Association Between Structural and Connectivity Characteristics in the Language Networks and the Statistical Learning Networks and Children’s Language Abilities

[KatherineTrice]

Research question(s)

Do structural characteristics (cortical thickness, grey-matter volume) of the language network predict concurrent language skills?
Do structural characteristics (cortical thickness, grey-matter volume) of the language network predict later language skills?
Does the connectivity of the language network predict concurrent language skills?
Does the connectivity of the language network predict later language skills?
Do structural characteristics (cortical thickness, grey-matter volume) of the statistical learning networks predict concurrent language skills?
Do structural characteristics (cortical thickness, grey-matter volume) of the statistical learning networks predict later language skills?
Does the connectivity of the statistical learning networks predict concurrent language skills?
Does the connectivity of the statistical learning networks predict later language skills?

Description

The overall goal of this project is to test whether the structural and connectivity characteristics in the language networks and the statistical learning networks are associated with children’s language abilities. We would most likely use the language networks defined by Fedorenko et al (2010) and the statistical learning networks defined in my lab's research as ROIs and extract the cortical thickness, grey-matter volume, and the connectivity between these ROIs as neural features to predict concurrent language skills (measured by NIH toolbox picture vocabulary and oral reading recognition) as well as the two-year follow-up scores on these two tests. This project could also easily be paired down to a smaller scale, and simply focus on one aspect of the question, if anyone who wishes to work with me decides that it is too broad for the desired time frame.

Suggested keywords/tags

Structural MRI, Functional MRI, Connectivity, Cortical Thickness, Grey-Matter Volume, Language Network, Statistical Learning, Language Skills

[tsalo]

Would you mind adding a bit of information about the planned analyses, necessary skills, and skills you could use help with in this project? It should make it easier for prospective collaborators to determine how and if they can contribute. Specifically, our new template has sections named "Tools and algorithms to be used", "Skills we could use help with", and "Link to analysis plan" (if you have been working on a more detailed plan than would fit in this issue).

[angielaird]

Hi @KatherineTrice, as you're thinking about your analysis plan for Project Week, perhaps consider joining forces with @mstnva (Issue #23 Cerebellar-frontal connections in light of language task performance) or @marissa-marko (Issue #13 Shared and distinct neural correlates of reading and attention) to see if there are common analytic goals that can be merged.

[KatherineTrice]

Revised Project week proposal and plan!

Research question(s)

Do structural characteristics (cortical thickness, grey-matter volume) of the statistical learning ROIs predict concurrent language skills?
Do structural characteristics (cortical thickness, grey-matter volume) of the statistical learning ROIs predict longitudinal language skills?
Does the connectivity of of the statistical learning ROIs predict concurrent language skills?
Does the connectivity of the statistical learning ROIs predict longitudinal language skills? (My favorite thus far)
What is the structural overlap of the statistical learning ROIs and the language network at baseline?
Are there longitudinal changes to the structural features of these overlapping regions?
What is the connectivity between the statistical learning ROIs and the language network at baseline?
Are there longitudinal changes to the connectivity of the statistical learning ROIs and the language network?

Description

Statistical learning is the robust human ability to implicitly learn and adapt to regularities from input. In our lab, we have found very strong relationships between statistical learning and reading skills and statistical learning and language skills in adults and children. But there is very little we know about how the neurological measures of statistical learning are related to language skills and language growth.

I hope to use language assessments from the ABCD dataset and the ROIs relevant to statistical learning defined in our lab based on a group level analysis of 22 adult learners who have shown robust learning in these tasks to validate the importance of this network in both concurrent language skills and longitudinal growth along with literacy skills.
A secondary goal, based on time, is to explore the overlap and connectivity between the language network in the brain based on Fedorenko et al (2010) and the statistical learning ROIs.

As this topic is rather broad, I’m open to team input on what aspect of the analysis we focus on. My preference is the association between the connectivity of the statistical learning ROIs and longitudinal language skills, but all these questions interest me.

Suggested keywords/tags

Structural MRI, Functional MRI, Connectivity, Cortical Thickness, Grey-Matter Volume, Language Network, Statistical Learning, Language Skills

Tools and algorithms to be used

Assessments (both baseline and 2-year):
-Picture vocabulary task (NIH toolbox)
-Oral reading recognition task (NIH toolbox)

Connectivity analysis:
-Resting state connectivity analysis tool (CONN?)*

Structural analysis:
-Freesurfer

Algorithms:
TBD

*I’m also open to people’s recommendation on resting state connectivity analysis, as CONN matlab based, which I have less experience in, and isn’t always as successful when analyzing high performance clusters.

Skills we could use help with

-I’m looking for partners that know more about different algorithms to help determine the best fit
-MRI processing experience is a plus! Especially with connectivity analyses.

[mxd2019]

Hi; @KatherineTriceI would like to join you in this work. I have freesurfer segmentation experience and tractography reconstruction skill using DSI studio, and trackvis. Let me know what brain areas you would like to understand the connectivity between....I have a good background in neuroanatomy as well.

[KatherineTrice]

That is fantastic, thank you so, so much for your interest! I'm getting the parcels that we can use for the SL portion from my lab mate at the moment, but while that is going on, I'll open a slack channel for us and we can decide 1) which questions(s) we are most interested in asking and 2) what our analysis pipeline should be!

[KatherineTrice]

Since I don't have your slack username, I'll set the channel to public for now; just post in it and, once you join, I'll make it private.

[KatherineTrice]

If you have trouble joining for some reason, just message me directly on Slack - my username is Katherine Trice - and I'll add you!