Are there independent effects of constraint and predictability on eye movements during reading?

Abstract

Evidence of processing costs for unexpected words presented in place of a more expected completion remains elusive in the eye-movement literature. The current study investigated whether such prediction error costs depend on the source of constraint violation provided by the prior context. Participants' eye movements were recorded as they read predictable words and unpredictable alternatives that were either semantically related or unrelated in three-sentence passages. The passages differed in whether the source of constraint originated solely from the global context provided by the first two semantically rich sentences of the passage, from the local context provided by the final sentence of the passage, from both the global and local context, or from none of the three sentences of the passage. The results revealed the expected processing advantage for predictable completions in any constraining context, although the relative contributions of the different sources of constraint varied across the time course of word processing. Unpredictable completions, however, did not yield any processing costs when the context constrained toward a different word, instead producing immediate processing benefits in the presence of any constraining context. Moreover, the initial processing of related unpredictable completions was enhanced further by the provision of a supportive global context. Predictability effects therefore do not appear to be determined by cloze probability alone but also by the nature of the prior contextual constraint especially when they encourage the construction of higher-level discourse representations. The implications of these findings for understanding existing theoretical models of predictive processing are discussed.

Looking for Immediate and Downstream Evidence of Lexical Prediction in Eye Movements during Reading

“…Instead, unpredictable completions that were semantically related to the best completion received shorter total reading times and fewer first-pass regressions in the strongly constraining contexts (i.e., "sermon" in 1a), suggesting that these items were easier to integrate due to their semantic overlap with the most predictable completion (see Federmeier & Kutas, 1999;Federmeier et al, 2002;Thornhill & Van Petten, 2012 for similar ERP findings). Similar findings were obtained in a recent eyemovement study by Wong et al (2022) in which unpredictable completions were presented in three-sentence passages that varied in whether the source of constraint violation originated from the global or local context. Luke and Christianson (2016) also found no evidence of prediction error costs when readers were presented with a corpus of naturalistic text passages for which cloze probability values had been calculated for each word.…”

“…On the basis of previous eye-movement studies using similar controlled experimental designs (Frisson et al, 2017;Wong et al, 2022), predictable words in strongly constraining contexts were expected to elicit the largest processing benefits relative to the same words in weakly constraining contexts. However, no immediate consequences of prediction failure were expected when plausible unpredictable words were presented in strongly compared to weakly constraining contexts, regardless of their semantic relatedness to the best completion.…”

“…As such, the fact that previously predictable words were facilitated downstream suggests that the effects of misprediction were only temporary because these expectations subsequently lingered to impact processing. The idea that the processing consequences of misprediction are short-lived could account for why evidence of prediction error costs across previous eyemovement studies has been inconsistent, and generally elusive, during the immediate processing of unexpected input (Andrews et al, 2022;Cevoli et al, 2022;Frisson et al, 2017;Luke & Christianson, 2016;Wong et al, 2022). Accordingly, it raises the possibility that readers' expectations could also remain active downstream in the eye-movement record even if they are disconfirmed by unexpected input.…”

“…The experiments presented short, connected sentence pairs in which the first sentence contained a target word that either confirmed or disconfirmed readers' expectations, while the second sentence presented previously predictable words close to their initial point of activation. This design allowed us to assess prediction error costs at target words that have not been observed in previous controlled eye-movement studies (e.g., Frisson et al, 2017;Wong et al, 2022), and, more specifically, to determine whether readers' predictions have downstream consequences even if they do not eventuate and are replaced by less expected input.…”

Early parafoveal semantic integration in natural reading

“…Firstly, the construction of pretarget– target word pairs consistently follows an adjective–noun phrase structure, potentially leading to semantic violations arising from immediate local incongruence rather than a broader incongruence derived from the entire sentential context. While the context preceding target words was deliberately minimized to ensure a pure effect of bottom-up parafoveal processing rather than the confounding impact of top-down prediction, it is essential to recognize that information from both local and global contexts can exert distinct effects on word processing during natural reading (Wong et al, 2022). Future investigations should incorporate more information-rich contexts to explore the extent to which the parafoveal semantic integration effect observed in this study can be generalized.…”