Robotic rattlesnake test shows zoo animals still fear predators they evolved alongside

PLOS

Place a robotic rattlesnake near a zoo enclosure and something interesting happens. Animals that evolved in rattlesnake territory react strongly, retreating and displaying alarm behavior. Animals from regions without rattlesnakes? Much less bothered.

That is the central finding from a study published in PLOS ONE on March 11, 2026, which used a 3D-printed mechanical rattlesnake to test whether captive zoo animals retain antipredator responses shaped by millions of years of coevolution. The answer, at least for the species tested, is yes.

A multimodal predator display

Rattlesnakes are unusual among predators because their warning display engages multiple senses simultaneously. The rattle produces sound. The coiled body and triangular head present a visual signal. The strike posture communicates imminent danger. The researchers built a robotic model that replicated these cues and presented it to various zoo species under controlled conditions.

The study, conducted at a U.S. zoo, examined the responses of multiple species across different enclosures. Animals were categorized as sympatric (sharing natural geographic range with rattlesnakes) or allopatric (from regions without rattlesnakes). The robot was placed near enclosures and observers recorded avoidance behaviors, approach behaviors, and time spent in proximity to the model.

Sympatric species retreated faster and farther

Species that coexist with rattlesnakes in the wild showed significantly stronger avoidance responses than those that do not. These animals moved away more quickly, maintained greater distance from the robot, and displayed more alarm behaviors. The multimodal nature of the display appeared important: combining visual, auditory, and postural cues produced stronger reactions than any single cue alone.

This pattern held even though the zoo animals had never encountered a real rattlesnake. Many were born in captivity and had no direct predator experience. The results suggest that the recognition of rattlesnake warning signals is at least partly innate, encoded through evolutionary history rather than learned through personal experience.

What this tells us about captive animal cognition

The study adds to a growing body of evidence that captive animals retain more of their wild behavioral repertoire than previously assumed. For zoo management and conservation breeding programs, this matters. If animals retain innate predator recognition, enrichment programs could potentially use controlled predator cues to maintain natural vigilance behaviors that would be critical if animals were ever reintroduced to the wild.

The researchers also note that the robotic approach offers a standardized, repeatable way to test antipredator behavior without exposing animals to actual danger. Traditional studies of predator-prey interactions in the wild are difficult to control and ethically constrained. A robot that reliably produces the same display every time allows cleaner comparisons across species and settings.

Limited scope and open questions

The study received no specific funding and was limited in several respects. The number of species tested was modest, and individual sample sizes within each species were small. Zoo enclosure designs varied, which could influence how animals perceived and responded to the robot.

The study could not determine which specific component of the rattlesnake display drove the strongest response. Was it primarily the rattle sound? The visual profile? The combination? Teasing apart these elements would require further experiments with modified robots presenting individual cues in isolation.

It is also unclear how well these results generalize beyond the specific zoo population studied. Different housing conditions, dietary regimes, and levels of human habituation could all influence antipredator behavior in captive animals.

Still, the core finding, that evolutionary history shapes behavioral responses even in animals separated from natural predators by generations of captivity, carries weight. It suggests that the neural circuits underlying predator recognition are durable features of animal cognition, not skills that fade quickly without reinforcement.