We've been tasked with performing a heuristic analysis for a project, and the requirements specifically include the use Nielsen's heuristics.
Although this isn't a problem for us, we usually use a reduced version of Kaniasty's criteria ( moved from the criteria of Bastien & Capin), since they contain a fairly complete criterion for accessibility.
Now I never noticed that Nielsen's heuristic doesn't include a clear heuristic criterion for accessibility (and neither does Shneidermann, although we've used "Keep Locus of Control" for accessibility).
In Nielsen's, the criterion that comes most readily to mind is # 6, "Recognition rather than recall."
So my question is: is this the right one? Or is there a better rule to group all accessibility concerns?
The Nielsen guidelines are so broad and vague that almost any guideline could fall under these accessibility criteria, so I'm not sure it's the right one.