Superstitious mice

One of the nicest things about being part of a large University is that, if you can drag yourself away from your desk long enough, you get the opportunity to attend some pretty amazing guest lectures discussing cutting edge scientific findings. Last week I sat in on a particularly engaging talk given by a researcher from UCL on ‘superstitious’ mice. Although the title was a bit confusing, leaving me with images of mice refusing to leave their beds of Friday the 13th and saluting whenever they saw a magpie, the actual research gave an amazing insight into how the brain balances its own internal prejudices with its actual experiences of the world. The ‘take home message’ of the talk was that mice and men don’t always believe what they see and, on occasions, will act on what they expect rather than what is actually in front of them. The research behind this finding is both elegant and eye opening and I will attempt to do it justice here:

Research in this lab was not initially intended to test the ‘superstitious’ nature of laboratory mice. The lab was instead interested in how well mice could distinguish between two images and how similar these needed to be before the animals became confused; research such as this is important for understanding how the visual system works. The experiments relied on the notion that mice can be taught to perform specific tasks in response to different commands (similar to training a dog…just on a smaller scale). Mice were kept in special cages with two separate treat dispensers and were taught to watch a screen which flashed one of two images. Each image corresponded to a different treat dispensers, I.e. when image 1 appeared the mouse could get a treat from dispenser 1 and when image two appeared the mouse could get a treat from dispenser 2. To make the task a little bit trickier the scientists sometimes manipulated the images making them harder to tell apart, with the aim of confusing the mice.

This figure expresses the concept behind the test however the images used in this research were not coloured dots.

What was particularly impressive about this experiment was that the scientists worked with two groups of mice, one which performed the behavioural task and another which watched the same images whilst the researchers recorded activity from the visual areas of their brain. This meant that researchers could compare how well cells in the brain responded to the different images with how well the mice performed on the task. Now this is where the findings get interesting! The mice weren’t very consistent when it came to performing the task; meaning that some times they would perform well, even when the images were similar, whilst other times they seemed to be unable to recognise even the clearly separated images. However, when the researchers looked at the corresponding brain activity they found that the visual cells they recorded from were consistently good at differentiating between the images. This caused some serious head scratching as the scientists tried to work out how, when the mouses’ brain could distinguish the images, the mouse itself sometimes behaved as though it could not.

What the group found was that mice based the decision of which treat dispenser to visit, not only on the image they saw, but also on their past experiences – taking into account what choices had previously lead them to receive a reward or not. The mice tried to assign a pattern to the task making assumptions based on what they had already experienced, then combined this internal prediction with what they actually saw. Amazingly these internal predictions (which the researchers called superstitions) could be strong enough to win out over the animals own vision causing it to make the wrong choice. We can perhaps understand this behaviour better by thinking about the times in our lives when we assign patterns to things which are in fact entirely random. Take for example the national lottery. There was a time when you heard news reports speculating on lottery number, making the assumption that since a certain number had not been drawn for weeks it was ‘due’ whilst another number which appeared more regularly may be less likely to appear again. Of course the lottery draw is entirely random, meaning that the frequency of certain numbers being drawn on previous weeks has no influence on what the current draw will be. However, this did not stop us speculating and assigning our own patterns to the draws. It seems that the brain just loves to create patterns!

However, we would never be silly enough to ignore what our eyes were telling us in favor of a ‘superstitious’ belief, would we? Well… before you sit back, quietly mocking the poor mice for being slaves to their internal pattern maker, it is worth noting that they are not the only species to fall foul to the problems of over thinking a scenario. Yes you guessed it, it appears that we do this too! A follow-up experiment used a similar protocol with people and amazingly found the we also sometimes make the wrong decision even though our eyes are obviously capable of telling us we are wrong. So it seems that when it comes to both mice and men our superstitions can occasionally get the upper hand!

Post by:  Sarah Fox

For original work see here (subscription necessary to view full article)

Comments are closed.

Share This