Spelling-to-sound correspondences affect acronym recognition processes

Abstract

A large body of research has examined the factors that affect the speed with which words are recognized in lexical decision tasks. Nothing has yet been reported concerning the important factors in differentiating acronyms (e.g., BBC, HIV, NASA) from nonwords. It appears that this task poses little problem for skilled readers, in spite of the fact that acronyms have uncommon, even illegal, spellings in English. We used regression techniques to examine the role of a number of lexical and nonlexical variables known to be important in word processing in relation to lexical decision for acronym targets. Findings indicated that acronym recognition is affected by age of acquisition and imageability. In a departure from findings in word recognition, acronym recognition was not affected by frequency. Lexical decision responses for acronyms were also affected by the relationship between spelling and sound-a pattern not usually observed in word recognition. We argue that the complexity of acronym recognition means that the process draws phonological information in addition to semantics.

Phonological effects in visual word recognition: evidence from the processing of two types of Hebrew acronyms

“…These findings are consistent with previous studies in English indicating that acronyms automatically activate their corresponding phonological representations (e.g., Playfoot & Izura, ; Slattery et al, ; Slattery et al, ). In particular, our findings replicate and extend those reported by Playfoot and Izura (), who also showed that the recognition of a printed acronym is influenced by the way its orthographic representation is mapped onto its phonological representation. Specifically, they observed that ‘phonological’ acronyms (NASA) were recognised faster than ‘letter‐by‐letter’ acronyms, irrespective of whether the ‘letter‐by‐letter’ acronym was orthographically legal (HIV) or not (BBC).…”

“…In addition, ‘phonological’ and ‘lexical’ acronyms were balanced in terms of objective printed frequency, bigram frequency and length: First, printed frequency estimates were obtained by running the Google search engine. Specifically, the log transformation of the number of hits returned for each acronym was used as an index of printed frequency (for a similar procedure, see Playfoot & Izura, ). There was no significant difference in printed frequency between ‘phonological’ and ‘lexical’ acronyms (Phonological: M = 5.559, SE = .101; Lexical: M = 5.779, SE = .202; t (38) = −0.972, p = .339).…”