ABSTRACT
Background
The use of prediction can aid language comprehension through preactivation of relevant word features. However, predictions can be wrong, and it has been proposed that resolving the mismatch between the predicted and presented item requires cognitive resources. Older adults tend not to predict and instead rely more on passive comprehension. Here, we tested, using an intraindividual approach, whether older adults consistently use this less demanding processing strategy while reading or whether they attempt to predict on some trials.
Methods
We used a cross-task conflict paradigm. Younger and older participants self-paced to read sentences that ended with either an expected or unexpected word. Each sentence was then followed by a flanker stimulus that could be congruent or incongruent. We examined responses within and across the two tasks.
Results
Unexpected words were in general read as quickly as expected words, indicating that typical processing of these words was similar. However, for both younger and older adults, there was a greater proportion of very slow trials for unexpected words, revealing different processing on a subset of trials. Critically, in older adults, these slowly read unexpected words engaged control, as seen in speeded responses to incongruent flanker stimuli.
Conclusion
Using a cross-task conflict paradigm, we showed that older adults are able to predict and engage cognitive resources to cope with prediction violations, but do not opt to use these processes consistently.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1. We also tested a group of students from the University of Illinois at Urbana-Champaign who received credit for participation. However, this sample did not appear to be sensitive to the self-paced reading manipulation, as there was no difference in the reading time distributions between expected and unexpected words. As a result, we believe that studying interindividual and interindividual differences in this group would not be of interest, as no prediction took place in the first place. Nevertheless, Appendix A contains the same analyses for this group as run for the younger and older samples in the main paper.
2. Due to a technical incompatibility between jsPsych and Pavlovia, we were not able to monitor screen leaving behavior in the young adult Prolific sample. In brief, the function used to record this information (jsPsych.data.getInteractionData) is not fully supported by Pavlovia at this time. However, there’s strong evidence that our data are reliable because of our described data cleaning methods.
3. To address any concerns that the data pattern might have been affected by differing cognitive abilities in the subset of older adults who were not able to complete the working memory performance measure, we re-ran the intraindividual analyses in the older adults sample excluding those eight participants (N = 40). The analysis with these participants excluded yielded the same result pattern. For reading times, there is a significant effect for τ (ß = 61.06, SE = 24.09, t = 2.64, p < .05) and not for µ (ß = 17.76, SE = 14.56, t = 1.22, p = .023). A greater proportion of slow trials was thus found for unexpected words rather than for expected words. Importantly, the expectancy effect found for slow reading trials carried over to the following flanker trial, as we again found a significant interaction between reading expectancy and flanker congruency: ß = -0.052, SE = 0.020, t = -2.58, p < .01. Incongruent flanker trials were responded to more quickly after an unexpected word (918 ms, SD 232) than after expected words (950 ms, SD 266), whereas congruent flanker trials were responded to more slowly after an unexpected word (747 ms, SD 250) than after an expected word (716 ms, SD 217). This pattern fully mirrors the complete database.