Inside a transparent arena at the University of Oulu, a bumble bee faced a problem its kind had never encountered: a blue artificial flower, glowing with the promise of reward, had been moved to the ceiling—just out of reach. The bee circled, explored, then did something startling. It found a ball on the arena floor, rolled it beneath the flower, and climbed atop it. Success. It had solved the task without ever being shown how, without training, without instruction. It was spontaneous insight in an insect, challenging everything scientists thought they knew about which brains could solve problems.
The discovery, published in Science by researchers from the University of Oulu, the University of Helsinki, and the University of Turku, upends a century of assumptions about animal cognition. More than 100 years ago, psychologist Wolfgang Köhler demonstrated that chimpanzees could solve novel problems by stacking boxes to reach an out-of-reach banana—a landmark finding that became the gold standard for measuring spontaneous problem-solving in the animal kingdom. Now bumble bees (Bombus terrestris) have shown strikingly similar abilities in a task researchers describe as an insect version of that very problem.
What makes the result so striking is what the bees didn't know. Before the test, they had learned only two separate pieces of information: that a blue flower contained reward, and that a ball was movable and harmless. They had never been trained to roll the ball, never shown the solution, never practiced combining these two experiences into action. Yet many bees, when placed in that novel situation, did exactly what the task required. "This was a completely new challenge," explains lead author Akshaye Bhambore from the University of Oulu. "Their behavior appeared goal-directed, with successful individuals showing more directed movement patterns."
The researchers designed the experiments with unusual rigor, building in multiple control conditions to eliminate simpler explanations. They tested bees in scenarios where the flower was hidden from view while the bees moved the ball—preventing them from simply steering toward a visible target. Even with the flower out of sight, the bees moved the ball to the correct location. They ruled out accidental success, trial-and-error learning, play behavior, and direct visual guidance. The bees were not reacting to stimuli. They were solving.
"Another important aspect is that our bees were fully naïve," says senior author Olli Loukola, Docent at the University of Oulu. "In many previous studies of insight-like problem-solving, the animals have had extensive experience with objects, test environments, or other problem-solving tasks. Here, the bees had never been trained to use the ball to reach the flower, and they had no previous experience with this kind of solution."
The finding carries profound implications. If a brain with fewer than one million neurons can generate novel solutions to entirely new problems, then spontaneous problem-solving may not require the massive neural architecture of vertebrates. It suggests that the cognitive flexibility once thought to be the exclusive domain of humans and large-brained animals operates by different principles than neuroscience has assumed—principles that evolution discovered not once, but in radically different forms. The bees have shown us that insight is not about brain size. It is about how a mind works.