A replication of morphological flanker effects in French

ABSTRACT

Grainger et al. examined flanker effects in an experiment where target words could be morphologically complex (e.g. farmer), have a pseudo-morphological structure (e.g. corner), or have no morphological structure (e.g. cashew). Target words were flanked on both sides by the left-aligned embedded word in the related flanker condition (e.g. farm farmer farm; corn corner corn; cash cashew cash). Effects of related flankers were compared with matched unrelated flankers (e.g. book farmer book). Significantly greater effects of flanker relatedness were found with morphologically related flankers compared with both pseudo-morphological flankers and non-morphological embedded word flankers. However, more recent work in Spanish has failed to replicate this specific pattern. We therefore sought to replicate our findings in French, while increasing experimental power and modifying the nonword targets. The results of the present study successfully replicate the findings of Grainger et al. [(2021). Morphological processing in the flankers task. Language, Cognition and Neuroscience, 36(3), 288–295. https://doi.org/10.1080/23273798.2020.1810292].

KEYWORDS:

• Reading
• morphology
• flankers task
• lexical decision
• multi-word processing

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 However, it is possible that a lack of statistical power as well as uncontrolled differences in the strength of semantic relatedness (Lazaro et al., 2022) might be factors driving the observed absence of a difference between the morphologically (and semantically) related and pseudo-morphological conditions in these studies. We further note that at least one study (Feldman et al., 2009) has found a significant advantage for truly morphological primes compared with pseudo-morphological primes.

2 It should be noted that Lazaro et al. appealed to the length of their stimuli as the reason for the absence of morphological flanker effects in their Experiment 1, but the lengths were comparable to those in the Grainger et al. study.