Contemporary associative learning theory predicts failures to obtain blocking: Comment on Maes et al. (2016).

Abstract

In a recent article, Maes et al. (2016) report the results from 15 experiments in the blocking effect, all of which failed to replicate the basic phenomenon. Whereas Maes et al. did not dispute the reality of the blocking effect, they concluded that the effect is more difficult to obtain than what could be assumed from the literature and that we lack insight into its boundary conditions. This conclusion is incorrect, because contemporary associative learning theory both agrees with the authors' conclusion that blocking is parameter-dependent and it makes specific predictions about the experimental parameters likely to produce a small or no blocking effect. Ten of the 15 experiments presented by Maes et al. use exactly those parameters (same-modality stimuli for the compound AX), making their results completely unsurprising in the light of contemporary associative learning theory. The results from 3 other experiments are difficult to interpret because of a floor effect. A failure to replicate blocking in only 2 experiments is unsurprising and can be explained as the result of statistical variability or changes in experimental procedure. (PsycINFO Database Record