© 2020 The Authors Is persistent soil organic matter (SOM), characterised by an old age and long-turnover time, more or less sensitive to changes in temperature than fast-cycling, recent SOM? Largely due to our limited understanding of the mechanisms of SOM formation, this question remains controversial. Laboratory incubation studies, through sieving the soil, may create conditions in which substrate accessibility is modified. The recent recognition of SOM accessibility as a defining factor of SOM persistency calls into question conclusions from these studies. Previously, in a study using root exclusion plots of increasing age, we showed in the field that the temperature sensitivity of SOM decomposition decreased with increasing persistence of SOM (Moinet et al., 2020), in opposition to many laboratory incubation studies. Here we sampled soils from the same root exclusion plots and conducted a laboratory incubation experiment to test the hypotheses that (i) the relationship between temperature sensitivity and SOM persistence is inverted as compared to the field, and (ii) the discrepancy is due to sieving the soil. We showed that, in the laboratory, the relationship was indeed inverted, with the temperature sensitivity being higher for the old root exclusion plots. However, sieving the soil at 2 mm did not affect estimates of the temperature sensitivity of SOM decomposition, suggesting that discrepancies between field and laboratory estimates are unlikely to stem from artificially modified substrate accessibility due to sieving.