Previous research has established that cats predict the presence of invisible objects based on what they hear.
In the new study, researchers from Kyoto University in Japan wanted to find out if cats use a causal rule to infer if a container holds an object, based on whether it is shaken along with a sound or not.
The team also wanted to establish if cats expect an object to fall out or not, once the container is turned over.
In others it did not, to simulate that the vessel was empty. After the shaking phase, the container was turned over, either with an object dropping down or not.
Also Read
Two experimental conditions were congruent with physical laws, where shaking was accompanied by a sound and an object falling out of the container.
The other two conditions were incongruent to the laws of physics. Either a rattling sound was followed by no object dropping out of the container or no sound while shaking led to a falling object.
This helped them predict whether an object would appear once the container was overturned.
The animals also stared longer at containers in incongruent conditions, meaning an object dropped despite its having been shaken noiselessly or the other way around.
It is as if the cats realised that such conditions did not fit into their grasp of causal logic.
"Cats use a causal-logical understanding of noise or sounds to predict the appearance of invisible objects," said Saho Takagi, of Kyoto University who led the research.
The ecology of cats' natural hunting style may therefore also favour the ability for inference on the basis of sounds.
Takagi said that hunting cats often need to infer the location or the distance of their prey from sounds alone because they stake out places of poor visibility.
The study was published in the journal Animal Cognition.