Inferring spatial memory and spatiotemporal scaling from GPS data: comparing red deer Cervus elaphus movements with simulation models

Origin stories: how does learned migratory behaviour arise in populations?

Although decades of research have deepened our understanding of the proximate triggers and ultimate drivers of migrations for a range of taxa, how populations establish migrations remains a mystery. However, recent studies have begun to illuminate the interplay between genetically inherited and learned migrations, opening the door to the evaluation of how migration may be learned, established, and maintained. Nevertheless, for migratory species where the role of learning is evident, we lack a comprehensive framework for understanding how populations learn specific routes and refine migratory movements over time (i.e., their origins). This review draws on advances in behavioural and movement ecology to offer a comprehensive framework for how populations could transition from resident to migratory by connecting cognitive research on fine-scale perceptual cues and movement decisions with literature on learning and cultural transmission, to the emergent pattern of migration. We synthesize the multiple cognitive mechanisms and processes that allow a population to respond to seasonal resource limitation, then encode spatial and environmental information about resource availability in memory and engage in social learning to navigate their landscapes and track resources better. A rise in global reintroduction efforts, along with human-induced rapid shifts in environmental cues and changing landscapes make evaluating the origins of this threatened behaviour more urgent than ever.

Identifying signals of memory from observations of animal movements

Incorporating memory (i.e., some notion of familiarity or experience with the landscape) into models of animal movement is a rising challenge in the field of movement ecology. The recent proliferation of new methods offers new opportunities to understand how memory influences movement. However, there are no clear guidelines for practitioners wishing to parameterize the effects of memory on moving animals. We review approaches for incorporating memory into step-selection analyses (SSAs), a frequently used movement modeling framework. Memory-informed SSAs can be constructed by including spatial-temporal covariates (or maps) that define some aspect of familiarity (e.g., whether, how often, or how long ago the animal visited different spatial locations) derived from long-term telemetry data. We demonstrate how various familiarity covariates can be included in SSAs using a series of coded examples in which we fit models to wildlife tracking data from a wide range of taxa. We discuss how these different approaches can be used to address questions related to whether and how animals use information from past experiences to inform their future movements. We also highlight challenges and decisions that the user must make when applying these methods to their tracking data. By reviewing different approaches and providing code templates for their implementation, we hope to inspire practitioners to investigate further the importance of memory in animal movements using wildlife tracking data.

Locating Nesting Sites for Critically Endangered Galápagos Pink Land Iguanas (Conolophus marthae)

Invasive alien species control is recognized worldwide as a priority action to preserve global biodiversity. However, a lack of general life history knowledge for threatened species can impede the effectiveness of conservation actions. Galápagos pink land iguanas (Conolophus marthae) are endemic to Wolf Volcano, Galápagos, Ecuador. These iguanas are threatened by invasive alien species, particularly feral cats, that may affect their small population size. To guarantee the long-term survival of C. marthae, the Galápagos National Park Directorate is considering, along with an ongoing campaign of feral cat control, the implementation of a head-start program. However, the success of this management strategy necessarily relies on the identification of pink iguana nesting grounds, which were still unknown at the onset of this study. We modeled the movement patterns of male and female iguanas during the reproductive season, using location data collected from custom-made remote tracking devices installed on adult pink iguanas in April 2021. We first calculated for each individual the vector of distances from its starting location, which was defined as net displacement. We then used net displacement as the response variable in a generalized additive mixed model with day of the year as the predictor. Based on the hypothesis that males and females may behaviorally differ after mating, we looked for female-specific migratory behavior suggesting females were moving toward nesting areas. The results obtained confirmed our hypothesis, as females exhibited a distinct migratory behavior, reaching a small plateau area inside of Wolf Volcano's caldera and ca. 400 m below the volcano's northern rim. Moreover, once inside the caldera, females displayed a more aggregated distribution pattern. The movement data obtained allowed Galápagos National Park rangers to locate individual pink iguana nests and subsequently to sight and collect the first observed hatchlings of the species. This work constitutes a necessary baseline to perform dedicated studies of pink iguana nests and emerging hatchling iguanas, which is an essential step toward the development of an effective head-start program.

Contextual memory engrams, and the neuromodulatory influence of the locus coeruleus

Here, we review the basis of contextual memory at a conceptual and cellular level. We begin with an overview of the philosophical foundations of traversing space, followed by theories covering the material bases of contextual representations in the hippocampus (engrams), exploring functional characteristics of the cells and subfields within. Next, we explore various methodological approaches for investigating contextual memory engrams, emphasizing plasticity mechanisms. This leads us to discuss the role of neuromodulatory inputs in governing these dynamic changes. We then outline a recent hypothesis involving noradrenergic and dopaminergic projections from the locus coeruleus (LC) to different subregions of the hippocampus, in sculpting contextual representations, giving a brief description of the neuroanatomical and physiological properties of the LC. Finally, we examine how activity in the LC influences contextual memory processes through synaptic plasticity mechanisms to alter hippocampal engrams. Overall, we find that phasic activation of the LC plays an important role in promoting new learning and altering mnemonic processes at the behavioral and cellular level through the neuromodulatory influence of NE/DA in the hippocampus. These findings may provide insight into mechanisms of hippocampal remapping and memory updating, memory processes that are potentially dysregulated in certain psychiatric and neurodegenerative disorders.

Spatial memory predicts home range size and predation risk in pheasants

Most animals confine their activities to a discrete home range, long assumed to reflect the fitness benefits of obtaining spatial knowledge about the landscape. However, few empirical studies have linked spatial memory to home range development or determined how selection operates on spatial memory via the latter's role in mediating space use. We assayed the cognitive ability of juvenile pheasants (Phasianus colchicus) reared under identical conditions before releasing them into the wild. Then, we used high-throughput tracking to record their movements as they developed their home ranges, and determined the location, timing and cause of mortality events. Individuals with greater spatial reference memory developed larger home ranges. Mortality risk from predators was highest at the periphery of an individual's home range in areas where they had less experience and opportunity to obtain spatial information. Predation risk was lower in individuals with greater spatial memory and larger core home ranges, suggesting selection may operate on spatial memory by increasing the ability to learn about predation risk across the landscape. Our results reveal that spatial memory, determined from abstract cognitive assays, shapes home range development and variation, and suggests predation risk selects for spatial memory via experience-dependent spatial variation in mortality.

Defining null expectations for animal site fidelity

Site fidelity-the tendency to return to previously visited locations-is widespread across taxa. Returns may be driven by several mechanisms, including memory, habitat selection, or chance; however, pattern-based definitions group different generating mechanisms under the same label of 'site fidelity', often assuming memory as the main driver. We propose an operational definition of site fidelity as patterns of return that deviate from a null expectation derived from a memory-free movement model. First, using agent-based simulations, we show that without memory, intrinsic movement characteristics and extrinsic landscape characteristics are key determinants of return patterns and that even random movements may generate substantial probabilities of return. Second, we illustrate how to implement our framework empirically to establish ecologically meaningful, system-specific null expectations for site fidelity. Our approach provides a conceptual and operational framework to test hypotheses on site fidelity across systems and scales.

Technological innovation facilitates the practice of "three-dimensional ecology"

The development of "three-dimensional ecology" reveals refreshing phenomena and challenges us to use three-dimensional information for studying animal perception. We created a new processing framework to quantify the shielding effect using a reconstructed environmental structure. The framework achieves three objectives: 1) the observed is introduced, 2) the observed space size can be flexibly dealt with, and 3) three-dimensional attributes are assigned to the environmental structure. Our processing framework is an applicable method to "three-dimensional ecology" based on the three-dimensional attributes of physical structures. We advocate for greater emphasis on "three-dimensional ecology" to recreate realistic animal living conditions and better reveal their behaviors.

Displacement Effects of Conservation Grazing on Red Deer (Cervus elaphus) Spatial Behaviour

Conservation grazing uses semi-feral or domesticated herbivores to limit encroachment in open areas and to promote biodiversity. However, we are still unaware of its effects on wild herbivores. This study investigates the influence of herded sheep and goats on red deer (Cervus elaphus) spatial behavior by testing three a-priori hypotheses: (i) red deer are expected to avoid areas used by livestock, as well as adjacent areas, when livestock are present, albeit (ii) red deer increase the use of these areas when sheep and goats are temporarily absent and (iii) there is a time-lagged disruption in red deer spatial behavior when conservation grazing practice ends. Using GPS-telemetry data on red deer from a German heathland area, we modelled their use of areas grazed by sheep and goats, using mixed-effect logistic regression. Additionally, we developed seasonal resource selection functions (use-availability design) to depict habitat selection by red deer before, during, and after conservation grazing. Red deer used areas less during conservation grazing throughout all times of the day and there was no compensatory use during nighttime. This effect mostly persisted within 21 days after conservation grazing. Effects on habitat selection of red deer were detectable up to 3000 meters away from the conservation grazing sites, with no signs of either habituation or adaption. For the first time, we demonstrate that conservation grazing can affect the spatio-temporal behavior of wild herbivores. Our findings are relevant for optimizing landscape and wildlife management when conservation grazing is used in areas where wild herbivores are present.

Individual Network Topology of Patch Selection Under Influence of Drifting Site Fidelity

Network theory has led to important insight into statistical-mechanical aspects of systems showing scaling complexity. I apply this approach to simulate the behavior of animal space use under the influence of memory and site fidelity. Based on the parsimonious Multi-scaled random walk model (MRW) an emergent property of self-reinforcing returns to a subset of historic locations shows how a network of nodes grows into an increased hierarchical depth of site fidelity. While most locations along a movement path may have a low revisit probability, habitat selection is maturing with respect to utilization of the most visited patches, in particular for patches that emerge during the early phase of node development. Using simulations with default MRW properties, which have been shown to produce space use in close statistical compliance with utilization distributions of many species of mammals, I illustrate how a shifting spatio-temporal mosaic of habitat utilization may be described statistically and given behavioral-ecological interpretation. The proposed method is illustrated with a pilot study using black bear Ursus americanus telemetry fixes. One specific parameter, the Characteristic Scale of Space Use, is here shown to express strong resilience against shifting site fidelity. This robust result may seem counter-intuitive, but is logical under the premise of the MRW model and its relationship to site fidelity, whether stable or shifting spatially over time. Thus, spatial analysis of the dynamics of a gradually drifting site fidelity using simulated scenarios may indirectly cast light on the dynamics of movement behavior as preferred patches are shifting over time. Both aspects of complex space use, network topology and dynamically drifting dispersion of site fidelity, provide in tandem important descriptors of behavioral ecology with relevance to habitat selection.

Cognitive maps in the wild: revealing the use of metric information in black howler monkey route navigation

When navigating, wild animals rely on internal representations of the external world – called ‘cognitive maps’ – to take movement decisions. Generally, flexible navigation is hypothesized to be supported by sophisticated spatial skills (i.e. Euclidean cognitive maps); however, constrained movements along habitual routes are the most commonly reported navigation strategy. Even though incorporating metric information (i.e. distances and angles between locations) in route-based cognitive maps would likely enhance an animal's navigation efficiency, there has been no evidence of this strategy reported for non-human animals to date. Here, we examined the properties of the cognitive map used by a wild population of primates by testing a series of cognitive hypotheses against spatially explicit movement simulations. We collected 3104 h of ranging and behavioural data on five groups of black howler monkeys (Alouatta pigra) at Palenque National Park, Mexico, from September 2016 through August 2017. We simulated correlated random walks mimicking the ranging behaviour of the study subjects and tested for differences between observed and simulated movement patterns. Our results indicated that black howler monkeys engaged in constrained movement patterns characterized by a high path recursion tendency, which limited their capacity to travel in straight lines and approach feeding trees from multiple directions. In addition, we found that the structure of observed route networks was more complex and efficient than simulated route networks, suggesting that black howler monkeys incorporate metric information into their cognitive map. Our findings not only expand the use of metric information during route navigation to non-human animals, but also highlight the importance of considering efficient route-based navigation as a cognitively demanding mechanism.

Highlighted Article: Black howler monkeys rely on route-based cognitive maps, which constrain their movement decisions, but likely incorporate metric information to navigate more efficiently along frequently used routes.

A random walk model that accounts for space occupation and movements of a large herbivore

Animal movement has been identified as a key feature in understanding animal behavior, distribution and habitat use and foraging strategies among others. Large datasets of invididual locations often remain unused or used only in part due to the lack of practical models that can directly infer the desired features from raw GPS locations and the complexity of existing approaches. Some of them being disputed for their lack of biological justifications in their design. We propose a simple model of individual movement with explicit parameters, based on a two-dimensional biased and correlated random walk with three forces related to advection (correlation), attraction (bias) and immobility of the animal. These forces can be directly estimated using individual data. We demonstrate the approach by using GPS data of 5 red deer with a high frequency sampling. The results show that a simple random walk template can account for the spatial complexity of wild animals. The practical design of the model is also verified for detecting spatial feature abnormalities and for providing estimates of density and abundance of wild animals. Integrating even more additional features of animal movement, such as individuals’ interactions or environmental repellents, could help to better understand the spatial behavior of wild animals.

Dynamics of animal joint space use: a novel application of a time series approach

BACKGROUND

Animal use is a dynamic phenomenon, emerging from the movements of animals responding to a changing environment. Interactions between animals are reflected in patterns of joint space use, which are also dynamic. High frequency sampling associated with GPS telemetry provides detailed data that capture space use through time. However, common analyses treat joint space use as static over relatively long periods, masking potentially important changes. Furthermore, linking temporal variation in interactions to covariates remains cumbersome. We propose a novel method for analyzing the dynamics of joint space use that permits straightforward incorporation of covariates. This method builds upon tools commonly used by researchers, including kernel density estimators, utilization distribution intersection metrics, and extensions of linear models.

METHODS

We treat the intersection of the utilization distributions of two individuals as a time series. The series is linked to covariates using copula-based marginal beta regression, an alternative to generalized linear models. This approach accommodates temporal autocorrelation and the bounded nature of the response variable. Parameters are easily estimated with maximum likelihood and trend and error structures can be modeled separately. We demonstrate the approach by analyzing simulated data from two hypothetical individuals with known utilization distributions, as well as field data from two coyotes (Canis latrans) responding to appearance of a carrion resource in southern Texas.

RESULTS

Our analysis of simulated data indicated reasonably precise estimates of joint space use can be achieved with commonly used GPS sampling rates (s.e.=0.029 at 150 locations per interval). Our analysis of field data identified an increase in spatial interactions between the coyotes that persisted for the duration of the study, beyond the expected duration of the carrion resource. Our analysis also identified a period of increased spatial interactions before appearance of the resource, which would not have been identified by previous methods.

CONCLUSIONS

We present a new approach to the analysis of joint space use through time, building upon tools commonly used by ecologists, that permits a new level of detail in the analysis of animal interactions. The results are easily interpretable and account for the nuances of bounded serial data in an elegant way.

From Birds to Bacteria: Generalised Velocity Jump Processes with Resting States

There are various cases of animal movement where behaviour broadly switches between two modes of operation, corresponding to a long-distance movement state and a resting or local movement state. Here, a mathematical description of this process is formulated, adapted from Friedrich et al. (Phys Rev E, 74:041103, 2006b). The approach allows the specification any running or waiting time distribution along with any angular and speed distributions. The resulting system of integro-partial differential equations is tumultuous, and therefore, it is necessary to both simplify and derive summary statistics. An expression for the mean squared displacement is derived, which shows good agreement with experimental data from the bacterium Escherichia coli and the gull Larus fuscus. Finally, a large time diffusive approximation is considered via a Cattaneo approximation (Hillen in Discrete Continuous Dyn Syst Ser B, 5:299–318, 2003). This leads to the novel result that the effective diffusion constant is dependent on the mean and variance of the running time distribution but only on the mean of the waiting time distribution.

Identifying Space Use at Foraging Arena Scale within the Home Ranges of Large Herbivores

An intermediate spatiotemporal scale of food procurement by large herbivores is evident within annual or seasonal home ranges. It takes the form of settlement periods spanning several days or weeks during which foraging activity is confined to spatially discrete foraging arenas, separated by roaming interludes. Extended by areas occupied for other activities, these foraging arenas contribute towards generating the home range structure. We delineated and compared the foraging arenas exploited by two African large herbivores, sable antelope (a ruminant) and plains zebra (a non-ruminant), using GPS-derived movement data. We developed a novel approach to specifically delineate foraging arenas based on local change points in distance relative to adjoining clusters of locations, and compared its output with modifications of two published methods developed for home range estimation and residence time estimation respectively. We compared how these herbivore species responded to seasonal variation in food resources and how they differed in their spatial patterns of resource utilization. Sable antelope herds tended to concentrate their space use locally, while zebra herds moved more opportunistically over a wider set of foraging arenas. The amalgamated extent of the foraging arenas exploited by sable herds amounted to 12-30 km2, compared with 22-100 km2 for the zebra herds. Half-day displacement distances differed between settlement periods and roaming interludes, and zebra herds generally shifted further over 12h than sable herds. Foraging arenas of sable herds tended to be smaller than those of zebra, and were occupied for period twice as long, and hence exploited more intensively in days spent per unit area than the foraging arenas of zebra. For sable both the intensity of utilization of foraging arenas and proportion of days spent in foraging arenas relative to roaming interludes declined as food resources diminished seasonally, while zebra showed no seasonal variation in these metrics. Identifying patterns of space use at foraging arena scale helps reveal mechanisms generating the home range extent, and in turn the local population density. Thereby it helps forge links between behavioural ecology, movement ecology and population ecology.

Uniting statistical and individual-based approaches for animal movement modelling

The dynamic nature of their internal states and the environment directly shape animals' spatial behaviours and give rise to emergent properties at broader scales in natural systems. However, integrating these dynamic features into habitat selection studies remains challenging, due to practically impossible field work to access internal states and the inability of current statistical models to produce dynamic outputs. To address these issues, we developed a robust method, which combines statistical and individual-based modelling. Using a statistical technique for forward modelling of the IBM has the advantage of being faster for parameterization than a pure inverse modelling technique and allows for robust selection of parameters. Using GPS locations from caribou monitored in Québec, caribou movements were modelled based on generative mechanisms accounting for dynamic variables at a low level of emergence. These variables were accessed by replicating real individuals' movements in parallel sub-models, and movement parameters were then empirically parameterized using Step Selection Functions. The final IBM model was validated using both k-fold cross-validation and emergent patterns validation and was tested for two different scenarios, with varying hardwood encroachment. Our results highlighted a functional response in habitat selection, which suggests that our method was able to capture the complexity of the natural system, and adequately provided projections on future possible states of the system in response to different management plans. This is especially relevant for testing the long-term impact of scenarios corresponding to environmental configurations that have yet to be observed in real systems.

The Lévy flight foraging hypothesis: forgetting about memory may lead to false verification of Brownian motion

BACKGROUND

The Lévy flight foraging hypothesis predicts a transition from scale-free Lévy walk (LW) to scale-specific Brownian motion (BM) as an animal moves from resource-poor towards resource-rich environment. However, the LW-BM continuum implies a premise of memory-less search, which contradicts the cognitive capacity of vertebrates.

RESULTS

We describe methods to test if apparent support for LW-BM transitions may rather be a statistical artifact from movement under varying intensity of site fidelity. A higher frequency of returns to previously visited patches (stronger site fidelity) may erroneously be interpreted as a switch from LW towards BM. Simulations of scale-free, memory-enhanced space use illustrate how the ratio between return events and scale-free exploratory movement translates to varying strength of site fidelity. An expanded analysis of GPS data of 18 female red deer, Cervus elaphus, strengthens previous empirical support of memory-enhanced and scale-free space use in a northern forest ecosystem.

CONCLUSION

A statistical mechanical model architecture that describes foraging under environment-dependent variation of site fidelity may allow for higher realism of optimal search models and movement ecology in general, in particular for vertebrates with high cognitive capacity.