A dataset to explore the reasons behind language learning abilities
The WP Aptitudes, a project within the NCCR Evolving Language, has just finished collecting their dataset. The team has a lot of projects in mind for the data set – including monolingual and multilingual participants, some of which with dyslexia – starting with a big exploratory analysis using the behavioral and brain data collected.
By Aurora Petroz
In the MRI room with the Aptitudes WP. © Golestani Lab on Twitter.
On the 12rd of June, the Golestani Lab scanned their last participant for the NCCR Evolving Language Aptitudes work package dataset. Participating in the data collection were PhD candidate Irene Balboni and post-doc Alessandra Rampini, supervised by the two P.I. Raphaele Berthele and Narly Golestani.
Irene Balboni works on individual differences in language learning and is particularly interested in dyslexia. Alessandra Rampini studies how we learn a language (our own or others), what is common to all of them and how it interacts with other skills such as music.
The construction of the dataset began almost one year ago, and the data collection took place in several stages. “We had 4 sessions with each of our participants: the first one was a questionnaire session, the second and third one were behavior sessions and the last one was an MRI session at the Campus Biotech”, Irene reports.
But what is the data going to be used for?
Irene and Alessandra have a lot of projects for it! “We are interested in understanding what lies behind language aptitude at the brain and behavioral level, what makes us different in our language learning abilities and what differentiate people gifted for language learning from some others that might be less successful at it, be it predisposition or experience ”. The difference could lie in training or in predisposition, so they collected a wide range of measures, including language-specific measures and domain-specific measures like musicality. “The idea was to gather data that could make a difference in language learning ability. From there, we’ll be able to see if there is a link between them or not” says Irene. The data were collected on a vast population including monolinguals and multilinguals among which some were also dyslexic readers. “Our dataset contains a wide and valuable range of languages,” Alessandra adds.
And what are the next steps?
“Now that we have all the data, we need to curate them and put them in a form shareable with the scientific community.” They plan to first do a big exploratory analysis with the behavioral data, as well as some important brain data that were collected. For Irene, this analysis can be very interesting to know the possible reasons for language learning differences.