For a pilot to have 10 hours experience on type, let alone 1000+ hours experience, and not be familiar with a critical S76 system such as the autopilot (or so it seems), and not safely and optimally configuring the S76 for instrument flight each and every time such flight is necessary, is quite remarkable. What is also plainly evident is it seems to be a deficiency of knowledge more common than one might expect. It seems to me the primacy of this accident is more relevant to lack of systems basic knowledge and application rather than lack of instrument recency. One might be able to demonstrate satisfactory practical standards in a Robinson on foggles with climb, descend, turns, intercepts, etc, but go try and do that in a S76 the following day in cloud without understanding the operation of its autopilot, it probably 'aint gonna work. Or it might work once, but not the next time. Eventually you're gonna get bit.