Rolf Zwaan, ‘A Replication With A Wrinkle’, https://rolfzwaan.blogspot.co.uk/2018/01/a-replication-with-wrinkle.html

Oh dear oh dear I had planned to be posting about some of the stuff mentioned here by now, but it’s brewing a bit slowly. I don’t like to leave it too long between posts, so here’s one about some old friends.
      Kidd and Castano are getting to be legends on this blog, and rightly so. Their 2013 paper, arguing that reading literary fiction enhances our capacities to empathise with others, looks like a massive validation of my life choices. And yet, reluctantly, I find it hard to believe. So I have kept track of some ups and downs in its subsequent reception (e.g. here and here, and there are onward links to follow).
      This blog post by Rolf Zwaan links to two versions of a report on his attempt to replicate one key effect, where a difference was observed between literary fiction and non-literary fiction. In the Kidd and Castano paper, there was a difference between the empathy-bonus that resulted from reading proper serious fiction, and the one from just any old fiction. Wow, but don’t ask me to draw the line. In the first paper about his replication attempt, in Dutch, Zwaan et al. reported no effect of that sort; but in the second, written in English and after further analysis, they reported an effect.
      A strange and arresting scenario! The reason for the discrepancy, Zwaan explains, was a decision about what to do about the definition of reading. It seems reasonable to say that only those who had spent a certain amount of time looking at a passage could be described as having read it, and only those people should be included in the study. Zwaan et al. had set one threshold, Kidd and Castano another (actually a lower threshold, requiring less time per page); the former researchers applied the latter’s threshold to their own analysis, and got a new result.
      This is all very thought-provoking: about the subtleties of replication attempts, about the persistence of Kidd and Castano’s conclusions, but most of all, to me at least, about how you decide someone has read something properly. How long does it take to read a novel, after all? What relationship is there between the time taken and the benefits acquired, whatever they may be? The difficulty of answering that question seems like a reason why experience of fiction, or of any art, is an awkward basis for experiment. It seems right to hold each subject to some basic standards, but time, and place, and so on, may be experienced very differently by different people.

*

Also in replication news, I found this article interesting, a reflection on the current wave of replication attempts. It’s interesting, seeing different perspectives on the question ‘what are we doing to ourselves?’.