A study has suggested that our fear of spiders and snakes is apparently from an early age, even before we are aware of the danger some species pose.

Published in Frontiers in Psychology, the study showed images of these creatures to infants that were six months old. They found that, despite the infants not knowing about them, their pupils dilated in a way that suggested the development of fear.

The findings were led by Stefanie Hoehl from the Max Planck Institute for Human Cognitive and Brain Science in Germany.

Two experiments were conducted, with the first involving 16 infants at six months old. The infants were shown two sets of eight photographs, one with spiders and flowers, the others with snakes and fish – chosen for their aesthetic similarities.

Parents stayed with their children, wearing opaque glasses so they wouldn’t respond to the stimuli. An infrared eye tracker was then used to monitor the dilation of the infants’ pupils as they were shown each photo.

With the first set of photos, there was an average dilation of 0.14 millimeters for the spiders, and just 0.03 millimeters for the flowers. For the spiders and fish, both were 0.16 millimeters. The researchers weren’t clear on what caused this, suggesting they may have been more excited by the animals (“hinting at a possible ‘life detector mechanism,”, they say), or it could have been something else.

To find out, they focused solely on snakes and fish in the second experiment, with 32 infants aged six months used this time. The experiment was essentially the same as the first one, except this time only the images of snakes and fish were used.

On this occasion, the snakes yielded an average pupil dilation of 0.29 mm, while the fish came it at just 0.17 mm. As a result, the researchers suggest there may indeed be an innate fear of spiders and snakes.

“Our results support the notion of an evolved mechanism that is sensitive to spiders and snakes,” they write. “Six-month-old infants react with increased physiological arousal to these ancestral threats compared to non-threatening control stimuli.”

Of course, there are some caveats. For one, the sample size was quite small. They also note that it might be better to use more similar images in future, like a spider and something that looks like a spider.

Also, it’s not super clear what pupil dilation actually means. “It is difficult to interpret some of the characteristics of the response,” the authors note. And they add that they can’t be certain the infants hadn’t been exposed to snakes or spiders before.

There does seem to be other research supporting these findings, though, so it is interesting at any rate.