Whether animals feel emotion, and are capable of suffering, is a question the answer to which has far-reaching implications. I recently read Victoria Braithwaite’s ‘Do Fish Feel Pain?’, a question that I didn’t worry about much until reading this book, but now bothers me a lot more. This book raised a number of quandaries I had previously not considered. One of the most pertinent of these was the measures we deem necessary for an animal’s welfare, particularly in farming practices, and how these are determined by how much we think the animal in question is capable of suffering. The ability to suffer is something that is difficult to quantify, and instead of going into details here I’d recommend reading this illuminating book instead. However, one way in which we do measure the ability to suffer is that once an animal experiences an unpleasant event it has lasting effects on that animal’s outlook on the world. For example, we know in humans that a negative event (such as pain or fear) causes a pessimistic cognitive bias, so that events after this that might usually be seen as neutral or even positive, are now viewed in a negative light.

Cognitive biases have also been found in animals like rats, dogs, sheep and starlings but until recently no one had looked to see if any invertebrate had them, presumably because we assumed that, of course, insects don’t experience pessimism or suffering. However, in a recent study published1, Bateson and her colleagues looked to see just this- would bees show a pessimistic outlook after experiencing a negative event?

Bees, whilst having their own remarkable communication system, unfortunately cannot talk to us, and so the first hurdle to overcome in an experiment like this is to know how to judge when a bee is being pessimistic in a situation. Luckily, bees are known for being great at learning, so the scientists trained bees to associate a particular odour with a sugar reward, to which they would stick their proboscis out to eat. After training the bees to associate these two things, the bees would then stick their proboscis out when they smelled that particular odour, even when there was no honey reward to follow. The bees were also trained to associate another odour with a negative outcome (instead of a high concentration sugar solution, they were given a low-concentration sugar solution or quinine which doesn’t taste very nice to bees).


After the bees were trained that one odour lead to a good reward, whereas the other odour lead to ‘punishment’ (its hard to imagine quinine being that bad, but pretend for a moment you’re a bee), they either were left undisturbed, or they were exposed to a negative experience. This negative experience was being put into a bottle and being shaken, as might happen if the bee were attacked by a predator.

These bees were then exposed to the odours as before, but now in different mixtures varying from the first odour, an equal mix of the first and second odour, through to the second odour alone. The bees that had been shaken up responded much more negatively to the odour that predicted a negative outcome than did the bees that hadn’t been shaken. Interestingly, this effect was much greater for the negative stimulus (the second odour by itself) rather than the ambiguous one, unlike what was found in studies with vertebrates.

Nevertheless, it does seem that these bees which were shaken definitely seemed to avoid the ‘bad’ odour more than the other bees. Whether this really shows that bees are capable of having a ‘glass half-empty’ outlook on the world in the same way that humans do is unknown, but it definitely leaves room for discussion.

Bateson et al. 2011 Agitated Honeybees Exhibit Pessimistic Cognitive Biases. Current Biology 21, 1070-1073.