The question matters because the existing literature has been strikingly inconsistent. Researchers have long known that bats modify their echolocation behavior when flying in groups compared with flying alone, but studies have reported different, sometimes conflicting, patterns in exactly which call properties change and in which direction. Some of this confusion is methodological. Echolocation recordings from a colony are a mixture of many overlapping signals, and separating the influence of social context from the influence of physical surroundings is genuinely difficult. A bat flying through a cluttered forest edge will naturally adjust its calls whether or not any companions are nearby, so disentangling the two effects requires data collected in a setting where both vary and an analytical framework flexible enough to detect their influence without prejudging its form.

That is precisely the approach the team took. Working with a colony of wild gray bats, they recorded echolocation calls as animals emerged from the roost and flew through their environment, while also tracking how many bats were emerging at a given time and noting the presence of obstacles along the flight path. Gray bats are an ideal subject for this kind of work. They are highly social, roosting and emerging in large numbers, which means group size varies substantially across emergence events. Their cave-roosting lifestyle also means they routinely navigate around and through physical structures, so environmental clutter is a natural feature of their daily flights rather than an artificial laboratory condition.

The researchers focused on two call properties that capture the energetic and temporal character of echolocation: the number of calls a bat produces and the acoustic power of those calls. Call rate and call intensity are the levers a bat can pull to change how much information its sonar gathers and how far that sonar reaches. Producing more calls refreshes the acoustic picture of the world more frequently, which is useful when maneuvering quickly or flying through tight spaces. Increasing acoustic power extends the range of detection but also increases the chance that the call interferes with, or is masked by, the calls of nearby conspecifics. The central hypothesis was that both the number of flying companions and the geometry of the surroundings would leave measurable fingerprints on these two properties.

To test that hypothesis without assuming in advance what the relationship should look like, the team turned to a time-series analysis tool called transfer entropy. Transfer entropy quantifies how much knowing the past state of one variable reduces uncertainty about the future state of another, beyond what the variable’s own history already reveals. In this context, it allowed the researchers to ask whether group size carries information about future call properties, and whether the presence of obstacles does, without imposing a linear or any other predetermined functional form on the relationship. This is a crucial advantage in a system as complex as a flying bat colony, where the true dynamics are almost certainly nonlinear and shaped by feedback between individuals and their environment.

The analysis revealed significant transfer entropy values linking both social and environmental factors to call properties. In plain terms, the number of bats emerging together and the presence of obstacles each carried genuine predictive information about how many calls individual bats produced and how acoustically powerful those calls were. The influence was detected without the researchers assuming any particular mathematical shape for it, which strengthens the conclusion that the relationships are real features of the bats’ behavior rather than artifacts of a chosen model. This finding helps resolve the inconsistency in earlier literature by suggesting that reported differences across studies may reflect genuine context dependence: the direction and magnitude of call adjustments plausibly depend on the specific combination of group density and environmental clutter a colony experiences.

Transfer entropy established that the influences exist, but the team wanted to go further and sketch what the underlying dynamics might actually look like. For this they applied a data-driven algorithm known as sparse identification of nonlinear dynamics, or SINDy. The core idea behind SINDy is elegant: given time-series measurements, it searches an enormous library of candidate functions and terms for the simplest set of equations that can reproduce the observed dynamics. Rather than a human modeler guessing that call rate should depend linearly on group size, say, or quadratically on obstacle proximity, SINDy lets the data nominate the terms that matter and discard the rest. The result is a compact mathematical description of echolocation behavior assembled directly from recordings of wild bats going about their nightly business.

The SINDy-based models suggested concrete ways in which gray bats adjust their echolocation in response to conspecifics and obstacles, offering a candidate structure for the phenomenon that earlier descriptive studies could only gesture at. This modeling step is what elevates the study from a correlation hunt to a constructive contribution toward predictive theory. If the identified equations capture real behavioral rules, they can be tested against new recordings, compared across bat species with different social systems, and eventually incorporated into larger models of collective movement in which echolocation serves simultaneously as a sensory channel and, potentially, as a medium of unintended information sharing among group members.

The broader significance of the work lies in how it reframes the interplay between sociality and sensory ecology. Schools of fish and flocks of birds coordinate largely through vision and lateral-line or aerodynamic sensing, channels that do not degrade in the same way when many individuals use them at once. Echolocation is different: it is an active sensing modality in which every individual’s measurement apparatus is also a potential source of noise for everyone else. The new results indicate that gray bats manage this trade-off in a way that is jointly sensitive to who else is flying and what the physical environment demands. Sociality, in other words, is not just a background condition for echolocation but an active driver of its dynamics, intertwined with the geometry of caves, cliffs, and forest corridors through which the animals move.

For researchers studying collective behavior, bioacoustics, and sensory ecology, the study offers both a methodological template and a set of testable expectations. The combination of transfer entropy and SINDy provides a pipeline for extracting dynamical rules from noisy field recordings of animals whose signals overlap and interfere, a challenge that extends well beyond bats to any species using active sensing in groups. For the gray bats themselves, the findings paint a picture of animals continuously recalibrating their sonar against two competing pressures: the need to navigate a cluttered, obstacle-strewn world and the need to keep their acoustic measurements usable in the company of hundreds of calling companions. As the authors note, these results provide new insights into the role of sociality and physical surroundings in shaping echolocation dynamics, and they suggest that the nightly exodus from a gray bat cave is not just a mass of sound but a richly structured information environment, one the bats themselves are constantly measuring, adjusting to, and rewriting call by call.

Subject of Research: How group size and environmental obstacles influence echolocation call properties in flying gray bats

Article Title: Group size and environmental obstacles drive acoustic call properties for gray bats in flight: A data-driven analysis

Article References: Grey, M., Aung, E., & Abaid, N. (2026). Group size and environmental obstacles drive acoustic call properties for gray bats in flight: A data-driven analysis. PLOS Complex Systems, 3(4), e0000100. https://doi.org/10.1371/journal.pcsy.0000100

Image Credits: AI Generated

DOI: 10.1371/journal.pcsy.0000100

Keywords: gray bats, echolocation, Myotis grisescens, transfer entropy, SINDy, collective behavior, bioacoustics, group size, environmental obstacles, nonlinear dynamics, sensory ecology, PLOS Complex Systems