Can giraffes do maths? Maybe, when it comes to getting their favourite snacks

Giraffes are many things. Elegant. Enigmatic. Tall, obviously. But maths nerds?

Well, a team from Barcelona Zoo argues they can use basic statistics by predicting the odds of getting their favourite snack, proving they have brains and beauty.

The ability to make statistical inferences is considered a highly-developed cognitive function, but has only previously been studied in large-brained animals such as primates and parrots.

Testing the abilities of two male and two female giraffes, the team showed each individual two containers containing varying proportions and amounts of their preferred snack, carrot sticks, and less-preferred courgette sticks.

The researcher then ‘covertly’ picked up a treat from each container in a closed fist so the giraffes could not see what had been selected – although the see-through containers were still in view – and held out both options for them to choose from.

In at least 17 out of 20 trials, the giraffes were reliably able to select the container that was more likely to produce their favoured carrot sticks. 

The authors used control conditions to rule out whether the giraffes were using other information such as smell, and eliminated alternative explanations based on learning processes.

Writing in the journal Scientific Reports, the team, led by Federica Amici, said: ‘In this study, giraffes could reliably make statistical inferences based on the relative frequencies of two different food types. 

‘Like chimpanzees and keas, giraffes spontaneously selected the container more likely to provide the preferred food in the experimental conditions, even when subjects could not rely on simpler quantity heuristics – for example, because the correct container did not contain a higher number of highly-liked food, and the wrong container did not contain a higher number of less-preferred food.

‘Giraffes were surprisingly fast at solving the first experimental task, requiring on average 1.2 sessions to reliably select the correct container in at least 17 out of 20 trials. In contrast, keas tested with the same procedure required an average of 3.9 sessions, and up to 11 sessions, to solve the task.

‘These results suggest that large relative brain sizes are not a necessary prerequisite for the evolution of complex statistical skills, and that the ability to make statistical inferences may be widespread in the animal kingdom.’

Source: Read Full Article