Cats Like To Sit In Squares — Even Ones That Are Really Optical Illusions

By Emma Young

The world is not exactly short of videos of cute cats up to strange antics. But one particular set of videos collected by cat owners during a COVID-19 lockdown reveals something genuinely interesting: a famous optical illusion that fools us also gets cats. The citizen science project, in which cats were experimented on in their own homes, shows that they, too, are tricked by “Kanizsa squares”, an illusion that suggests the presence of a square that doesn’t in fact exist.

It’s well known that cats love to sit in enclosed spaces, like boxes. They even like to sit in square shapes made of tape, as documented during a 2017 Twitter craze (#CatSquare). So in the new study, published in Applied Animal Behaviour Science, Gabriella E. Smith at City University of New York and colleagues took this further, to see if cats also show a preference for illusory squares.

Of the 500 pet cats and owners who signed up to take part, 30 completed all the trials. The owners “were not aware of the study’s investigative purpose at any point before or during the experiment”, the authors write. (Well, they weren’t told what it was about, at least.)

Every day for six days, each owner put their cat out of the room while they taped a pair of stimuli to the floor. The researchers instructed them each day on which two of three stimuli to use: an actual square, a Kanizsa square (in which four Pacman-type cut-outs are arranged to suggest the lines of a square) and the Kanizsa control (in which the Pacman-type cut-outs face the other way.)

Examples of Kanizsa squares (right) paired with the Kanizsa control (top panel) and actual square (bottom panel). Via Smith et al (2021)

The owner then put on sunglasses (so their cat couldn’t use their gaze as a cue to what to do), brought the cat into the room, and started videoing. If, within five minutes, their pet sat or stood within either of the shapes, the trial ended. Either way, after five minutes, they submitted the result.

Of those 30 cats who completed all the trials, nine cats chose to inside a stimulus at least once — and, in total between them, 16 times. On eight of these occasions, a cat sat inside an actual square. On seven, it sat inside a Kanizsa square. In contrast, the Kanizsa control was chosen only once. This suggests that cats treat the Kanizsa squares just like real ones. Like us (and dogs, chimps, and many other animals, in fact) they show “susceptibility to illusory contours.”

Recruiting pet owners to do experiments themselves clearly has some downsides. Very few of those who signed up actually completed the full study. Also, light levels might have varied quite a bit across the various homes, and this could have influenced the cats’ decisions, the team notes. However, there are also good reasons to study cats in their own homes, rather than a lab. As they’re more relaxed at home, their behaviour is likely to be more natural.

The study provides further evidence that animals experience all kinds of illusions. But it also gives further insight into cat cognition, which has received little attention compared to that of dogs. Quite why not isn’t clear. However, given the useful, if limited, data from this new study, the researchers write: “the use of citizen science as a precursor to in-lab investigations of cat cognition could greatly help bridge this divide.” 

If I fits I sits: A citizen science investigation into illusory contour susceptibility in domestic cats (Felis silvestris catus)

Emma Young (@EmmaELYoung) is a staff writer at BPS Research Digest