Abstract
Research employing psycholinguistic techniques to assess the on-line processing of collocation by native and non-native speakers has flourished in the past few years. This line of research aims (among other things) at exploring actual performance in real time as opposed to the traditional paper-and-pencil testing techniques that have been extensively employed in collocation research. The present paper reviews some of the pertinent research on the on-line processing of collocations and argues for the need for more replication studies in the area. It then looks at how two experimental studies on the topic - Millar (2011) and Wolter and Gyllstad (2011) - may be replicated in order to gain deeper understanding of the key factors behind collocation processing and to obtain more valid and generalizable results that can find their way into language teaching practice.