Frontiers in Psychology (Oct 2016)
Predicting lexical priming effects from distributional semantic similarities: A replication with extension
Abstract
In two experiments, we attempted to replicate findings by Günther, Dudschig &Kaup (2016) that word similarity measures obtained from distributional semantics models -Latent Semantic Analysis (LSA) and Hyperspace Analogue to Language (HAL) - predictlexical priming effects. To this end, we used the pseudo-random method to generate itemmaterial while systematically controlling for word similarities introduced by Günther et al.,which was based on LSA cosine similarities (Experiment 1) and HAL cosine similarities(Experiment 2). Contrary to the original study, we used semantic spaces created from farlarger corpora, and implemented several additional methodological improvements. InExperiment 1, we only found a significant effect of HAL cosines on lexical decision times,while we found significant effects for both LSA and HAL cosines in Experiment 2. Asfurther supported by an analysis of the pooled data from both experiments, this indicatesthat HAL cosines are a better predictor of priming effects than LSA cosines. Takentogether, the results replicate the finding that priming effects can be predicted fromdistributional semantic similarity measures.
Keywords
