Having recently joined the happy hordes of mp3 player owners, our household has been getting an object lesson in the nature of random events. For those who have yet to succumb to the charms of these amazing little gadgets, they can hold thousands of songs in memory and play them back in many different ways. You can, for example, just play a single album, or make up a playlist of songs for a party, or to encapsulate a particular mood.

But one of the most popular features is Shuffle, which plays all the songs in random order. Turn the player off, and it reshuffles and starts again. This produces some weird results. We have lots of Neil Young on our iPod Nano, but it seems almost the only song that gets played is Harvest Moon, which we’re fast going off. Pink Floyd’s 23-minute epic Echoes is a fine piece of music, but we don’t want to hear it every time we turn on the player, as seems to happen. Very often it plays two songs by the same artist with only one or two tracks in between by someone else. Once in a while a song pops up that we’d forgotten we even had.

Way back in 2004, the New York Times had an article about how people were convinced their players could detect their moods, or had favourite songs, or had simple neural networks and were learning based on a user’s history of song skipping. This perception persists.

The trouble is, random doesn’t mean what people think it does (strictly speaking, Shuffle isn’t entirely random, because it won’t play the same song twice in a single session, but with a couple of thousand songs to run through it’s near enough). One commenter on an on-line forum said her iPod definitely wasn’t playing songs at random because some songs had been played 10 times while others hadn’t been played at all. A reply pointed out that a random distribution has a lot of ‘clumpiness’, and the numbers of plays should follow a Poisson distribution. Randomly select 1000 songs from a list of 1000, and you would expect 368 songs would never be played, while three would play five times.

None of this should be news to skeptics. The human propensity to perceive order in random events lies at the base of many paranormal beliefs. First one thing happens, then another thing happens. When you notice this occurring a couple of times, it’s natural to think the two are causally related, and you tend to forget the times the pattern isn’t followed. Only yesterday my brother-in-law was expressing concern about his son travelling unaccompanied on an Airbus, given their recent spate of crashes, despite the lack of any common factor in these incidents. Being able to detect patterns and make plans accordingly is a valuable survival skill for humans, but just as often it can lead us astray.

Recommended Posts