Answers are grounded in this paper and cite their sources.
Investigates reading comprehension of causal and concessive relations, indicating flexible processing based on connectives.
This study investigates the online processing of causal and concessive discourse relations in authentic German sentences using eye tracking during reading. Based on the causality-by-default hypothesis, we examined whether connectives facilitate sentence comprehension differently depending on discourse relation type. Participants read corpus-based sentences either containing or omitting causal (weil “because”) and concessive (obwohl “although”) connectives. Early and late eye-movement measures were analyzed in a critical region immediately following the connective. Results revealed a robust and sustained facilitative effect of explicit connectives across all reading measures, with faster processing in sentences containing a connective compared to connective-free versions. With respect to discourse relations, the findings provide some degree of support for the causality-by-default hypothesis: Causal relations showed a clear processing advantage at the beginning of the experiment. However, this advantage did not persist as a stable linear effect. Instead, pronounced nonlinear adaptation effects were observed: Concessive relations initially elicited greater processing difficulty but were rapidly accommodated, leading to convergence with causal relations over time. These findings suggest that causal default expectations are highly flexible and quickly updated in response to distributional properties of the input, reader experience, and task demands. At the same time, they confirm a routinized use of connectives in discourse processing.
No takes yet. Share an insight, caveat, or question.
Frank et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: