Constraints on parallel activation in bilingual spoken language processing: Examining proficiency and lexical status using eye-tracking

Abstract

During spoken word-recognition, bilinguals have been shown to access their two languages simultaneously. The present study examined effects of language proficiency and lexical status on parallel language activation. Language proficiency was manipulated by testing German-native and English-native bilingual speakers of German and English. Lexical status was manipulated by presenting target words that either overlapped in form across translation equivalents (cognate words) or did not overlap in form across translation equivalents (English-specific words). Participants identified targets (such as hen) from picture-displays that also included similar-sounding German competitor words (such as Hemd, “shirt”). Eye-movements to German competitors were used to index co-activation of German. Results showed that both bilingual groups co-activated German while processing cognate targets; however, only German-native bilinguals co-activated German while processing English-specific targets. These findings indicate that high language proficiency and cognate status boost parallel language activation in bilinguals.

Acknowledgements

This work was supported by grants NICHD 1R03HD046952-01A1 and NSF BCS-0418495 to the second author, and by a Northwestern University Graduate Research Grant to the first author.

Thanks go to Margarita Kaushanskaya, James Booth, Cynthia Thompson, and Judith Kroll for insightful comments and discussion, and to Valerie Burt, Gayatri Menon, Naveen Malik, Vridhi Chhabria, Nicole Kaligeropolous, Nadia Cone, Olga Boukrina, and Avital Rabin for their assistance at various stages of the project. We are grateful to Manuel Carreiras and to two anonymous reviewers for helpful comments and feedback on a previous version of this paper.

Parts of this work were presented at the Fifth International Symposium on Bilingualism, and at the 27th Annual Meeting of the Cognitive Science Society.

Notes

¹For a linking hypothesis between linguistic processing and eye-movements, see Tanenhaus, Magnuson, Dahan, and Chambers (2000).

²This translation was made by a fluent German–English bilingual, and back-translated to English by two other fluent German–English bilinguals for reliability. The German and English versions were then balanced by-item on word frequency [CELEX lexical database, Baayen, Piepenbrock, & Van Rijn, 1995, t (203) = 0.6, p>.5].

³For purposes of subsequent analyses, English recordings of target words were compared to German recordings of competitor words (German recordings were made by native speakers of German). Onset similarity between English target-words and German competitor-words was assessed at the acoustic level (Ju & Luce, 2004) using Sound Studio software. The duration of acoustic overlap between target and competitor onsets was measured, and is shown in Table 2. A one-way ANOVA yielded significant differences between durations of target-competitor overlap in the low-, medium- and high-overlap conditions (cognate target condition, F(2, 30) = 45.5, p<.001; English target condition, F(2, 28) = 45.2, p<.001). Planned post-hoc comparisons yielded differences between low- and medium-overlap conditions, as well as between medium- and high-overlap conditions and between low- and high-overlap conditions (LSD post-hocs: p<.01). Significant differences between the three conditions were also found in terms of phonemes (cognate target condition, F (2, 30) = 18.6, p<.001; English target condition, F (2, 28) = 4.8, p<.05, and in terms of phonetic features: cognate target condition, F (2, 30) = 10.2, p<.001; English target condition, F (1, 28) = 6.1, p<.01.

⁴Of these, the target-competitor pairs cylinder / Zylinder (top-hat) and lock / Lockenwickler (curlers) were excluded due to within-language competition. The picture of curlers contained locks (of hair) and the top hat was cylinder-shaped. For evidence on eye-movements due to shape-similarities, see Dahan & Tanenhaus, 2005. The target-competitor pairs bear / Band (ribbon), file / Pfeil (arrow), turtle / Tuer (door) and tooth / Tuch (cloth) were excluded because these competitor pictures were fixated more than control items in the monolingual control group.

⁵By-item analyses were significant or marginally significant in all between-group comparisons, but were not significant in within-group comparisons. This is likely due to the fact that duration of overlap (in milliseconds) was a continuous variable. Variations on the continuum within each category were likely to elicit different responses, resulting in large standard errors. This variability among items contradicted the inherent assumption in by-item analyses that all items be similar, prompting analyses of variance by subject only. To verify effects of phonological overlap across items, regression analyses were performed (where degree of phonological overlap was regressed on likelihood of parallel activation).

⁶Follow-up t-tests were only conducted to compare percentages of looks to competitor vs. control items, with competitor-control item differences reflecting parallel language activation. Looks to competitors in one condition vs. another (or to control items in one condition vs. another) were not compared statistically, because they do not reflect parallel language activation. Any differences between looks to competitors in different conditions (or looks to control items in different conditions) may be due to the presence of different stimuli across these conditions. Further, differences between looks to competitors in different groups (or looks to control items in different groups) may be due to general differences in response latencies across groups. Follow-up t-tests were 1-tailed.

⁷Alternatively, it may be the case that the German L1 cohort was highly co-activated, even with target-competitor low phonological overlap. The increased duration of competition may have had no additional effect on competitor activation, and no sensitivity to phonological overlap would have been observed. Such an account has to be questioned for two reasons. First, absence of phonological overlap effects should also be apparent in monolingual auditory word recognition; however, this is not the case. Monolingual priming studies suggest that degree of phonological overlap influences co-activation of two words (Slowiaczek & Hamburger, 1992). Further, Weber and Cutler (2004) found a target-competitor overlap effect in Dutch-native bilinguals who processed words with confusable vowels in their second language, English. Therefore, since phonological overlap effects have been found within native and non-native languages, the absence of phonological overlap sensitivity, due to overall high L1 co-activation, is not a plausible account.