Do Anglo-Saxons free-ride more?

We report on experiments replicating the Partners and Strangers design and we find some evidence that may help to accommodate previous diverging partners/strangers results. This finding comes out of a preliminary investigation into whether nationality makes a difference as far as free-riding is concerned. We seem to have identified a strong effect on behaviour resulting from national differences (which, in turn, presumably reflect cultural and sociological differences between subject groups), both on average contribution and, possibly, on the attitude toward playing in stranger or partner sessions. This seems to point out the existence (and relevance) of different social norms in different social and cultural contexts. We have also explored the influence of a change to Public Bads and we have further investigated the Restart Effect. Our results both in the final rounds and especially in the restart indicate that learning requires much more time and trials than expected, and does not seem to provide a full explanation of the observed behaviour. (C)1997 Elsevier Science S.A.