Down the garden-path and back again: Factors contributing to successful recovery from ambiguity-related misinterpretations

For most adults, understanding the meaning of words and sentences seems like an easy task. However, misinterpretations are common. More than 80% of common English words have more than one meaning – even the word “abstract” is ambiguous. The pervasiveness of ambiguity means that selecting appropriate word meanings is a fundamental skill for readers. But what happens when the language processing system selects the wrong meaning?Sentences like “The plant had been difficult to build” lead readers down a metaphorical “garden path” in interpretation: because they are likely to initially select the meaning of “plant” that first comes to mind (“botanical organism”), readers need to initiate appropriate reinterpretation processes once they realise that this meaning is not compatible with “build”. This thesis addresses two gaps in the literature on the recovery from such misinterpretations: Are the processing costs that comprehenders experience when they need to reinterpret a sentence consistent across a) tasks or stimulus presentation formats with different processing demands, and b) individuals?Results from the present thesis revealed that readers initiated reinterpretation procedures at an earlier point during processing when an explicit task required them to understand the sentence in detail (Chapter 4). Additionally, reinterpretation costs to brain responses were found to be exaggerated in a visual word-by-word presentation format compared with listening or whole-sentence reading (Chapter 5). Individual differences investigations showed that readers with greater vocabulary knowledge tended to be more sensitive to errors in processing, and able to adapt their reading behaviour on-line (Chapters 2 and 3). These findings support theories that view language experience and the development of robust knowledge structures as central to language processing, and highlight the importance for both theorists and experimentalists to carefully consider the influence of task-related and format-related processing demands on processing behaviour and comprehension.