Probably, but you would have to map out -exactly- where the spoon was going, the size of the portion taken, create the morph for that one spoonful, then repeat it for every consecutive action. You would also have to involve texture changes and either normal mapping or out and out displacement mapping to keep the visible food surface from looking like deformed geometry.

Basically, you are getting into the same reasons you don't see very many CG characters barefoot. The amount of work required to do it well enough to keep from breaking the suspension of disbelief isn't worth it most times. Now it was worth it in 'Avatar' , as the only time you really see feet in motion is during Jake's run after getting his avatar.....which works as a dramatic device to separate the kitty cat from the chair bound marine with the busted backbone.  But those few seconds of toe wriggling took a =lot= of manhours to get that good; about the only thing as complex is hands and facial expression, and both of those take far longer, as they do more.