This summer I read Shaun Nichols's excellent new book, Bound: Essays on Free Will & Responsibility (2015, OUP). It includes systematic discussion of the relevant experimental philosophy, as well other empirical research, and how this relates to the problem of free will and related topics (including responsibility and punishment).
I wrote a short (400-word) book note on Bound for the AJP (draft here), which prompted my further thinking and motivated a blog post. You can read a more detailed and comprehensive summary by Manuel Vargas in NDPR that raises some nice worries for Shaun. Here I'd like to summarize just one key component of Bound and raise a distinct worry for it (despite agreeing with many other claims made in the book).
This key component is Nichols's discretionary view. It states roughly that, assuming our actions probably are determined by the past and laws of nature, there are contexts in which it's true that we have free will but also contexts in which it's false. Can we really have our cake and lack it too?