Crosslinguistic evidence against interference from extra-sentential distractors.

Abstract

Cue-based parsing theories posit that dependency resolution during real-time sentence comprehension relies on cue-based retrieval of linguistic items encoded in memory. This retrieval mechanism is prone to similarity-based interference, which can occur when there are items in memory that are similar to the retrieval target. Interference during sentence comprehension seems to be well-established across numerous syntactic dependencies; however, the evidence for interference on within-sentence dependencies from sentence-external lexical items (encoded in memory prior to reading a target dependency) is inconclusive (Van Dyke &McElree, 2006; Van Dyke et al., 2014). The goal of the present study is to provide a large-scale cross-linguistic investigation of retrieval interference from sentence-external distractors under varying processing demands. Three larger-sample eye-tracking studies in English (N=66),German (N=122) and Russian (N=109) show no support for similarity-based interference from sentence-external material during sentence comprehension. We discuss the implications of our findings for cue-based parsing theories.

Agreement attraction in grammatical sentences and the role of the task

“…Generally, the present results are in line with several studies reporting interference effects in grammatical sentences only in question responses, but not in reading times (Jäger et al, 2015;Laurinavichyute et al, 2017;Mertzen et al, 2020).…”

Dynamical Cognitive Modeling of Syntactic Processing and Eye Movement Control in Reading

“…Therefore, rather than assuming that the parser has direct control of the eye-movement targeting system, we propose an indirect, stochastic influence on saccade goals via memory activations. This is in line with our results on rather moderate effects of syntactic processing on eye guidance (Mertzen et al, 2020), and contrasts with earlier claims of a more direct relationship (Frazier & Rayner, 1982;Just & Carpenter, 1980).…”

“…To test the integrated model, its predictions, and explanatory power with regard to psycholinguistic data, we use empirical data from an experimental study of proactive memory interference during sentence processing. In a large-scale attempt to replicate earlier findings by Van Dyke and McElree (2006), Mertzen et al (2020) displayed sentences to subjects in four different experimental conditions in a crossed memory load (2) × feature match (2) design. Sentences were presented either with memory load (2a,b) or without memory load (2c,d).…”

“…As a proof of concept that the integrated SWIFT-LV05 is generally capable of generating the predicted pattern of effects, data were simulated for the sentences used in Mertzen et al (2020). Summary results are illustrated in Figure 3, which plots first-pass reading times (FPRT) for words across four different experimental conditions.…”

Share the code, not just the data: A case study of the reproducibility of articles published in the Journal of Memory and Language under the open data policy

“…A. Klein et al, 2014;R. A. Klein et al, 2018;Mertzen et al, 2020;Nieuwland et al, 2018;Stack et al, 2018;Vasishth et al, 2018). Such failures to replicate raise important questions about the extent of the non-replicability problem in psycholinguistics and related areas.…”