specific behaviours of large mammals remotely in real time, such as elephant running, could inform on poaching threats.

Evidence-based severity assessment: Impact of repeated versus single open-field testing on welfare in C57BL/6J mice.

According to current guidelines on animal experiments, a prospective assessment of the severity of each procedure is mandatory. However, so far, the classification of procedures into different severity categories mainly relies on theoretic considerations, since it is not entirely clear which of the various procedures compromise the welfare of animals, or, to what extent. Against this background, a systematic empirical investigation of the impact of each procedure, including behavioral testing, seems essential. Therefore, the present study was designed to elucidate the effects of repeated versus single testing on mouse welfare, using one of the most commonly used paradigms for behavioral phenotyping in behavioral neuroscience, the open-field test. In an independent groups design, laboratory mice (Mus musculus f. domestica) experienced either repeated, single, or no open-field testing - procedures that are assigned to different severity categories. Interestingly, testing experiences did not affect fecal corticosterone metabolites, body weights, elevated plus-maze or home cage behavior differentially. Thus, with respect to the assessed endocrinological, physical, and behavioral outcome measures, no signs of compromised welfare could be detected in mice that were tested in the open-field repeatedly, once, or, not at all. These findings challenge current classification guidelines and may, furthermore, stimulate systematic research on the severity of single procedures involving living animals.

When and Where Learning is Taking Place: Multisynaptic Changes in Strength During Different Behaviors Related to the Acquisition of an Operant Conditioning Task by Behaving Rats.

Although it is generally assumed that brain circuits are modified by new experiences, the question of which changes in synaptic efficacy take place in cortical and subcortical circuits across the learning process remains unanswered. Rats were trained in the acquisition of an operant conditioning in a Skinner box provided with light beams to detect animals' approaches to lever and feeder. Behaviors such as pressing the lever, eating, exploring, and grooming were also recorded. Animals were chronically implanted with stimulating and recording electrodes in hippocampal, prefrontal, and subcortical sites relevant to the task. Field synaptic potentials were evoked during the performance of the above-mentioned behaviors and before, during, and after the acquisition process. Afferent pathways to the hippocampus and the intrinsic hippocampal circuit were slightly modified in synaptic strength during the performance of those behaviors. In contrast, afferent and efferent circuits of the medial prefrontal cortex were significantly modified in synaptic strength across training sessions, mostly at the moment of the largest change in the learning curve. Performance of behaviors nondirectly related to the acquisition process (exploring, grooming) also evoked changes in synaptic strength across training. This study helps to understand when and where learning is being engraved in the brain.

Bioacústica como método de avaliação do comportamento em pastejo de novilhas Girolando / Bioacoustics as a tool for evaluating grazing behavior of Girolando heifers

Objetivou-se validar o método de análise de dados bioacústicos na descrição do comportamento em pastejo de novilhas em área de integração lavoura-pecuária (iLP) e floresta (iLPF). Foram utilizadas oito novilhas Girolando (¾Holandês x »Gir) com média de 25±6,8 meses de idade e 268±83kg de peso vivo (PV), distribuídas em delineamento crossover 2x2 (dois sistemas x dois períodos de avaliação). A coleta de dados para análise bioacústica foi realizada utilizando-se gravadores MP3. Esses dados foram analisados pelo software Audacity® para identificação das atividades de pastejo, ruminação e ócio. Concomitantemente, o comportamento também foi avaliado pelo método de observação visual, realizado no período de 8 até 16 horas, em intervalos de 15 minutos. Não houve diferença (P>0,05) entre médias dos tempos (minutos) de pastejo, ruminação e ócio registrados pelos métodos de bioacústica e avaliação visual, sendo, respectivamente, 127 vs. 137; 122 vs. 113 e 238 vs. 216, no sistema iLP, e 134 vs. 103; 130 vs. 165 e 233 vs. 203, no sistema iLPF. Por meio desses resultados, foi possível concluir que o método da bioacústica é uma importante ferramenta para avaliar o comportamento de pastejo em diferentes sistemas de produção a pasto, com a vantagem de possibilitar o registro de atividades durante o período noturno, além de maior praticidade e precisão do que o método observacional.(AU)

Aiming to validate the analysis of bioacoustic data for describing grazing behavior of heifers in integrated crop, livestock (ICL) and forestry (ICLF) systems. Eight 25±6.8 month-old Girolando (¾Holstein × »Gir) heifers with 268±83kg of live weight (LW) were distributed in 2x2 crossover design (two systems x two evaluation periods). Data collection for bioacoustic analysis was done with MP3 recorders. These data were analyzed by Audacity® software for identification of the activities of grazing, rumination, and resting. Concomitantly, behavior was also analyzed by the method of visual observation from 8AM to 4PM in intervals of 15 minutes. There was no difference (P>0.05) between the methods of bioacustic and visual observation considering the means of time (minutes) of grazing, rumination, and resting, respectively: 127 vs. 137; 122 vs. 113; and 238 vs. 216 in ICL system and; 134 vs. 103; 130 vs. 165 and 233 vs. 203 in ICLF system. In conclusion, the bioacustic method is a tool that can be useful for evaluating bovine grazing behavior within different grazing systems, with the advantage of allowing report of activities during the night with higher practicality and precision than the visual observation method.(AU)

What is a gesture? A meaning-based approach to defining gestural repertoires.

Current systems of categorising ape gestures are typically subjective, relying on human intuition. We have systematised the features on which categorization depends (movement; body part; one/both limbs; use of detached object; rhythmic repetition; contact with recipient), showing that a potential repertoire of over 1000 gestures is physically possible, as large as the lexicon of some languages. In contrast, little more than a tenth of these gestures is used in chimpanzee communication. The striking overlaps in repertoire found between populations and even species of great ape are evidently not a result of a restricted set of possible gestures. Using the reactions of signallers to identify which gestures are intended to be different by the apes themselves, we revised the current classification, making some new distinctions and abolishing others previously considered important, giving a final repertoire of 81. A small number of gestures are used deictically, such that the recipient must pay attention to specific locations to satisfy the signaller; raising the possibility of a stepping-stone to the evolution of reference.