Abstract

Although individual differences have been documented throughout the language system, it remains unclear whether native speakers vary in the efficiency with which they process naturalistic sentences. In the current study, we capitalized on the fact that language comprehension is constrained by the rapid and transient nature of incoming input and used speed pressure to reveal individual differences in sentence comprehension that would otherwise be difficult to detect. Two hundred native English speakers completed speeded listening and speeded reading tasks in which temporal pressure was systematically increased and comprehension was assessed using semantic plausibility judgments. Performance showed substantial individual differences in both tasks, and scores on the two tasks were highly correlated, suggesting a modality-independent source of variance in sentence comprehension efficiency. We also examined whether this variability was associated with domain-general processing speed, short-term memory, and print exposure. Domain-general processing speed was a strong predictor of performance in both tasks, suggesting that the speed of lower-level processing constrains successful sentence comprehension. Print exposure predicted performance in both reading and listening, consistent with the view that reading experience strengthens lexical representations in ways that benefit comprehension across modalities. Together, these findings indicate that temporal pressure can reveal meaningful variation in sentence-comprehension efficiency and suggest that native speakers differ not only in the linguistic knowledge they possess, but also in the efficiency with which that knowledge is deployed during online comprehension.

Abstract

“List as many words as you can that start with M.” The verbal fluency (VF) task is simple, yet even a typical university student only manages to produce about 15 words within one minute, and there is substantial variability around this mean. We investigated the individual differences in linguistic and domain-general skills that support verbal fluency and lead to this variability. In particular, we focused on the role of linguistic knowledge, which was calculated through six tests capturing vocabulary, amount of reading, and knowledge of grammar rules. We compared this to the role of three key domain-general skills: working memory (WM), processing speed, and nonverbal reasoning (IQ). 571 young native Dutch speakers completed semantic (category) and phonemic (letter) VF tasks, along with an individual differences test battery. We assessed the VF trials for the total number of correct words as well as two temporal variables, the time to first response and the time by which half of the responses have been produced. Linear mixed effects models revealed that linguistic knowledge was the strongest predictor of performance across all scores, and it had a larger effect than the domain-general skills. We discuss the role of linguistic knowledge and how it shapes the mental lexicon, concluding that better language skills lead to more efficient and detailed associative links between words that share semantic or phonological/orthographic features. We also provide updated norms for VF performance in Dutch and practical suggestions for using the task.