I think it is interesting to conduct experiments to determine the channel limit of each human sense; that is, the maximum amount of information that can be conveyed by each sense. It is commonly said that 80% of the information we process comes from vision, which makes sense. For example, reading text is much more efficient than hearing text, which is more efficient than reading Braille. However, the statistic itself is quite useless without knowing the information capacity of sight.
Actually, in hindsight, I am beginning to doubt the 80% figure, as the visual channel is clearly much more capable than the other senses, not only in sheer capacity, but also in responsiveness. In contrast, taste and smell have poor responsiveness and poor differentiation ability. I am, however, very much interested in knowing the dynamic range of smell (the maximum and minimum concentrations detectable for a given substance).
Why is this important? Most likely it is not. However, it is possible to develop special assistive devices to replace damaged senses by transmitting through the other still functional senses. The fundamental transmission limits of the senses should provide a useful guide as to the full potential of the assistive device.
Knowing the transmission capacity of each sense should also allow us to design better machine-human interfaces. I posit that an interface providing visual, aural, and tactile feedback would be better than one that provides the same capacity using only the visual channel. In particular, the additional sensory channels may be used to provide interrupts or higher priority signaling, as they may not use the same attentional channels as vision does.