Poster D60, Friday, August 17, 4:45 – 6:30 pm, Room 2000AB
Neural correlates of between-language competition in foreign language attrition
Anne Mickan1,2, James M. McQueen2, Vitória Piai2,3, Kristin Lemhöfer2;1Max-Planck Institute for Psycholinguistics, 2Donders Centre for Cognition, Radboud University Nijmegen, 3Department of Medical Psychology, Radboudumc
Recent psycholinguistic research suggests that language attrition (i.e. forgetting) is driven by interference and competition from other, more recently used languages (Levy et al, 2004; Mickan et al, 2018). Mickan et al (2018), for example, have shown that retrieval failure in a foreign language can be caused by the recent, repeated practice of translation equivalents in another language, especially another foreign one. Here we aim to track these competition dynamics at the neural level, in an effort to further our understanding of the precise processes underlying foreign language attrition. Twenty-seven Dutch native speakers with knowledge of English (B2-C1) and no prior knowledge of Italian first learned 70 new Italian words. The learning session was spread out over two consecutive days and consisted of a mix of comprehension (multiple choice) and production (picture naming) tasks. One day after the second learning session, they performed naming tasks on half of these words in English. Finally, memory for all initially learned Italian items was tested again, in a delayed picture-naming task. Recall accuracy and (delayed) naming latency, as well as EEG measurements were taken. The latter were recorded during the 2s delay period after picture presentation onset and before speech onset, which is when retrieval and competition-related dynamics should be most evident. Following Mickan et al (2018) we expected more mistakes and slower responses on Italian words that had just been interfered with (i.e. retrieved in English) compared to items in the no-interference condition. On the neural level, we expected two types of modulations: an increase in the frontal N2 amplitude for interfered compared to not interfered items, and an increase in theta power for the interference condition. Both electrophysiological measures have been linked to competition: the N2 is usually interpreted as an indicator for increased inhibition demands, during for example language switching tasks (Kroll et al, 2008). Theta power has been linked to competition increases in studies on retrieval-induced forgetting (Ferreira et al, 2014; Staudigl et al, 2010). Results showed that recall was indeed slower and tended to be less complete for Italian words that received interference than for Italian words that did not. In the EEG, in line with expectations, we found an increased frontal N2 between 200-300ms post picture onset for the interfered items compared to the not interfered items. Unexpectedly, we also found an increased late positivity (400-700ms, central) for the not interfered compared to the interfered items. Finally, again in line with the literature, we observed an increase in theta power for the interfered items from 500ms onwards (central). The fact that the behavioral forgetting effects were accompanied by an increased N2 supports the idea that inhibition is at play in foreign language attrition, most likely as a mechanism to resolve the increased competition from the recent practice of English words. Theta power changes, we propose, index this increase in competition. Results will be further discussed in light of previous research and theories on language competition in multilingual language production.
Topic Area: Multilingualism