03/30/2023: Bratislav Misic

[GabeNicholson]

Comment below with a well-developed question or comment about the reading for this week's workshop.

If you would really like to ask your question in person, please place two exclamation points before your question to signal that you really want to ask it.

Please post your question by Tuesday of the coming week, at 11:59 PM. We will also ask you all to upvote questions that you think were particularly good. There may be prizes for top question-askers.

[sdbaier]

Dear Professor Misic, thank you for sharing your research with us. Great to have some Canadian representation at our workshop! A few questions:

Functional Connectivity emerges from higher-order interactions, i.e., structure is linked to function via complex associations, but their correlation is imperfect at best. Estimating FC across time seems logical, but how can these associations be estimated across individuals (presumably) within a single time slice? Is the underlying assumption that neural activity in different individuals corresponds to different “stages”, analogous to estimation across time?

The paper presupposes structural connections, and indicates that functional connections emerge from higher-order interactions. How can we be sure about the directionality of this link? This is the sociologist in me, so this might not apply to neuroscience, but could structure equally emerge from function?

[yuzhouw313]

Hello, Professor Misic, Thank you for sharing with us your research on the relationship between brain's structural and functional properties at the macro-scale level. I look forward hearing more of your research on Thursday!

I am not familiar with the field of neuroscience or neuro-imaging techniques, but I found your discussion on the relationship and non-one-to-one mapping between the structural and functional connections of the brain interesting and relatively intuitive for a layperson like me. However, I am particularly curious about the disruption or change on this relation, especially its consequences. Specifically, when you stated that "disruption of the protein’s structure results in loss of function," what function are you referring to and can you elaborate on the impact of brain's structure disruption on its function? Thank you again!

[secorey]

Hi Professor Misic, In my limited exposure to neuroscience, I remember learning about how for some people with portions of their brain removed, other parts of their brain learn to execute the same functions that the missing piece used to control. Is this related to your research and, if so, how has it informed your knowledge of the nature of structure-function relationships? Thanks for coming to present this paper. Sam

[LynetteDang]

Hi Professor Misic, thank you for sharing your work with us. I have a couple of questions:

1. how do different types of brain activity (e.g. resting-state vs. task-based) relate to brain networks?
2. how do individual differences in SC and FC relate to differences in cognitive function, personality, and behavior?
3. what are the ethical implications of studying structural connectivity in the brain, particularly with regard to privacy and informed consent?

Thanks!

[bmisic]

Dear Professor Misic, thank you for sharing your research with us. Great to have some Canadian representation at our workshop! A few questions:

Functional Connectivity emerges from higher-order interactions, i.e., structure is linked to function via complex associations, but their correlation is imperfect at best. Estimating FC across time seems logical, but how can these associations be estimated across individuals (presumably) within a single time slice? Is the underlying assumption that neural activity in different individuals corresponds to different “stages”, analogous to estimation across time?

The paper presupposes structural connections, and indicates that functional connections emerge from higher-order interactions. How can we be sure about the directionality of this link? This is the sociologist in me, so this might not apply to neuroscience, but could structure equally emerge from function?

Great question! We've actually done just such a study, looking at how frame-by-frame fluctuations in functional co-activation relate to the underlying structural connectivity:

https://doi.org/10.1038/s42003-022-03466-x

More broadly, all the methods typically used to study structure-function coupling can be applied to individuals, and there is an emerging literature on how individual differences in structure-function coupling relate to behaviour, cognition, etc, and how they are affected by disease.

[bmisic]

Hi Professor Misic, In my limited exposure to neuroscience, I remember learning about how for some people with portions of their brain removed, other parts of their brain learn to execute the same functions that the missing piece used to control. Is this related to your research and, if so, how has it informed your knowledge of the nature of structure-function relationships? Thanks for coming to present this paper. Sam

There is indeed a perpetual interplay between structure and function, and this is why the notion of polysynaptic signaling in brain networks is so important. Namely, a complex network can support multiple functional configurations, so some perturbations can theoretically be compensated for by alternative signaling patterns.

[Thiyaghessan]

Dear Professor Misic,

Thank you for the talk! In your paper you briefly mention reservoir computing at the tail end:

"Particularly exciting are advances in artificial intelligence, such as reservoir computing, that make it possible to implement connectomes as artificial neural networks"

I was wondering if you could elaborate on how reservoir computing will allow us to bridge the gap between neural networks as understood today in deep learning and biological neural networks in the human brain.

Thank you!

[kuitaiw]

Dear Prof Misic, Neuroeconomics is a discipline that attempts to understand the neural mechanisms underlying economic behavior. So I would like to ask whether we can use techniques such as brain imaging to study the neural activity of individuals when making economic decisions? If so, what exactly should we do?

[borlasekn]

Prof. Misic - thank you for sharing your expertise! I appreciate hearing from perspectives outside of my own field and enjoying thinking about how work can translate between fields. That being said, as an individual interested in educational research, your section on "Individual Differences in Structure-Function Coupling" was particularly interesting. Individual-specific neuroscience is important in child development, especially in understanding how life events effect brain function during youth development! You mention that respecting individual differences is a major question for future research studies on the structure-function relationship. Would you mind giving some examples on how this research would be carried out, and what key insights you'd anticipate from this research? Thanks!

[taizeyu]

Dear Prof Misic,

Thank you for sharing your research. This is a very profound paper. I am curious that how can we apply the technology mentioned in the paper to the world. And in what areas can it be applied?

Thank you very much

[JoeHelbing]

Professor Misic,

I also have very limited exposure to Neuroscience, so the concept of functional vs structural connectivity is sort of hard to grasp. My best understanding from the paper was that SC is what neurons are connected to what neurons, whereas FC is what neurons work together with what neurons? Apologies is that is massively off-base. It's interesting that these would not be 1:1, as a non-neruo person, more SC more FC would make sense to me.

Could you give a bit more background for why this is an area of study? I'm making an assumption here that it is more "fundamental" science, laying foundations for more applied work?

[zoeyjiao1104]

Hi Professor Misic,

Thank you so much for sharing your work. I was curious about how you would address the ethical issues that could potentially arise in your study in terms of participant consent. In addition, could you please elaborate on the real world application of your study. Thank you!

[sushanz]

Dear Professor Misic, I appreciate your time and thanks for sharing! Really interesting work you have done and presented in the paper. As we are aware, deep learning has made remarkable progress in the field of cognitive science. People deploy today's techniques to study human behavior and brain function. You mentioned in your paper that the current models lack biological details therefore hinder the functional ability to fully emulate the brain and predict functions.However, I am curious to know why this field is relevant if we can never gather enough information to create a flawless model? Especially, there are many inexplicable and mental processes involved in the behavior we study, such as the availability heuristic, which has both biological and cognitive influences. Although without the biological details, some of the techniques today has already been able to replicate the process and taught our machine to recognize and recall relevant patterns

[edelahayeUChicago]

Dear Professor,

I come from an economics background so I won't pretend that I understand most of your paper's contents. I'm interested though what the key takeaways in terms of high level behaviour as might be analysed in other social and behavioural sciences? Is this something that you could see becoming a more fundamental direction of research in how social behaviour is determined?

Thanks, Elliot

[awaidyasin]

Dear Prof., thanks for sharing your work! On page 306, you mention, "Perhaps the simplest way to link structure and function is statistically. Varying forms of reduced rank regression have emerged as particularly useful, including canonical correlation [52] and partial least squares". I was wondering how would the empirical evidence (i.e., the underlying dataset) be generated for such applications given that this area studies quite specific connections in the brain.

[zhiyun0707]

Hi Professor Misic, thanks for sharing your work with us! As I am not familiar with neuroscience, I found it interesting when you talk about the structural connectivity and functional connectivity and linking them together statistically. Could you please explain more on the practical applications for your work?

[xin2006]

Hi Professor Misic, thanks for sharing your work with us! As the paper focuses primarily on macroscale brain networks, I am wondering how to understand the role and the importance of the connection in the complete picture of the brain's activity? And what makes the Macroscale part being captured in the paper instead of the smaller-scale connections between individual neurons?

[helyap]

Hi Professor Misic,

Thank you for sharing your work with us. I'm not familiar with neuroscience, but I am very curious about the methodology employed. In particular, I'm curious about the evidence and the data gathering process that would be needed to support arguments regarding structural-functional mappings in brain networks especially as it relates to the sparse yet requisite biological detail. For example, if MRIs are needed to collect mappings of neural activity, what would be the require sample size for the claim to be accepted by the neuroscience community? I imagine high time and price costs in running these studies in addition to the challenge of participant recruitment. Furthermore, are there thoughts on any potential heterogeneity in the structural-functional mappings in the data collection process?

[AlexBWilliamson]

Dear Dr. Misic

Thank you for sharing your research with us! I know that the brain is very poorly understood by doctors, and I was wondering what sort of practical implications work like this might have on medicine or other fields? For example, could a better understanding of the brain help us to better understand mental illness or neurological diseases?

[Yuxin-Ji]

Dear Professor Misic, Thank you so much for sharing your work with us. I was curious about what are some implications of your findings? If our understanding towards structure-function relationship increases, are we able to get a better prediction? In other words is the current inaccuracy in prediction caused mainly by the data, method, or theory? Thank you!

[ValAlvernUChic]

Hello Professor Misic,

Thank you for sharing your work! Neuroscience is quite foreign to me so I might've missed this in the paper; how exactly is neuromodulation modeled/visualized? If I understand it correctly to be the "texture" or even congestion of the neural pathways, how does one get an image of this? Thank you!

[LuZhang0128]

Hi Professor Misic,

Thank you for the amazing work! I'm not familiar with the field. I am particularly interested in the "Individual Differences in Structure–Function Coupling" section. I wonder how do individual differences in structural and functional connectivity impact the relationship between structure and function in the brain? Meanwhile, what are the implications for understanding individual variability in brain function and dysfunction? Thank you!

[hazelchc]

Hello Prof. Misic,

Thank you for sharing your fascinating work with us. I am interested in the reasons behind studying the link between structural connectivity and functional connectivity in the brain, and how these findings can be applied in practical settings. Thank you!

[beilrz]

Hello Professor

Thank you for your work. I was wondering does neuroplasticity suggest a somewhat independences between brain's structure and its functions (since the damage to one part of the brain will not affect its overall functioning much) ; or does it suggest that brain's structure is somewhat uniform, so that the damaged part can easily be replaced by other part.

[JunoWuu]

Hi Prof. Misic,

This is such an interesting topic! I am really interested in studying regional homogeneity. However, it does not attracts as much attention as connectivity analysis. It would be great to know more about how can we actually make sense of the regional homogeneity and how it connects to functions.

[BaotongZh]

Hi Professor Misic,

Thank you for sharing your work. As a student of computational social science, I am intrigued by the potential societal and economic implications that could result from your findings.

[Coco-Jiachen-Yu]

Hello Professor Misic,

The paper is really amazing! Thank you for sharing and I look forward to your presentation tomorrow. I was wondering how big the individual differences in structure-function relationship would be. Do you think a big deviation from the average population would imply biological or psychology pathology?

[Emily-fyeh]

Hello Prof. Misic,

Thank you for sharing your work with us. I would love to know more about the link between structural connectivity and functional connectivity in human brains--what are their potential applications in the real-world setting? Thank you!

[linhui1020]

Hi Prof. Misic, thanks for sharing your work! I agree with the point that nn model could do much better if it could function more like how our brain works. Are there any gender or age differences in the link between structural and functional connectivity in the brain? What are some potential applications of the findings from this area of research in the field of artificial intelligence?

[yhchou0904]

Dear Prof. Misic,

Thank you for sharing your idea with us! Though I realize the importance of understanding the relationship between the organizational features of neuronal networks and the spectrum of cortical functions, I am still wondering how understanding of this relationship could help us make better decisions or realize the world.

[tangn121]

Hello Professor Misic, thanks for sharing your work with us. I understand that investigating the correlation between structural and functional networks on an individual level can be quite challenging. However, I am still curious if you have any insights on how to tackle this issue at an individual level.

[ShiyangLai]

Hello Professor Misic, thank you for sharing your interesting work. I am not familiar with neuroscience, however, I am intrigued to learn more about your insights on the structural and functional differences between human brain neuron networks and AI neuron networks.

[zihua-uc]

Hi Prof. Misic,

I find it really interesting that the extent of alignment between brain structure and function is correlated with individual differences in cognitive flexibility. Is there any research that looks at how to treat and reduce this source of variation across individuals?

[Hongkai040]

Hi Prof. Misic, thank you for sharing your work! I wonder do our brains have structures of redundancy and randomness? I don't know much about neuroscience, but I can't help think that if it's the truth that our brains are sophisticatedly programmed by genes that all parts are delicate and well-functioned. If the prev answer is yes, does that mean we cannot find a perfect mapping from structures to functions?

[yunshu3112]

Hi Professor Misic,

Thank you for sharing! I wonder how can this study facilitate the artificial intelligence development? Do you think the future of AI is to become closer and closer to the real human brain or diverge from how the human brain functions?

[yujing-syj]

Hi Professor Misic, thanks so much for sharing this interesting topics with us. I am not very familar with this field. Could you give us some other examples and your insights about how to apply this study into the social science realm?

[JerryCG]

Dear Prof. Misic,

How could we relate this kind of biological findings to the social society? Do these specific brain functions/structures lead to some social/cultural patterns or can explain some social issues?

Best, Jerry Cheng

[ChongyuFang]

Hi Prof. Misic for sharing your work with us! How do the distinct attributes of structural connectivity networks, such as high local clustering, short path length, and rich club organization, relate to the organization and specialization of neural circuits in the brain, and how can this be quantified and modeled using network neuroscience approaches? Thanks

[shenyc16]

Hi Professor Misic, thank you for sharing this interesting work with us. The paper provides a comprehensive overview of current research on the relationship between the structure and function of the brain's macroscopic network, highlighting the importance of this relationship for understanding brain function and dysfunction. Looking forward to your lecture tomorrow!

[shaangao]

1. Does structural connectivity only include direct connections, but not indirect ones? Perhaps some functional connectivities result from indirect connections?
2. Functional connectivity can differ a lot across tasks. Did you find any difference in the percentage of functional connectivity captured by structural connectivity across tasks, and what might be the mechanism?

[cgyhumble0612]

Hi professor, thanks for sharing us this interesting research. I have questions about application. How might the approach of studying structure-function relationships be applied in other fields beyond neuroscience? How might a better understanding of this lead to advancements in fields? Thank you!

[fiofiofiona]

Hi Dr. Misic, thank you for sharing your work. I am curious about the direction of connectivity in FC specifically -- in some cognitive "tasks", one region may trigger the activity in another region, while in some other cases, multiple regions are coactivated. I wonder if the edges between nodes in an FC would be annotated as uni/bidirectional, and why or why not.

[yutaili]

Thanks for sharing your work Dr. Misic. One question I have about this paper is that how do you propose to address the challenge of integrating data from different imaging modalities and time scales to create a more unified model of brain network structure and function? Are there any emerging techniques or approaches that hold promise for overcoming these challenges? Thanks.

[y8script]

Hello Prof. Misic, thank you for sharing your work with us! I wonder how far we currently are from getting a functional neuroimaging scan with sufficient resolution as a typical anatomical scan. For example, nowadays 7T fMRI machines seem to be capable of looking at different cortical layers. As the neuroimaging techniques get more advanced, is it possible that we can directly explore the structural organization of the brain with fine-grained functional time series data?

[pranathiiyer]

Thanks so much for sharing your work, professor!

i was wondering how generalizable can this idea of the relationship between structure and function be generalized across populations? How inclusive can such theories be for people with disabilities ? For that matter, could it also vary for people with different education levels etc?

[HongzhangXie]

Hi Professor Misic, thank you so much for presenting interesting research. I am little background in neuroscience, but I am very impressed by the discussion of the relationship between brain function and structure. I am interested in the individual difference in structural-function coupling. How can we make causal inferences analysis about brain structure-function heterogeneity by adding individual differences into the model (for example, gender, age, or medical history data)?

[AlexPrizzy]

Thanks for presenting this work to the workshop, Bratislav! The mapping of structural to functional networks may be exciting to researchers. But does network neuroscience face a similar issue of modularity that's seen throughout the field of neuroscience?

[zbchen0129]

Hi Professor Misic, thank you so much for presenting interesting research. My questions are below:

1. Can you explain the concept of multiplexed models, and how they contribute to a better understanding of structure-function coupling in the brain? 2.How can brain networks annotated with micro-architectural features help in studying structure-function coupling?

[iefis]

Hi Professor Misic, Thank you for sharing your work with us. I would like to learn more about the alignment between structural and functional configurations in individuals. It seems that individual variation in functional boundaries makes comparing individuals a challenging task. Could you elaborate more on how to make use of individual differences in stucture-function coupling in quantifying the effects of manipulations amd perturbations? Thank you!

[erweinstein]

Hi Dr. Misic,

As a non-neuroscientist, the most interesting part of your research is how strongly we see that, "the modern theory of structure–function relationships in the brain is still in its infancy" (as you and your co-authors put it). For example, you note the significance of the paper by Finn et. al. published in late 2019 that demonstrated layer-dependence/depth-dependence in prefrontal cortex activity during working memory tasks. It would seem (again, as an outsider to the field) like such dependence would be a very logical, even obvious property to look for, yet Finn and her colleagues only demonstrated it less than 4 years ago. I assume technological advances (e.g., of MRIs) is part of the reason. As I feel that you have conveyed to us, this is a very exciting time to work in advanced computational methods for neuroscience, where such big discoveries are "out there" awaiting people to put in the focus and (usually, years of) work. (Like the way economics and Bayesian statistics didn't take off in their modern forms until the 1980s.)