This was one the first papers I published and resulted from my undergraduate honours project at the University of Bristol. We’re all familiar with the saying that an optimist sees the glass as half full while a pessimist sees it as half empty but, remarkably, there’s a growing body of evidence that suggests this is literally true: that your mood and expectations about the world can actually influence your perceptions.

 

In this study we set out to look at whether anxiety levels affected how undergraduate students responded to ambiguous stimuli. A very large proportion of psychological studies use undergraduate students as subjects, because a very large proportion of psychological studies are done in universities where undergraduate students are plentiful and easily bribed or coerced into taking part. You might wonder if undergraduate students are representative of humanity as a whole and whether this might affect the reliability of the results and you’d be very wise so to do. It’s a big problem, but I was an undergraduate myself and didn’t have the resources to try to tackle huge systematic problems in my field. I’m afraid an awful lot of decisions in science are made on the basis of what’s achievable with the time and resources we have, rather than what’s the best way to do things.

 

We used a standardized questionnaire to assess anxiety levels in our volunteers and then set them a computer task in which they were shown a line with a smiley face at one end and a frowny face at the other. The line then disappeared and was replaced with a cross, and the students had a few seconds to decide which end of the line it was closest to.  If they correctly identified te cross as being closest to te smiley, we “rewarded” them by showing a photograph of a cute animal; on the other hand, if they said the cross was closest to the frowny face and were wrong, we “punished” them with a picture of a scary predator like a shark baring its teeth.  We did a fun preliminary study in which we used a skin conductance response machine - essentially a lie detector machine (although bear in mind that lie detection is bullshit pretend science) – to demonstrate that people did actually have a physiological response to the scary predators!

 

The crucial thing about this odd task is that selecting the frowny face never results in either reward or punishment – either the cross really was closest to the frowny and you avoid your punishment, or it was closer to the smiley and you miss out on the cute, fluffy reward picture, but don’t get punished for your mistake. Choosing the smiley face is a high-risk, high-reward action: if you are right, you get the reward but if you are wrong you will be punished for it.

 

In most cases, the cross was unambiguously closer to one end or the other, but every now and again we sneakily put a cross dead center. What we found was that volunteers who had higher anxiety scores were more likely to play it safe and report that the cross was closer to the frowny face, avoiding any risk of punishment but also ensuring they would go unrewarded. Less anxious students were more likely to expect to be rewarded and report that the cross was closer to the smiley face. (Incidentally, don’t worry too much about these anxious students; in fact, anxiety scores in our sample were lower than generally found in the population as a whole. They were relatively more anxious than their peers, but still not especially anxious).

 

One of the leading theories about why emotions evolved is that they aid decision making, mediating physiological and behavioural responses to the complex web of threats and opportunities that animals face every day. In natural situations, it is rare to have full information about the risks involved in any given situation. An animal that is too bold and ignores dangers is likely to be quickly eaten, but one that is too cautious may avoid hazards but will also pass up opportunities to feed, mate and so on. Every individual much find a balance that allows them to take advantage of opportunities without putting them in too much danger. Experiments like ours suggest that a state of anxiety might play a role in temporarily altering that balance to favour caution – for example, in the wake of a close escape from a predator it is rational to err on the side of caution – and that it might do so by affecting an animal’s interpretation of ambiguous sensory experiences,  making the animal more likely to anticipate danger and less likely to anticipate opportunity.

 

Interestingly, this experiment was inspired by a study into rat behaviour and my colleague Cwyn Solvi more recently carried out a conceptually similar study on bumblebees, showing that they too might perceive ambiguous stimuli differently depending on recent experience. This is a slightly unusual, but welcome change to the more common pattern in which studies on humans are adapted to ask the often-unedifying question of whether animals can do what humans do. In this case, evidence seems to be accumulating to suggest that emotional states are common across the animal kingdom and probably serve similar roles in assisting decision making under conditions of uncertainty and risk.