The Processing of English-Turkish (False) Cognates: Evidence from a Backward Lexical Translation Task

Abstract

Cognate facilitation and false cognate inhibition effects have been tested in various language pairs with different experimental tasks and participant profiles so far. However, studies focusing on the recognition or production of (false) cognates are nearly absent for Turkish-English despite the prevalence of these words. Thus, using a backward lexical translation task (from L2 to L1), this study aimed to investigate whether cognate facilitation and false cognate inhibition effects could be observed in Turkish-English by testing 50 adult Turkish L2 speakers of English. The materials were made up of cognates, false cognates, and controls. The effect of L2 proficiency was also manipulated by dividing the participants into two proficiency groups (high vs. low) based on OPT scores. Also, the role of morphology was introduced by using mismatch items (polymorphemic in L2 but monomorphemic in L1). The findings showed a robust cognate facilitation and false cognate inhibition but no significant effect of L2 proficiency. The role of morphology was not conclusive and came with its limitations. These results provided supporting evidence for the language non-selective view and pointed towards the presence of these effects irrespective of language, task or participant profile. Also, a compelling need for measuring proficiency using multiple measures emerged.