Nice article from a persitent champion. I have a fascination with exploring the evidence for the maximum information rate of human learning, as expressed in bits per second. The world record performances of memory athletes provide an excellent source of data. The useful events are the ones for memorising symbols such as playing cards or binary or decimal digits, because these allow us to precisely state the number of information bits per symbol. As the events are timed we can therefore derive a figure in bits per second. As the top performers’ times are remarkably similar, it suggests that we might be observing a limit of human mental physiology (in a similar way that champions runners are displaying something of the limits of human physical physiology.
The task of memorising a single pack of cards (in around half a minute) is almost ideal for my purpose. For longer events, the memorising task absorbs a greater proportion of the time spent, so reducing the deduced bit rate. The Very short dutration memory contests (around a second) are of limited use to me because we absorb a brief memory snapshot (around 50 bits) that we post-process. Simon’s performance memorising a single pack of cards is equivalent to an information rate of 14 bits per second. So thanks for a useful data point Simon.
Other mental athletes who specialise in mental arithmatic events achieve around 18 to 19 bits per second. Their task is easier as only the result of the mental calculation needs to be remembered. These figures may seem suprisingly low compared with our electronic communications, but are remarkably similar to the information rate of language (reading or listening), when the redundancy in language is taken into account. They are also similar to the maximum information rate of monitored human skills, but that is another story.