The idea that came to my mind when I was viewing the lectures on recursive feedback was that by its very nature, recursive feedback is multilayered. Once I give feedback to a peer on her case study (layer 1), another peer can give me feedback on my feebdback (layer 2). The process can go on, as more peers participate and build more layers. At some stage, assuming that we have reached layer (n), we also need to consider the instructor's involvement. Should this involvement put a break on the feedback layers, or should this be the beginning of a new cycle of layers? I cannot tell. But I am sure that this is an area that can be researched (if not under research already) and provide some exciting findings.