We investigated the processing of free combinations and collocations, which differ in lexical fixedness. Eye movements of L1 and L2 speakers of Italian were recorded while they read target sequences embedded in context. The degree of lexical fixedness and leaners’ proficiency were found to affect the processing of word combinations.
The present study explores the processing of Verb+Noun free combinations and collocations by first language (L1) and second language (L2) speakers of Italian. Phraseological models (Cowie, 1981; Howarth, 1998; Mel’cuk, 1998; Ježek, 2005; Masini, 2009) – which we adopt in the study – distinguish word combinations in terms of lexical fixedness, which refers to whether or not a word within a phrase can be substituted with a synonym. Sixty Verb+Noun free combinations and collocations (the original condition) were embedded in sentence contexts. In the variant condition, the verb within the phrase was substituted with a synonym. Same contexts were used. We sought to answer the following questions: Do L1 and L2 speakers process free combinations and collocations in the two conditions differently? Does learner proficiency affect the processing of word combinations? L1 and L2 speakers of Italian of different proficiency levels read target combinations in the original and variant conditions, while their eye movements were recorded. The results of mixed-effects modelling showed that L1 speakers read original and variant free combinations in a comparable way, whereas they read original collocations significantly faster than variant collocations. In addition, L2 speakers’ proficiency affected learners’ reading times. Both lower and higher proficiency learners read original and variant free combinations in a comparable way. However, while lower proficiency learners read variant collocations faster than original collocations, higher proficiency learners read original collocations faster than variant collocations. Our results suggest that the degree of lexical fixedness does indeed affect the processing of word combinations in L1 and L2, lending support to phraseological models (Cowie, 1981; Howarth, 1998; Mel’cuk, 1998; Ježek, 2005; Masini, 2009). Our results further point to the conclusion that higher proficiency L2 learners may process word combinations in a way similar to L1 speakers.
