Encoding variability and the concreteness effect in paired-associate learning

In an attempt to explain the concreteness effect in PA learning, encoding variability (as measured by contextual variety) was manipulated factorially on the stimulus and response sides within abstract and within concrete noun pairs. The results showed a strong concreteness effect, with increasing encoding variability having no effect on the stimulus side, and a direct positive effect on the response side. The applicability of these results to the encoding-variability hypothesis was discussed.