Periodic Reporting for period 1 - BilMemBrain (When language meets memory: the role of language exposure in semantic- episodic memory interaction in bilinguals)
Reporting period: 2015-10-01 to 2017-09-30
The main aim of this project was to study the influence of linguistic pattern in the information access and retrieval. Linguistic pattern was defined by language age of acquisition, proficiency and language use and exposure in everyday life.
More specifically, the aims of this project were 1) to understand the consequences of being Monolingual versus Bilingual in true and false memories; and 2) as experience and environmental input change the brain, this projects studied if being bilingual modulates brain areas engaged in remembering correctly and falsely.
In order to study this, we used Deese-Roediger-McDermott paradigm (DRM) because it is sensitive to linguistic manipulations on memory. In this paradigm, after studying several word lists (e.g. bed, rest, tired, dream, wake,…) converging on a semantic theme captured in a word that is never presented during study (e.g. sleep), participants perform an old/new recognition test that includes studied words (e.g. rest; target), semantically related non-studied words (e.g. sleep; critical lure), and other new non-semantically associated words (e.g. flower; unrelated lure). Participants typically exhibit a high rate of false memories to Critical Lures (CL), which is thought to reflect semantic relational processing during encoding. DRM lists could vary in their tendency to elicit false memories based on their associative strength. In this project the associative strength was manipulated using high and low associative strength lists.
We conducted a series of behavioral experiments using DRM paradigm: comparing Spanish and Basque languages in Bilingual group; comparing Monolingual and Bilingual groups in the same language, Spanish; and comparing Monolingual group in Spanish with Bilingual group in Basque.
And we conducted an fMRI experiment as well, comparing Monolinguals and Bilinguals in Spanish. In all the experiments associative strength was manipulated. And language proficiency, age of acquisition, and language use/exposure were controlled.
The results from this project show that being Bilingual decreases false memory rate only when intrusions are semantically associated with the studied words. Importantly, higher false memories for Monolinguals than Bilinguals were observed across high and low associative strength conditions and language comparisons. Overall, the use of a second language in our everyday life modulates the access and retrieval of words.
fMRI results revealed that left Inferior Frontal Gyrus (IFG) was significantly more activated in Bilinguals than Monolinguals for both, true and false memories. And this area was more activated for low than high associative strength conditions. These results are consistent with previous evidence showing that the IFG is engaged in semantic tasks requiring active strategies and effortful processing. And this suggests a less automatized or more effortful semantic access in Bilinguals than in Monolinguals.
Results: Bilinguals exhibited a higher rate of CL false alarms than Unrelated Lures (UL) false alarms showing that their memory errors were strongly associated with relational semantic processing. Another main results was that Bilinguals showed more CLs false alarms in L1 than in L2.
Conclusions: It is not clear which is the main factor that explains the decrease in false memories in L2. This result could be modulated by language use (higher/lower) and by language (Spanish/Basque). It is possible that the DRM lists in each Language are not completely equivalent. In order to study this we carried out the Study 2.
Dissemination: 2016 Annual Meeting of the Cognitive Neuroscience Society.
Study 2a: DRM behavioral study comparing Monolingual and Bilingual groups in Spanish (L1) (93 participants: 47 Bilinguals and 46 Monolinguals)
Results: Results showed that Monolinguals exhibited a higher percentage of CLs false alarms than Bilinguals but similar Hits and UL false alarms.
Conclusions: Considering that Monolinguals showed a stronger use of L1 than Bilinguals, these results probably reflect the effect of language use in semantic relational processing leading to a stronger DRM false-memory effect in the Monolingual group than in the Bilingual group.
Study 2b: DRM behavioral study comparing Monolinguals in Spanish (L1) and Bilinguals in Basque (L2) (93 participants)
Results: Monolinguals exhibited a stronger DRM false-memory effect in their L1 than Bilinguals in their L2 but there were no statistically significant differences between groups for Hits or UL false alarms.
Conclusions: Monolinguals, through an extensive use of Spanish, showed primed access to sematic network and due to this advantage they exhibited a higher number of CL false alarms than the Bilingual group.
Dissemination of Study 2a and 2b: International Conference on Memory. 2016.
Dissemination of Study 1 and 2: Marín-García, E., & Paz-Alonso P.M. Is semantic processing modulated by bilingualism? A comparative study between Spanish-Basque early and proficient bilinguals with Spanish monolinguals. In press.
Study 3: DRM behavioral study comparing encoding instructions (standard-implicit versus explicit relational processing) and Monolingual and Bilingual groups in Spanish (L1). (32 participants: 16 Bilinguals and 16 Monolinguals)
Results: Only Bilinguals exhibited more CL false memories for lists studied under implicit versus explicit encoding instructions.
Conclusions: Further research is needed to understand if this result is related to the small size of the groups.
Dissemination: Psychonomics Society 57th Annual Meeting. 2016.
Study 4: DRM fMRI study comparing Monolingual and Bilingual groups in Spanish (L1)
Results and conclusions: Monolinguals and Bilinguals engage similar brain areas during the DRM retrieval task. Left Triangularis area is more activated for Bilinguals, which probably is reflecting a more effortful and recollection-to-reject operations than Monolinguals that may be using more automatic processing.
Dissemination: The International Conference for Cognitive Neuroscience. 2017.