Serial reproduction of traumatic events: does the chain unravel?

Our goal in this paper is to show that a careful analysis of recall accuracy within a serial reproduction chain can add to a detailed qualitative analysis of the reproductions within the chain. The texts we chose are based on newspaper reports concerning current events, which are far from being mundane: reports of tragic events, even traumatic events. The participants were 216 students who were randomly assigned to 54 four-person reproduction chains, 18 for each of three 160-word texts. The reproduction chain is highly dependent on the recall accuracy of the first generation (i.e., the first participant in the reproduction chain). Thus, we argue (and show) that there should be a qualitative difference between chains starting off with a high level of recall accuracy from those starting off with a low level of recall accuracy. Our data-analytic approach is based on trend analysis, which we argue is an apt quantitative, holistic, dynamic, process-oriented type of analysis that is required in such research.