As someone who writes murder mysteries (see  https://www.brianclegg.net/fiction.html ) I also love reading  them and watching them on TV, but perhaps because of spending time thinking about book plots, I've noticed that sometimes the relationship of the reader/viewer to the medium makes it possible to get clues that the characters can never access. In TV mysteries, for example, there is always the 'well known actor' syndrome. This says that actors who are famous are likely to have an important part in the proceedings, as they wouldn't be hired just to do a bit part. Then there's the weird placement effect. What we are shown on TV or read about in a book is carefully controlled. So although we have to deal with red herrings, the fact that something is mentioned that doesn't need to be alerts us to its possible significance. However, I've just come across an even more meta* example of a piece of information providing the solution to a whodunnit where not only

A little while ago I had the pleasure of giving a talk at the Royal Institution in London - arguably the greatest location for science communication in the UK. At one point in the talk, I put this photograph on the screen, which for some reason caused some amusement in the audience. But the photo was illustrating a serious point: the odd nature of mirror reflections. I remember back at school being puzzled by a challenge from one of our teachers - why does a mirror swap left and right, but not top and bottom? Clearly there's nothing special about the mirror itself in that direction - if there were, rotating the mirror would change the image. The most immediately obvious 'special' thing about the horizontal direction is that the observer has two eyes oriented in that direction - but it's not as if things change if you close one eye. In reality, the distinction is much more interesting - we fool ourselves into thinking that the image behind the mirror is what's on ou