Listener characteristics modulate the semantic processing of native vs. foreign-accented speech

Abstract

Foreign accents have been shown to have considerable impact on how language is processed [1]. However, the impact of a foreign accent on semantic processing is not well understood. Conflicting results have been reported by previous event-related potential (ERP) studies investigating the impact of foreign-accentedness on the N400 effect elicited by semantic violations. Furthermore, these studies have only examined a subset of the four characteristics of the N400 (i.e. onset latency, latency, amplitude, and scalp distribution), and have been conducted in linguistic environments where foreign-accented speech is relatively uncommon. The current study therefore compared the N400 effect elicited by semantic violations in native Australian English vs. Mandarin-accented English, in a context where foreign-accented speech is common. Factors which may be responsible for individual variability in N400 amplitude were also investigated. The results showed no differences between the N400s elicited by native and foreign-accented speech in any of the four aforementioned characteristics. However, the analysis of individual variability revealed an effect of familiarity with foreign-accented speech on the amplitude of N400 effects for semantic violations. An effect of working memory capacity on N400 amplitude was also found. These findings highlight the relevance of the ambient linguistic environment for studies of speech processing, and demonstrate the interacting influences of both speaker- and listener-related factors on semantic processing.

Replicability of neural responses to speech accent is driven by study design and analytical parameters

“…If it is the case that out-group speech produces a late N400 or some other semantically modulated difference in signal around this time frame, the fact that some studies overlap with it and others do not could lead to significantly different results among studies, even with similar data sets. For example, it is plausible that a critical component of the response to congruity manipulations in out-group speech could fall between 500 and 600 ms, as consistent with the findings of Holt et al 13 . If this is the case, it could potentially explain the conflict between the observation by Grey and van Hell 12 of no differential response to congruence and incongruence in the N400 window in out-group speech, and the observation by Romero-Rivas et al 10 of both a differential response in out-group speech and a greater N400 response to incongruent sentences in out-group speech than in-group speech, as the latter study included this 500–600 ms range and the former did not.…”

“…Spurred by the increasing societal relevance of this topic, several recent studies have attempted to identify differences and similarities that may exist in neural processing of speech in an accent that either matches or does not match the accent of the listener 8 – 13 , referred to here respectively as in-group and out-group accents. As a note before proceeding, studies in this literature often draw a distinction between so-called “native-accented” and “foreign-accented” speech.…”

“…In Holt et al 13 , who performed a cluster-based analysis instead of using a priori analysis parameters, clusters reflecting a significant negativity for incongruent as compared to congruent sentences were identified between 200 and 900 ms for in-group speech but between 560 and 660 ms in out-group speech. While care must be taken in interpreting cluster-based analyses as they do not establish the temporospatial boundaries of the detected effect 17 , this observation is consistent with a late negativity to semantic incongruence in out-group speech.…”

“…Grey and van Hell 12 ascribe their results’ divergence from previous studies to differences in the participant populations, as their study focused on monolinguals with limited foreign accent experience, while Hanulíková et al 9 and Romero-Rivas et al 10 employed bilinguals listening in their native language. Holt et al 13 suggest that discrepancies in outcomes among studies including their own could result from differences in participant population, which in their study was drawn from a cosmopolitan city with presumably high exposure to different accents, and from other studies’ use of time ranges and regions that were either pre-determined or selected from visual inspection of results. However, they do not subject their data to any of the analyses used by previous studies, so it is unclear whether they would have gotten the same results had they done so.…”

“…(Parameters used in each study, including time ranges and region definitions, can be found in Table 1 ). Meanwhile, the cluster-based analysis used by Holt et al 13 identifies that differences in signal evoked by semantically incongruent as compared to semantically congruent sentences appear reliably in out-group speech only between 560 and 660 ms. If it is the case that out-group speech produces a late N400 or some other semantically modulated difference in signal around this time frame, the fact that some studies overlap with it and others do not could lead to significantly different results among studies, even with similar data sets.…”