Relational and Item-specific Information: Trade-off and Redundancy

Abstract

In two experiments, the trade-off and the redundancy hypotheses for relational and item-specific information proposed by Hunt and Seta (1984) were tested. Lists consisting of categories of varying sizes were presented under categorising instructions and under pleasantness-rating instructions. Memory was tested in free recall and a recognition test. Different measures were used for relational and itemspecific information. According to the trade-off hypothesis, the amount of relational information should increase with increasing category size, and at the same time, the amount of item-specific information should decrease. This hypothesis could not be confirmed. Whereas the amount of relational information increased with increasing category size, the amount of item-specific information did not decrease. The redundancy hypothesis assumes that relational and itemspecific information depend on category size only if the relevant information is not provided otherwise, as by orienting tasks. This hypothesis could not be conformed either. Rather, relational encoding is supplemented when both orienting task and the list structure focus encoding of that type of information. Item-specific encoding, on the other hand, is independent of category size and increases when the instruction focuses on it. The findings of free recall show that free recall is determined by more than the interplay of relational and item-specific information in an additive manner.