Researchers Create "General Intelligence" Test For Wild Birds


Janet Fang

Staff Writer

clockSep 21 2015, 20:10 UTC
2512 Researchers Create "General Intelligence" Test For Wild Birds
New Zealand North Island robin. Tony Wills/Wikimedia Commons CC BY 2.5

Human IQ-type tests are scored using trials assessing things like processing speed, working memory and verbal comprehension. One single factor, called “general intelligence” (or g), accounts for at least 40% of the differences between people’s scores. 


In order to study how animals acquire, process, store, and act on information, Victoria University of Wellington’s Rachael Shaw and colleagues developed a cognitive performance test for New Zealand North Island robins, Petroica longipes. Their findings, published in Animal Behaviour, suggest that a similar g-factor exists in animal species, especially those that maintain caches of food. 

The team conducted their experiments at the Zealandia wildlife sanctuary in Wellington. Up to 765 wild robins live within its predator exclusion fences, and 20 of them were trapped, measured, banded, and then released by the researchers. Luckily, they voluntarily participated in the team’s cognitive experiments. A wooden board, which served as a test platform, was placed on the ground under the canopy within the birds’ territories.

The cognitive test battery consisted of six tasks: motor task (a), color discrimination and color reversal (b), spatial memory (c), inhibitory control (d, e), and symbol discrimination (similar setup as b). And, to make sure the birds stayed motivated, the researchers trained them to hop onto a scale and eat a worm before and after each test (f). 

The test apparatuses, with a robin for scale. R.C. Shaw et al., 2015 Animal Behaviour


For the motor skill task (a), the robins had to flip white PVC lids on a wooden block, which contained six wells. Four of the six wells were baited with a freshly killed mealworm; sometimes the lid would be covering the well, other times it’d be next to it. In the spatial memory task (c), eight of 12 wells were covered, and one of covered wells contained a mealworm. The team made sure the same well was baited during each of several trials. For the inhibitory control task, a mealworm was hidden in opaque (d) and transparent (e) tubes with open ends; to pass the test, the birds had to detour to the ends to retrieve the worm without pecking at the wall of the tube first. For the color and symbol tasks (b), the birds had to show they could discriminate between two differently decorated lids – red or blue, cross or square – that were the covering wells. 

The robins varied greatly in their ability to solve these tasks, and the team found a weakly positive correlation between most tasks. That means if a robin did well on one test, it was likely to do well on another one. Between 34% and 45% of the differences in the robins’ overall performances could be explained by a "general cognitive" factor similar to our g-factor, Science explains, suggesting that their performance isn’t just because of talent for specific tasks – but because of their overall braininess. 

  • intelligence,

  • birds,

  • IQ test,

  • g-factor,

  • g factor,

  • general intelligence