For six years, researchers watched the same bull sharks return to Shark Reef Marine Reserve in Fiji, and they kept noticing something unexpected: certain sharks always showed up together. Not just in the same general area, but within a body length of each other, swimming in sync, one trailing the other in repeated lead-follow sequences. Out of 184 individually identified sharks, specific pairs kept reuniting, season after season, in patterns that could not be explained by coincidence.
A study published in May 2026 in the journal Animal Behaviour now puts hard numbers behind those observations. The research, led by scientists from the University of Exeter, Lancaster University, Fiji Shark Lab, and Beqa Adventure Divers, found that bull sharks form persistent, non-random social associations that include coordinated behaviors like parallel swimming. “We were struck by how consistent these pairings were across years,” said Yannis Papastamatiou, a marine scientist involved in the research. “These aren’t sharks that just happen to be in the same place. They are actively choosing to be near specific individuals.” The findings challenge the longstanding image of bull sharks as solitary, indiscriminate roamers and suggest that even among apex predators, social relationships matter.
Tracking 184 sharks by sight
The team identified individual sharks using distinctive markings, scars, and fin shapes, a photo-identification method that allowed them to follow specific animals across years without tagging. They recorded two scales of social behavior: broad-scale associations, meaning two sharks appearing within roughly one body length of each other during the same observation period, and fine-scale interactions, including lead-follow sequences and side-by-side parallel swimming.
To test whether the patterns were real, the researchers built social networks mapping how often each shark appeared with every other shark, then compared those networks against thousands of computer-generated simulations in which shark identities were shuffled at random. The real networks showed far stronger clustering and more repeated pairings than any of the simulations predicted. In plain terms, certain bull sharks were choosing to spend time together.
The work builds directly on a 2021 study in Frontiers in Marine Science that first documented preferred companionships among bull sharks at the same Fiji site. The newer paper extends that dataset from a few years to six, refines the statistical methods, and adds the fine-scale behavioral layer. Together, the two studies form the most detailed picture yet of social structure in bull sharks.
Not the first sharks caught socializing
Bull sharks are not the only species rewriting the script on shark sociality. A landmark 2009 study in Animal Behaviour showed that juvenile lemon sharks in the Bahamas formed non-random associations, providing some of the earliest statistical evidence that sharks can exhibit measurable social preferences. Subsequent work by Mourier et al., published in 2012 in Proceedings of the Royal Society B, used social network analysis to reveal fission-fusion dynamics in blacktip reef sharks, showing that groups form, dissolve, and re-form in structured ways rather than assembling at random.
What sets the Fiji bull shark work apart is the combination of long duration, large sample size, and behavioral detail. Documenting not just who is near whom, but how they move together, pushes the evidence beyond co-occurrence and toward something that looks more like active social coordination.
The provisioning question
The most important caveat is one the researchers themselves have acknowledged: every observation in both the 2026 and 2021 studies was collected at a site where dive operators regularly feed sharks. Shark Reef Marine Reserve is home to Beqa Adventure Divers, one of Fiji’s longest-running shark-dive operations, and the regular presence of food draws bull sharks to the area in concentrations that may not occur naturally.
That raises a legitimate question. Are these social bonds a feature of bull shark biology, or an artifact of artificial aggregation? Sharks that might otherwise range across hundreds of kilometers could be meeting more frequently because they are drawn to the same food source, and repeated co-occurrence at a buffet is not the same thing as friendship.
No comparable dataset exists from a site where bull sharks are not provisioned, so there is currently no way to measure how much the feeding operation amplifies or distorts the social patterns. The statistical methods control for the possibility that sharks simply cluster near food at random, and the results still show preferences beyond what random crowding would produce. But until researchers replicate the work at non-provisioned sites, the influence of ecotourism feeding on these social structures cannot be fully separated from the sharks’ natural tendencies.
Open questions that matter
Beyond provisioning, several gaps remain. The studies do not report whether preferred companions are genetically related. If bull sharks preferentially associate with kin, the biological explanation shifts toward inclusive fitness, helping relatives survive. If they choose unrelated partners, the relationships may hinge on mutual benefits like cooperative foraging or reduced harassment from larger individuals. Without genetic sampling tied to the social network data, that distinction stays unresolved.
Geographic scope is another limitation. All primary evidence for bull shark sociality comes from a single location. Whether populations in the Gulf of Mexico, off the coast of South Africa, or in Australian river systems form similar bonds has not been tested with comparable methods. Bull sharks occupy an enormous range of habitats, from warm coastal waters to freshwater rivers hundreds of kilometers inland, and social behavior could vary dramatically across those environments.
Function is perhaps the biggest unknown. The studies demonstrate that preferred associations exist and involve coordinated movement, but they do not yet explain what those relationships accomplish. Hypotheses range from improved foraging efficiency and information sharing to reduced predation risk, social learning about human activity, and support for mating hierarchies. Observational data alone cannot confirm which, if any, of these benefits drive the behavior.
What the Fiji findings mean for marine reserves
If bull sharks depend on social relationships for foraging, safety, or reproduction, then conservation strategies focused solely on protecting individual animals may miss something critical. Marine reserves that safeguard known aggregation sites could carry added value if those sites also function as social hubs where sharks maintain bonds that influence their survival.
Shark Reef Marine Reserve itself offers a case study. The site has operated as a no-fishing zone since 2004, and the long-term presence of identifiable sharks suggests that at least some individuals treat it as a reliable home base. If the social networks documented there prove to be important for the sharks’ well-being, the reserve model gains an argument that goes beyond simply keeping sharks alive: it may also be preserving the social fabric those sharks rely on.
For now, the most defensible conclusion is specific: bull sharks at a single, well-studied site in Fiji show clear, statistically robust preferences for certain companions and engage in repeated, coordinated interactions with them over multiple years. Extending that conclusion to bull sharks worldwide will require genetic sampling, multi-site comparisons, and research that links social patterns to measurable outcomes like survival and reproductive success.
But even within its geographic limits, the Fiji work carries a broader implication. The ocean’s most notorious generalist predator, an animal comfortable in salt water and fresh, famous for turning up in rivers and canals, appears to navigate its world with an awareness of specific individuals. Bull sharks, it turns out, are not just tracking prey and avoiding threats. They are also keeping track of each other.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
