I've seen this article before, and every time I've said to myself, "Gosh, how non-intuitive. I'll have to remember that" Today, I had a little time and a littl

I've seen this article before, and every time I've said to myself, "Gosh, how non-intuitive. I'll have to remember that" Today, I had a little time and a little more skepticism, and said "Gosh, I wonder if it's actually true". I, too, replicated the experiment, and for me, at least, it's completely wrong. Conditions: Microsoft Word 2003, 12-point default, whole page view, 15-inch laptop screen, 1200x1920 resolution Text: the paragraph of Peter Deutch's letter Keyboard: 128 seconds Normal mouse: 167 seconds (30% slower) Lenovo Trackpoint mouse: 228 seconds (78% slower) Yet another disadvantage of the mouse (and Trackpoint) was that the task required far more attention and generated more anxiety: it was difficult, error-prone, clumsy, and frustrating--I kept wanting to stop and rest my fingers. That is yet another reason that the mouse is a poor choice for this kind of task. With the keyboard, I could work much longer without anxiety. I hypothesize, but have no idea of how to test, that this greater comfort results from not relying on eye-hand coordination for the details of the task, but rather being able to coordinate with a mental model of knowing what the cursor keys will do without having to see it happen on the screen.