| Title | Creative problem solving in a jumping spider |
| Author | Wang, Yi-Rong |
| Publication | ZOOLOGICAL RESEARCH |
| Year | 2026 |
| Volume | 47 |
| Issue | 5 |
| Corresponding Author | Jiao, XG;Liu, JX; Chen, ZQ |
| Abstract | Behavioural innovation, a key indicator of advanced cognition, has been well documented across animal taxa and manifests as creative problem-solving. Typical innovators such as apes and corvids can spontaneously solve complex, multi-step tasks individually without prior training. Innovative behaviours in arthropods-honeybees, bumblebees, and ants-are probably underestimated because their innovative behaviours are often considered products of associative learning, group-level processes, or one-step problem-solving. Here, we challenge this view by demonstrating spontaneous, individual-level innovation in nest construction by the social-living jumping spider Toxeus maxillosus, without prior training or external demonstration. Under four challenging laboratory conditions, T. maxillosus females exhibited the core components of typical innovative behaviours: environmental exploration, behavioural flexibility, insight, and action planning. Furthermore, in a vertical string-pulling task (a benchmark test of animal innovation), half of the spiders pulled an out-of-reach sponge upward and incorporated it into the inner wall of the device, thereby creating a suitable microstructure for nest construction. These behaviours demonstrate multi-step, goal-directed object manipulation. Our findings suggest that arthropods are capable of individual-level, spontaneous innovation involving planning and multi-step sequences, akin to those observed in typical innovators. We propose that a social lifestyle centred on prolonged maternal care-a trait shared by T. maxillosus and many vertebrate innovators-rather than large brain size or long lifespan, may be a critical evolutionary driver of such cognitive complexity. This study challenges the vertebrate-centric view of typical innovation and provides a new framework for understanding the evolution of cognition across scales. |