← Three short discoveries on WatWat
A white Yorkshire pig standing in a dirt pen beside a wooden fence
Strange but true

Moving a joystick is one thing, and understanding that your movement moves a cursor on a screen is another thing entirely. Four pigs were trained to push a joystick with the snout and bring a cursor to a colored wall. What did they understand?

Tested in animalsSample: 4 pigs

More than chance, with an built for hands. Two Yorkshire pigs and two Panepinto pigs learned to move a joystick with the snout, bringing a cursor on screen to a colored wall. All four did better than in some conditions, including when the number of walls shrank and the target grew harder. There was also : sometimes they kept going correctly when the food machine broke, responding to the handler’s encouragement.

A pig at a screen with a joystick sounds like a stunt, and the question under test is narrow. Does it connect a movement of the snout with , the travel of a cursor on a screen?

Four pigs were trained to move the handle with the snout and bring a cursor to a colored wall. The interface does not suit them: a snout does not grip, so a failure could have come from the tool and not from understanding.

All four did better than random choice in some conditions, including when the number of walls shrank and the task grew harder, which is the stricter test.

Sometimes they kept performing correctly when the food machine broke, responding to the handler’s encouragement. The sample is very small, and vision and motor control limited the performance, so the result is not equivalent to a human understanding of a computer game.

wat. Four pigs learned to move a cursor on a screen with a joystick they worked with the snout, and did better than chance.

wat wat

Pigs can connect an action of the snout with a distant visual outcome, although an ordinary computer interface is built for a human hand and eye.

How do we know?

The researchers analyzed the choices of four pigs in stages with four, three, two or one target wall, and compared them with chance performance.

Tested in animalsPeer-reviewed studySample: 4 pigs

What is still unknown

The sample is very small and the pigs’ vision and motor control limited the performance. The result is not equivalent to a human understanding of a computer game.

Learn something?

Croney & Boysen, Frontiers in Psychology· 2021· DOI 10.3389/fpsyg.2021.631755

Where to next

Did a brain scan prove that writing by hand teaches better than typing?learningA cockatoo wants the seed inside a hard fruit. Why is one tool not enough?learningWho else can tell a benign slide from a malignant one?learning

Want another one worth repeating?

Start with three short discoveries. No sign-up, nothing to pick first.

Surprise me
Browse the topic libraryAbout WatWat, and how sources get checked