Nicely said.

It’s true that there is some additional processing involved switching from written instructions and the visual inputs and outputs they describe – maybe it is always difficult. I know in the case you described, I don’t want written instructions – I need to see it in order to visualize which way I am turning. I prefer, however, to have a person (if I am lucky enough to have a navigator) tell me which way to go rather than show me the map.

The biggest piece to focus on when deciding what medium to use is where (if anywhere) you want the maximum processing power to be spent. Text instructions require that a substantial portion be devoted to reading and processing the text, and then reacting to the information it contains. That often leaves little remaining processing power to figuring things out. Easiest way to shift the load is to reduce the words and do things visually.