iiinova — where your digital self keeps evolving

Agata Wolna @awolna.bsky.social · 23/02/2026

Our results point to highly adaptive nature of the control mechanisms in bilinguals, which flexibly respond to immediate contextual demands rather than only being determined by language proficiency or other fixed traits of bilingual speakers. (7/7)

000

Agata Wolna @awolna.bsky.social · 23/02/2026

Importantly, these contextual dominance shifts persisted after LST rather than snapping back immediately (with smaller dominance reversal following the L2- than L1-dominant context) (6/n)

100

Agata Wolna @awolna.bsky.social · 23/02/2026

Notably, the reversed dominance effect increased over time and was driven primarily by an increase in inhibition to the language more present in the context (in-depth analyses of the temporal dynamic in Supplemental analyses) (5/n)

100

Agata Wolna @awolna.bsky.social · 23/02/2026

Language context heavily affected performance in LST, reversing the typical switch cost asymmetry from L1 in the L1-context to L2 in the L2-context, and decreasing the reversed language dominance in the L2-dominant context (4/n)

110

Agata Wolna @awolna.bsky.social · 23/02/2026

We asked 88 Polish-English unbalanced bilinguals to complete LST in L1- and L2-dominant contexts, created by manipulating the frequency of naming pictures in the L1 vs. L2 before and during LST. (3/n)

100

Agata Wolna @awolna.bsky.social · 23/02/2026

Both are usually attributed to an imbalance in the accessibility of L1 and L2, driven primarily by stable traits of bilingual language experience, like proficiency. Here, we asked if immediate changes in the language context can modulate this balance and affect language control engagement. (2/n)

100

Agata Wolna @awolna.bsky.social · 23/02/2026

The language switching task (LST) is often used to measure two signatures of language control: 1️⃣ asymmetric switch costs (switching into L1 is costlier than into L2), and 2️⃣ reversed language dominance (L1 becomes slower than L2). (1/n)

100

Agata Wolna @awolna.bsky.social · 23/02/2026

000

Agata Wolna @awolna.bsky.social · 23/02/2026

100

Agata Wolna @awolna.bsky.social · 23/02/2026

We asked 88 Polish-English unbalanced bilinguals to complete LST in L1- and L2-dominant contexts, created by manipulating the frequency of the L1 vs. L2 used before and during LST. (3/n)

100

Agata Wolna @awolna.bsky.social · 23/02/2026

100

Agata Wolna @awolna.bsky.social · 23/02/2026

100

Reposted by Agata Wolna

Cory Shain @coryshain.bsky.social · 31/03/2025

New brain/language study w/ @evfedorenko.bsky.social! We applied task-agnostic individualized functional connectomics (iFC) to the entire history of fMRI scanning in the Fedorenko lab, parcellating nearly 1200 brains into networks based on activity fluctuations alone. doi.org/10.1101/2025... . 🧵

doi.org

A language network in the individualized functional connectomes of over 1,000 human brains doing arbitrary tasks

A century and a half of neuroscience has yielded many divergent theories of the neurobiology of language. Two factors that likely contribute to this situation include (a) conceptual disagreement…

14213

Agata Wolna @awolna.bsky.social · 03/04/2025

Lesson 3️⃣: Non-selective areas may also be important to understand language in the🧠 Some regions respond to both amodal language, as well as to non-linguistic tasks. These mixed functional profiles may reflect a role in integrating information across cortical networks. (14/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

Lesson 2️⃣: The components of spatially distributed systems are still highly focal. At the individual level, the full set of 22 language-selective areas, although distributed across many brain regions, takes up <1.5% of the brain volume—about the size of a large🍓. (13/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

Lesson 1️⃣: Not all “language tasks” are created equal Some brain areas where responses to “language tasks” have been reported don’t survive our criteria for what counts a high-level, amodal and selective language region These include occipital areas and the basal ganglia. (12/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

📌To summarize: these are the three main lessons coming from our work: (11/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

We also looked at the Harvard-Oxford Subcortical parcellation where we found selective responses to language in bilateral amygdalae and hippocampi, non-selective but robust responses to both language tasks left thalamus shows, but no proof for neither in the basal ganglia: (10/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

To make sure we did not miss anything, we also systematically searched for language-selective responses in 3 standard atalses: DKT, Harvard-Oxford Cortical, and Glasser parcellation. These results converge with our data-driven functional parcellation. (9/n)

110

Agata Wolna @awolna.bsky.social · 03/04/2025

These 12 new regions constitute new targets for systematic evaluation in future work, to understand how they contribute to language processing, and whether their role is distinct from that of the core language areas. (8/n)

110

Agata Wolna @awolna.bsky.social · 03/04/2025

… two regions in the right and one in the left cerebellum (for in-depth characterization of cerebellar language regions look out for an imminent preprint from @coltoncasto.bsky.social!): (7/n)

130

Agata Wolna @awolna.bsky.social · 03/04/2025

… bilateral areas in the ventral temporal poles and the left BTLA (see also Jin Li's work: tinyurl.com/2t98jkks): (6/n)

110

Agata Wolna @awolna.bsky.social · 03/04/2025

…which leaves 12 newly established language-selective regions! These include three medial frontal regions and a region in the left precuneus: (5/n)

110

Agata Wolna @awolna.bsky.social · 03/04/2025

Using group-constrained subject-specific parcellation, we identified 27 candidate language regions in the🧠. 22 of the 27 regions show robust, amodal response to language, and 20 are selective for language. These include the core language regions and their right homotopes. (4/n)

110

Agata Wolna @awolna.bsky.social · 03/04/2025

We then tested whether these areas—defined within individuals by the reading-based localizer (A), generalize to auditory language comprehension (B), and whether they are selective relative to a non-linguistic difficult task (C). (3/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

Here using a dataset of 772 participants who completed a validated language localizer task (reading sentences vs. nonwords), we identified a set of 27 candidate language-responsive regions in the brain. (2/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

Language processing recruits a left fronto-temporal🧠network. However, language tasks often engage regions beyond this core system. As many paradigms don’t isolate language from speech or task demands, it’s hard to know if these non-core areas are real "language" regions. (1/n)

110

Agata Wolna @awolna.bsky.social · 03/04/2025

000

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

📌To summarize: these are the three main lessons coming from our work: (11/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

… two regions in the right and one in the left cerebellum (for in-depth characterization of cerebellar language regions look out for an imminent preprint from @coltoncasto.bsky.social !): (7/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

… bilateral areas in the ventral temporal poles and the left BTLA (see also Jin Li’s work: tinyurl.com/2t98jkks): (6/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

…which leaves 12 newly established language-selective regions! These include three medial frontal regions and a region in the left precuneus: (5/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

Language processing recruits a left fronto-temporal 🧠network. However, language tasks often engage regions beyond this core system. As many paradigms don’t isolate language from speech or task demands, it’s hard to know if these non-core areas are real "language" regions. (1/n)

100

Agata Wolna @awolna.bsky.social · 03/04/2025

Thank you!! We are really excited about this work and what comes after - a systematic, in-depth exploration of these newly identified regions!

010

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

100

Agata Wolna @awolna.bsky.social · 03/04/2025

📌To summarize: these are the three main lessons coming from our work: (11/n)

100