Kalsi, T.K
(2023)
Investigating the age differences in good enough processing: eye-tracking evidence from young and older adults.
PhD thesis, University of Nottingham.
Abstract
Shallow processing and good enough processing are two theoretical accounts which advocate that a full syntactic and semantic analysis may not always be accomplished, instead readers often rely on heuristics, such as context and word knowledge (Ferreira et al., 2002; Sanford & Graesser, 2006; Sanford & Sturt, 2002). This type of processing often leads to readers establishing a shallow, underspecified or ‘good enough’ mental representation of text. Good enough processing has been widely investigated in a young adult population. Another particularly interesting population to explore good enough processing in are healthy older adults because across the lifespan, elements of crystallised intelligence, such as world knowledge and vocabulary expand (Paterson et al., 2020; Salthouse, 2009). However, elements of fluid intelligence, for instance, memory can deteriorate across the lifespan (Daneman & Merikle, 1996; Salthouse & Babcock, 1991). Coupled together, older readers may be more inclined to rely on heuristics, such as world knowledge than their younger counterparts to compensate for deterioration of elements of fluid intelligence (Christianson et al., 2006). This thesis comprises of four eye-tracking experiments which examine the online processing of young and older adults whilst reading sentences that could bias them to a good enough interpretation at the word-, sentence- and wider discourse-level. In Experiment 1, sentences containing hard-to-detect anomalies were presented to assess shallow processing at the word-level. In Experiment 2, sentences containing temporary syntactic ambiguities were presented and in Experiment 3 plausible and implausible sentences in active or passive sentence structure were used to assess good enough processing at the sentence-level. Finally, doubly quantified sentences can create semantic ambiguity and this was exploited to examine the discourse-level. There were three key findings from this thesis. Firstly, in contrast to the prediction of the author, older adults were no more likely to engage in good enough processing than younger readers (Experiments 2 and 4). Instead, in Experiments 1 and 3, older adults had higher accuracy scores (which were used as an index of good enough processing), therefore, suggesting they engaged in good enough processing less than their younger counterparts. Eye tracking results also revealed that young and older adults may use different strategies during online processing of ambiguities but ultimately achieve similar levels of comprehension (Experiments 2 and 3). The relationship between eye movements made during online processing of text and comprehension accuracy was examined in Experiment 1, 2 and 3. Another key finding was that various eye movements were found to predict good enough processing, especially regressions. Generally an increase in regressions was identified as a predictor of less good enough processing in all three experiments. The conclusion from this thesis is that whilst shallow and good enough processing accounts can account for some of the findings, results from Experiments 1, 3 and 4 suggest that these frameworks cannot account for all of the findings.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of This is my second submission, following My VIVA which I passed with minor corrections. I have made theses corrections and amended. Please see pdf attached]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "This is my second submission, following My VIVA which I passed with minor corrections. I have made theses corrections and amended. Please see pdf attached")