Hidden properties: The kea´s understanding of weight

Throughout nature, animals are faced with the universal challenge of navigating their physical environment. To do so, they must rely on some degree of knowledge about the physical world, including the properties of objects within it. Principal among these is weight: though it cannot be seen directly, weight pervades nearly every aspect of our daily lives. We adjust our lifting forces to pick up objects of different weights, predict the weights of new objects based on their appearance or behavior, and even flexibly select objects of different weights depending on the task at hand (e.g., choosing a stone to hold a tablecloth in place on a windy day). Though all animals experience weight on a daily basis, it has been suggested that humans have a uniquely sophisticated understanding of weight; however, we still know surprisingly little about other species understanding of this property, including how it might reveal itself in different situations. To what extent do other animals attend to and reason about weight, and where might the limits lie? The proposed project aims to examine weight understanding in kea (Nestor notabilis), a large-brained and highly innovative parrot species. It analyzes this capacity over three key levels, focusing on (i) whether and how kea attend to the weight of objects, (ii) whether they infer the weight of objects without handling them directly, and (iii), whether they use information about weight to flexibly solve problems. Addressing these questions will provide insight into how nonhuman animals perceive and reason about the objects they interact with, and ultimately pinpoint critical differences between human and nonhuman minds.

We presented kea parrots with a battery of tasks designed to assess their sensitivity to object weight, as well as the visual, auditory and causal cues they might use to infer weight. We found that kea rapidly learned to discriminate between objects based solely on weight with minimal experience. This finding is particularly striking in light of research with primates, which can take hundreds of trials to acquire similar discriminations. One possible explanation for this difference may be that weight is a particularly salient property for birds, which are exceptionally lightweight and often transport objects during flight. Moreover, kea demonstrated a high sensitivity to small differences in object weight. Although kea clearly attended to weight information while handling objects, we found no evidence that they retained information about an object's weight after prior interactions. Additionally, they did not appear to use visual, auditory, or causal cues to infer object weight in the absence of direct contact. Taken together, these results suggest that the kea's understanding of weight may be limited to first-order perceptual information - specifically, the proprioceptive feedback received through direct interaction with objects. This pattern may be related to the kea's highly exploratory nature, in which they tend to rely on the haptic information acquired through active manipulation rather than on visual information, which may be more commonly used by less exploratory or more neophobic species. Overall, these findings contribute to a broader discussion concerning whether only humans possess a more abstract understanding of weight as an enduring property of objects that exists independently of direct experience.