A Word or Two About Nonwords: Frequency, Semantic Neighborhood Density, and Orthography-to-Semantics Consistency Effects for Nonwords in the Lexical Decision Task

For the most part, the effects of lexical-distributional properties of words on visual word recognition are well-established. More uncertainty remains, however, about the influence of these properties on lexical processing for nonwords. The work presented here investigates the mechanisms that guide nonword processing through an analysis of lexical decision latencies for 18,547 words and 27,079 nonwords in the British Lexicon Project (Keuleers, Lacey, Rastle, & Brysbaert, 2012) using piecewise generalized mixed models (pamms; Bender, Groll, & Scheipl, 2018; Bender & Scheipl, 2018; Bender, Scheipl, Hartl, Day, & Küchenhoff, 2018). The pamm analysis of the data revealed 2 novel effects for nonwords in the lexical-decision task. First, whereas previous studies reported effects of base word frequency, the current study is the first to document a true nonword frequency effect. Second, we report effects of semantic neighborhood density and orthography-to-semantics consistency; not only for words, but also for nonwords. The effects of frequency, semantic neighborhood density and orthography-to-semantics consistency are facilitatory for words, but inhibitory for nonwords. The pamm analysis offers insights into the temporal development of the effects of lexical-distributional variables that are not available through more traditional analysis techniques and that shed new light on lexical processing in visual word recognition tasks. The implications of the reported results for models of visual word recognition are discussed.