Keep calm and move on: ‘Good-enough’ processing of clausal constructions in Korean

Abstract

The good-enough processing account maintains that the human parser favours cognitively less effortful parsing (a heuristic stream) over deep and costly parsing (an algorithmic stream) in real-time processing. This study investigates how the two parsing routes work for sentence comprehension in Korean, an understudied language for this topic, through two self-paced reading experiments. In Experiment 1, focusing on the suffixal passive, native Korean speakers show no significant difference in reading times at the verb region between the verb-initial and verb-final conditions. However, we find a significant by-condition difference in the spill-over region, with longer reading times for the verb-initial condition compared to the verb-final condition. Experiment 2, examining the morphological causative, we find constant reading times across both word-order conditions including at the spill-over region. These findings suggest an interplay between the two parsing routes, with heuristics taking precedence over algorithms when disequilibrium arises. This study's findings also shed light on how the operations of these two parsing streams are modulated by language-specific properties (e.g., word order, case-marking, verbal morphology) during sentence comprehension.