Abstract
The contribution of lexical and semantic knowledge to verbal short-term memory (vSTM) span is explained by language-based models, assuming that vSTM is deeply grounded within the linguistic system with to-be-remembered items being activated in a non-strategic and automatic manner. However, direct evidence for a non-strategic account of lexical and semantic contributions to vSTM span is scarce. In this study, we assessed the influence of several types of long-term linguistic knowledge (lexicality, lexical frequency, semantic similarity and imageability) on vSTM using a fast encoding running span procedure preventing any strategic processes during encoding. We observed reliable effects of lexicality (words vs. nonwords, Experiment 1), lexical frequency (high vs. low frequency words, Experiment 2) and semantic similarity (related vs. unrelated lists, Experiment 3) on running span performance. However, word imageability (high vs. low imageability words, Experiment 4) did not consistently impact running span performance. Experiment 5 showed that the imageability effect only appears in standard immediate serial recall conditions which do not prevent list-strategic encoding. This study provides novel evidence for linguistic accounts of vSTM by demonstrating a robust impact of lexical and surface-level semantic knowledge on vSTM in non-strategic, fast-encoding conditions.
Kowialiewski, Benjamin