A cauldron of bats. Image credit: Jens Rydell (CCBY 4.0)
Bats rely on echolocation to navigate and hunt in complete darkness. They emit high-frequency calls and use the returning echoes to detect the distance, shape and location of objects around them. However, when many bats are flying in the same area and calling at the same time – such as during group takeoffs from a crowded cave – the overlapping calls can interfere with each other, creating a phenomenon known as acoustic jamming. This interference may block important echoes, making it harder to detect obstacles or find a way out.
Scientists have long questioned how bats manage to avoid collisions and successfully navigate in such noisy, crowded conditions. If echolocation is still effective under such conditions, what strategies might these animals use to overcome interference? To investigate this, Mazar and Yovel developed a computer model that simulates how individual bats might behave and perceive their environment when flying in large, acoustically noisy groups.
The simulations showed that acoustic jamming is less problematic than previously assumed. Frequent call emission allows bats to collect redundant sensory information, and integrating these signals over short periods helps them form a reliable picture of their surroundings. In addition, simple behavioral strategies, such as following walls and avoiding collisions with other bats, allow them to navigate safely. The researchers also found that an agent-based model using signal redundancy, short-term memory, and local movement rules can robustly replicate the dynamics seen in real bats. These findings suggest that bats have evolved behavioral and sensory adaptations that make echolocation effective even in dense, noisy conditions.
These findings can benefit researchers studying bat behavior, sensory processing and neural adaptation to complex environments. They may also inform the design of drone swarms or autonomous agents that need to operate in crowded, noisy conditions. Importantly, this study shows how simulations, particularly agent-based models, are powerful tools for exploring complex group behavior and sensory challenges. To apply these insights practically, further empirical validation and real-world implementation in robotics or neuroscience are needed.
