Come and join us! 
Monday mornings at 9:30 am Amsterdam time,
we discuss exciting new science from last week with colleagues,
students and everyone interested in the brain.

Connect live via Zoom: LINK
ID: 895 8114 6322, Code: CNS
Previous events are available OnDemand on YouTube:  www.youtube.com/c/ClinicalNeuroanatomySeminars

Paper link: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002176 Abstract Music is core to human experience, yet the precise neural dynamics underlying music perception remain unknown. We analyzed a unique intracranial electroencephalography (iEEG) dataset of 29 patients who listened to a Pink Floyd song and applied a stimulus reconstruction approach previously used in the speech domain. We successfully reconstructed a recognizable song from direct neural recordings and quantified the impact of different factors on decoding accuracy. Combining encoding and decoding analyses, we found a right-hemisphere dominance for music perception with a primary role of the superior temporal gyrus (STG), evidenced a new STG subregion tuned to musical rhythm, and defined an anterior–posterior STG organization exhibiting sustained and onset responses to musical elements. Our findings show the feasibility of applying predictive modeling on short datasets acquired in single patients, paving the way for adding musical elements to brain–computer interface (BCI) applications. Presenter: Lilit Dulyan, PhD student at Donders Institute

Paperlink: https://www.science.org/doi/10.1126/science.adc8810#elettersSection Other papers mentioned: https://pubmed.ncbi.nlm.nih.gov/2064027/ https://www.nature.com/articles/nature14432 https://pubmed.ncbi.nlm.nih.gov/810497/ https://pubmed.ncbi.nlm.nih.gov/3782501/ https://pubmed.ncbi.nlm.nih.gov/6831534/

Paperlink: https://www.nature.com/articles/s41586-023-06031-6#article-info Learnable latent embeddings for joint behavioural and neural analysis Mapping behavioural actions to neural activity is a fundamental goal of neuroscience. As our ability to record large neural and behavioural data increases, there is growing interest in modelling neural dynamics during adaptive behaviours to probe neural representations1,2,3. In particular, although neural latent embeddings can reveal underlying correlates of behaviour, we lack nonlinear techniques that can explicitly and flexibly leverage joint behaviour and neural data to uncover neural dynamics3,4,5. Here, we fill this gap with a new encoding method, CEBRA, that jointly uses behavioural and neural data in a (supervised) hypothesis- or (self-supervised) discovery-driven manner to produce both consistent and high-performance latent spaces. We show that consistency can be used as a metric for uncovering meaningful differences, and the inferred latents can be used for decoding. We validate its accuracy and demonstrate our tool’s utility for both calcium and electrophysiology datasets, across sensory and motor tasks and in simple or complex behaviours across species. It allows leverage of single- and multi-session datasets for hypothesis testing or can be used label free. Lastly, we show that CEBRA can be used for the mapping of space, uncovering complex kinematic features, for the production of consistent latent spaces across two-photon and Neuropixels data, and can provide rapid, high-accuracy decoding of natural videos from visual cortex.

Paper link: https://www.nature.com/articles/d41586-023-02600-x?utm_medium=Social&utm_campaign=nature&utm_source=Twitter&error=cookies_not_supported&code=8dfe6066-e8d0-489d-b773-e88597275d77#Echobox=1692967976 3D brain atlas: https://atlases.ebrains.eu/viewer/#/a:juelich:iav:atlas:v1.0.0:1/t:minds:core:referencespace:v1.0.0:dafcffc5-4826-4bf1-8ff6-46b8a31ff8e2/p:juelich:iav:atlas:v1.0.0:6/rn:74cbdb6/@:0.0.0.-W000.._eCwg.2-FUe3._-s_W.2_evlu..7LIy..0.14gY0~.14gY0..1LSm New divisions in the inferior frontal sulcus: https://www.sciencedirect.com/science/article/pii/S0010945222001071?via%3Dihub

https://www.pnas.org/doi/10.1073/pnas.2218565120 Human and chimpanzee shared and divergent neurobiological systems for general and specific cognitive brain functions A long-standing topic of interest in human neurosciences is the understanding of the neurobiology underlying human cognition. Less commonly considered is to what extent such systems may be shared with other species. We examined individual variation in brain connectivity in the context of cognitive abilities in chimpanzees (n = 45) and humans in search of a conserved link between cognition and brain connectivity across the two species. Cognitive scores were assessed on a variety of behavioral tasks using chimpanzee- and human-specific cognitive test batteries, measuring aspects of cognition related to relational reasoning, processing speed, and problem solving in both species. We show that chimpanzees scoring higher on such cognitive skills display relatively strong connectivity among brain networks also associated with comparable cognitive abilities in the human group. We also identified divergence in brain networks that serve specialized functions across humans and chimpanzees, such as stronger language connectivity in humans and relatively more prominent connectivity between regions related to spatial working memory in chimpanzees. Our findings suggest that core neural systems of cognition may have evolved before the divergence of chimpanzees and humans, along with potential differential investments in other brain networks relating to specific functional specializations between the two species.

The anatomy of the brain necessarily constrains its function, but precisely how remains unclear. The classical and dominant paradigm in neuroscience is that neuronal dynamics are driven by interactions between discrete, functionally specialized cell populations connected by a complex array of axonal fibres1,2,3. However, predictions from neural field theory, an established mathematical framework for modelling large-scale brain activity4,5,6, suggest that the geometry of the brain may represent a more fundamental constraint on dynamics than complex interregional connectivity7,8. Here, we confirm these theoretical predictions by analysing human magnetic resonance imaging data acquired under spontaneous and diverse task-evoked conditions. Specifically, we show that cortical and subcortical activity can be parsimoniously understood as resulting from excitations of fundamental, resonant modes of the brain’s geometry (that is, its shape) rather than from modes of complex interregional connectivity, as classically assumed. We then use these geometric modes to show that task-evoked activations across over 10,000 brain maps are not confined to focal areas, as widely believed, but instead excite brain-wide modes with wavelengths spanning over 60 mm. Finally, we confirm predictions that the close link between geometry and function is explained by a dominant role for wave-like activity, showing that wave dynamics can reproduce numerous canonical spatiotemporal properties of spontaneous and evoked recordings. Our findings challenge prevailing views and identify a previously underappreciated role of geometry in shaping function, as predicted by a unifying and physically principled model of brain-wide dynamics. Paper link: https://www.nature.com/articles/s41586-023-06098-1 Some critical voices: - Landmark story shows that brain activity is spatiotemporally low-pass. https://t.co/1dK9lQojpv I am shocked! - https://twitter.com/SaadJbabdi/status/1668597401299369986 - https://twitter.com/KordingLab/status/1667170639340347392

Paper link: https://www.cell.com/patterns/fulltext/S2666-3899(23)00101-0 Connectome-based machine learning models are vulnerable to subtle data manipulations Neuroimaging-based predictive models continue to improve in performance, yet a widely overlooked aspect of these models is “trustworthiness,” or robustness to data manipulations. High trustworthiness is imperative for researchers to have confidence in their findings and interpretations. In this work, we used functional connectomes to explore how minor data manipulations influence machine learning predictions. These manipulations included a method to falsely enhance prediction performance and adversarial noise attacks designed to degrade performance. Although these data manipulations drastically changed model performance, the original and manipulated data were extremely similar (r = 0.99) and did not affect other downstream analysis. Essentially, connectome data could be inconspicuously modified to achieve any desired prediction performance. Overall, our enhancement attacks and evaluation of existing adversarial noise attacks in connectome-based models highlight the need for counter-measures that improve the trustworthiness to preserve the integrity of academic research and any potential translational applications.

Paper link: https://link.springer.com/article/10.1007/s12144-023-04698-8

A brain–computer interface that decodes continuous language from non-invasive recordings would have many scientific and practical applications. Currently, however, non-invasive language decoders can only identify stimuli from among a small set of words or phrases. Here we introduce a non-invasive decoder that reconstructs continuous language from cortical semantic representations recorded using functional magnetic resonance imaging (fMRI). Given novel brain recordings, this decoder generates intelligible word sequences that recover the meaning of perceived speech, imagined speech and even silent videos, demonstrating that a single decoder can be applied to a range of tasks. We tested the decoder across cortex and found that continuous language can be separately decoded from multiple regions. As brain–computer interfaces should respect mental privacy, we tested whether successful decoding requires subject cooperation and found that subject cooperation is required both to train and to apply the decoder. Our findings demonstrate the viability of non-invasive language brain–computer interfaces.

Cross-movie prediction of individualized functional topography Participant-specific, functionally-defined brain areas are usually mapped with functional localizers and estimated by making contrasts between responses to single categories of input. Naturalistic stimuli engage multiple brain systems in parallel, provide more ecologically plausible estimates of real-world statistics, and are friendly to special populations. The current study shows that cortical functional topographies in individual participants can be estimated with high fidelity from naturalistic stimuli. Importantly, we demonstrate that robust, individualized estimates can be obtained even when participants watched different movies, were scanned with different parameters/scanners, and were sampled from different institutes across the world. Our results create a foundation for future studies that allow researchers to estimate a broad range of functional topographies based on naturalistic movies and a normative database, making it possible to integrate high-level cognitive functions across datasets from laboratories worldwide. Paper link: https://www.biorxiv.org/content/10.1101/2022.11.21.517253v1 Paper on Hyperalignment: https://elifesciences.org/articles/56601

We discussed what special issues are, how they work, and why they are helpful.

The Cerebellum: Adaptive Prediction for Movement and Cognition Multidisciplinary evidence indicates a role for the cerebellum in various aspects of cognition. Due to its uniform cytoarchitecture and extensive reciprocal connections with frontal, parietal, and temporal associative cortices, theorists have sought to identify cerebellar computations that are universal across sensorimotor and associative processes. Two key concepts are prediction and error-based learning. Recent work has revealed physiological diversity across structurally similar cerebellar modules. The computational constraints that arise from this diversity may be important for understanding cerebellar processing in different functional domains. Knowledge has substantially evolved on cerebellar involvement in language and social cognition, providing representative domains to evaluate functional hypotheses of the ‘cognitive’ cerebellum and to consider how disturbances of cerebellar function may contribute to developmental and neuropsychiatric disorders. Over the past 30 years, cumulative evidence has indicated that cerebellar function extends beyond sensorimotor control. This view has emerged from studies of neuroanatomy, neuroimaging, neuropsychology, and brain stimulation, with the results implicating the cerebellum in domains as diverse as attention, language, executive function, and social cognition. Although the literature provides sophisticated models of how the cerebellum helps refine movements, it remains unclear how the core mechanisms of these models can be applied when considering a broader conceptualization of cerebellar function. In light of recent multidisciplinary findings, we examine how two key concepts that have been suggested as general computational principles of cerebellar function- prediction and error-based learning- might be relevant in the operation of cognitive cerebro-cerebellar loops. Paper link: https://doi.org/10.1016/j.tics.2017.02.005

Native language differences in the structural connectome of the human brain Is the neuroanatomy of the language structural connectome modulated by the life-long experience of speaking a specific language? The current study compared the brain white matter connections of the language and speech production network in a large cohort of 94 native speakers of two very different languages: an Indo-European morphosyntactically complex language (German) and a Semitic root-based language (Arabic). Using high-resolution diffusion-weighted MRI and tractography-based network statistics of the language connectome, we demonstrated that German native speakers exhibited stronger connectivity in an intra-hemispheric frontal to parietal/temporal dorsal language network, known to be associated with complex syntax processing. In comparison, Arabic native speakers showed stronger connectivity in the connections between semantic language regions, including the left temporo-parietal network, and stronger inter-hemispheric connections via the posterior corpus callosum connecting bilateral superior temporal and inferior parietal regions. The current study suggests that the structural language connectome develops and is modulated by environmental factors such as the characteristic processing demands of the native language. Paper link: 10.1016/j.neuroimage.2023.119955

Localization patterns of speech and language errors during awake brain surgery: a systematic review Awake craniotomy with direct electrical stimulation (DES) is the standard treatment for patients with eloquent area gliomas. DES detects speech and language errors, which indicate functional boundaries that must be maintained to preserve quality of life. During DES, traditional object naming or other linguistic tasks such as tasks from the Dutch Linguistic Intraoperative Protocol (DuLIP) can be used. It is not fully clear which speech and language errors occur in which brain locations. To provide an overview and to update DuLIP, a systematic review was conducted in which 102 studies were included, reporting on speech and language errors and the corresponding brain locations during awake craniotomy with DES in adult glioma patients up until 6 July 2020. The current findings provide a crude overview on language localization. Even though subcortical areas are in general less often investigated intraoperatively, still 40% out of all errors was reported at the subcortical level and almost 60% at the cortical level. Rudimentary localization patterns for different error types were observed and compared to the dual-stream model of language processing and the DuLIP model. While most patterns were similar compared to the models, additional locations were identified for articulation/motor speech, phonology, reading, and writing. Based on these patterns, we propose an updated DuLIP model. This model can be applied for a more adequate “location-to-function” language task selection to assess different linguistic functions during awake craniotomy, to possibly improve intraoperative language monitoring. This could result in a better postoperative language outcome in the future. Paperlink: https://link.springer.com/article/10.1007/s10143-022-01943-9

Paperlink: https://academic.oup.com/cercor/advance-article/doi/10.1093/cercor/bhac224/6609522 Dissociating the functional roles of arcuate fasciculus subtracts in speech production Recent tractography and microdissection studies have shown that the left arcuate fasciculus (AF)—a fiber tract thought to be crucial for speech production—consists of a minimum of 2 subtracts directly connecting the temporal and frontal cortex. These subtracts link the posterior superior temporal gyrus (STG) and middle temporal gyrus (MTG) to the inferior frontal gyrus. Although they have been hypothesized to mediate different functions in speech production, direct evidence for this hypothesis is lacking. To functionally segregate the 2 AF segments, we combined functional magnetic resonance imaging with diffusion-weighted imaging and probabilistic tractography using 2 prototypical speech production tasks, namely spoken pseudoword repetition (tapping sublexical phonological mapping) and verb generation (tapping lexical-semantic mapping). We observed that the repetition of spoken pseudowords is mediated by the subtract of STG, while generating an appropriate verb to a spoken noun is mediated by the subtract of MTG. Our findings provide strong evidence for a functional dissociation between the AF subtracts, namely a sublexical phonological mapping by the STG subtract and a lexical-semantic mapping by the MTG subtract. Our results contribute to the unraveling of a century-old controversy concerning the functional role in speech production of a major fiber tract involved in language. Discussed in contrast to Paper link: https://n.neurology.org/content/94/6/e594 Anatomical evidence of an indirect pathway for word repetition Objective To combine MRI-based cortical morphometry and diffusion white matter tractography to describe the anatomical correlates of repetition deficits in patients with primary progressive aphasia (PPA). Methods The traditional anatomical language model identifies a network for word repetition that includes Wernicke and Broca regions directly connected via the arcuate fasciculus. Recent tractography findings of an indirect pathway between Wernicke and Broca regions suggest a critical role of the inferior parietal lobe for repetition. To test whether repetition deficits are associated with damage to the direct or indirect pathway between both regions, tractography analysis was performed in 30 patients with PPA (64.27 ± 8.51 years) and 22 healthy controls. Cortical volume measurements were also extracted from 8 perisylvian language areas connected by the direct and indirect pathways. Results Compared to healthy controls, patients with PPA presented with reduced performance in repetition tasks and increased damage to most of the perisylvian cortical regions and their connections through the indirect pathway. Repetition deficits were prominent in patients with cortical atrophy of the temporo-parietal region with volumetric reductions of the indirect pathway. Conclusions The results suggest that in PPA, deficits in repetition are due to damage to the temporo-parietal cortex and its connections to Wernicke and Broca regions. We therefore propose a revised language model that also includes an indirect pathway for repetition, which has important clinical implications for the functional mapping and treatment of neurologic patients.

ChatGPT (Chat Generative Pre-trained Transformer) has caused a stir across many professions and has been a hot topic of debate across universities and publishing houses. The tool has exceeded expectations and even passed academic exams raising several ethical considerations about ownership, authorship, and plagiarism. We discuss these considerations and how the model was trained to be effective and politically correct, which also raised further critiques on potential biases and the method of training. Debate participants: - Stephanie Forkel, Ass. Professor at the Donders Institute, Netherlands - Lilit Dulyan, PhD student at the Donders Institute in the Netherlands - Chris Foulon, Postdoc at UCL in the UK - Michel Thiebaut de Schotten, Research director at CNRS in France - Barbara Molz, Postdoc at the Max Planck Institute, Netherlands - Ann-Katrin Ohlerth, Postdoc at the Max Planck Institute, Netherlands - Sabrina Beber, PhD student at the Donders Institute in the Netherlands - Marius Braunsdorf, Postdoc at the Donders Institute in the Netherlands - Barbara Eckstein, Donders Institute in the Netherlands - Joanne Kenney, Senior Clinical Scientist, UK - Eva Guzman Chacon, Master's student at Radboud University, Netherlands - Anna Matsulevits, PhD student at the University of Bordeaux, France - Ambra Ferrari, Postdoc at the Max Planck Institute, Netherlands - Giacomo Bignardi, Postdoc at the Max Planck Institute, Netherlands - Danny Thewissen, School teacher in the Netherlands - Valentina Pacella, Ass. Professor at the University of Padua, Italy #teaching #research #innovation #ai #techinnovation #chatgpt #ethics #trainingdata #rewardlearning #DAN #publishing #universities #cnsdebate #cnsnightcap ---- Chapters ----- 0:00 Introduction 2:09 Would you pay for ChatGPT? 8:30 What's under the hood: a reward learning model 9:35 Ethics and training datasets 9:52 Political Correctness: Trump vs Biden 11:02 DAN roleplay 12:04 What jobs might Chat replace 15:08 Editorial debate about plagiarism 16:48 Grammarly vs chat 21:34 Journal cover letter 25:55 Interim summary 32:14 Teaching innovation using AI 51:55 GPT competitors

Transfer Learning Approaches for Neuroimaging Analysis: A Scoping Review Deep learning algorithms have been moderately successful in diagnoses of diseases by analyzing medical images especially through neuroimaging that is rich in annotated data. Transfer learning methods have demonstrated strong performance in tackling annotated data. It utilizes and transfers knowledge learned from a source domain to target domain even when the dataset is small. There are multiple approaches to transfer learning that result in a range of performance estimates in diagnosis, detection, and classification of clinical problems. Therefore, in this paper, we reviewed transfer learning approaches, their design attributes, and their applications to neuroimaging problems. We reviewed two main literature databases and included the most relevant studies using predefined inclusion criteria. Among 50 reviewed studies, more than half of them are on transfer learning for Alzheimer's disease. Brain mapping and brain tumor detection were second and third most discussed research problems, respectively. The most common source dataset for transfer learning was ImageNet, which is not a neuroimaging dataset. This suggests that the majority of studies preferred pre-trained models instead of training their own model on a neuroimaging dataset. Although, about one third of studies designed their own architecture, most studies used existing Convolutional Neural Network architectures. Magnetic Resonance Imaging was the most common imaging modality. In almost all studies, transfer learning contributed to better performance in diagnosis, classification, segmentation of different neuroimaging diseases and problems, than methods without transfer learning. Among different transfer learning approaches, fine-tuning all convolutional and fully-connected layers approach and freezing convolutional layers and fine-tuning fully-connected layers approach demonstrated superior performance in terms of accuracy. These recent transfer learning approaches not only show great performance but also require less computational resources and time. Paperlink: https://www.frontiersin.org/articles/10.3389/frai.2022.780405/full

We had technical issues at the beginning resulting in the loss of the first few minutes (mainly the introduction of our presenter Majd Abdallah @majdabd33, a postdoc at the MIND Team at INRIA in Paris) Default-mode network streams for coupling to language and control systems The human brain is organized into large networks. One important brain network is the Default network, which enables cognitive functions such as social thinking, memory, and reward. In group-averaged data, this network emerges as a unitary whole, despite its involvement in multiple cognitive functions. Here, we tested whether Default networks found in individual humans, rather than group-average networks, contain organized substructure. In individuals, we consistently found nine subnetworks within the Default network. These subnetworks matched brain activity patterns during cognitive tasks. Some subnetworks resembled brain circuits involved in specific Default functions. Others linked Default network to other large networks. In summary, this study describes a set of brain circuits within the Default networks of individual humans. Paper Link: https://www.pnas.org/doi/full/10.1073/pnas.2005238117

How bilingual brains accomplish the processing of more than one language has been widely investigated by neuroimaging studies. The assimilation-accommodation hypothesis holds that both the same brain neural networks supporting the native language and additional new neural networks are utilized to implement second language processing. However, whether and how this hypothesis applies at the finer-grained levels of both brain anatomical organization and linguistic functions remains unknown. To address this issue, we scanned Chinese-English bilinguals during an implicit reading task involving Chinese words, English words and Chinese pinyin. We observed broad brain cortical regions wherein interdigitated distributed neural populations supported the same cognitive components of different languages. Although spatially separate, regions including the opercular and triangular parts of the inferior frontal gyrus, temporal pole, superior and middle temporal gyrus, precentral gyrus and supplementary motor areas were found to perform the same linguistic functions across languages, indicating regional-level functional assimilation supported by voxel-wise anatomical accommodation. Taken together, the findings not only verify the functional independence of neural representations of different languages, but show co-representation organization of both languages in most language regions, revealing linguistic-feature specific accommodation and assimilation between first and second languages. Paper link: https://www.nature.com/articles/s42003-023-04446-5#Sec20

Robust dynamic brain coactivation states estimated in individuals Paper link: https://www.science.org/doi/10.1126/sciadv.abq8566

Today's paper: https://osf.io/74z8m/ A primer on entropy in neuroscience: The role of the observer and the misconception of disorder by Erik D. FagerholmZalina DezhinaRosalyn MoranFederico E. Turkheimer, Robert Leech Join u live every Monday morning: https://www.clinicalneuroanatomyseminars.com/neuroccino

Everyone is welcome - join us every Monday morning at 9:30am Amsterdam time. The 4 papers directly related to the debate: Letter 1: Mesulam, M., Thompson, C., Weintraub, S., Rogalski, E., 2022. The Wernicke conundrum is misinterpreted. Brain, accepted Letter 2: Matchin, W., den Ouden, D.-B., Hickok, G., Hillis, A.E., Bonilha, L., Fridriksson, J., 2022. Reply: The Wernicke conundrum is misinterpreted. Brain, accepted Conundrum revisited: Matchin, W., den Ouden, D.-B., Hickok, G., Hillis, A.E., Bonilha, L., Fridriksson, J., 2022. The Wernicke conundrum revisited: evidence from connectome-based lesion-symptom mapping. Brain 145, 3916–3930. Original conundrum paper: Mesulam, M.-M., Thompson, C.K., Weintraub, S., Rogalski, E.J., 2015. The Wernicke conundrum and the anatomy of language comprehension in primary progressive aphasia. Brain 138, 2423–2437.

The neurons that restore walking after paralysis A spinal cord injury interrupts pathways from the brain and brainstem that project to the lumbar spinal cord, leading to paralysis. Here we show that spatiotemporal epidural electrical stimulation (EES) of the lumbar spinal cord1,2,3 applied during neurorehabilitation4,5 (EESREHAB) restored walking in nine individuals with chronic spinal cord injury. This recovery involved a reduction in neuronal activity in the lumbar spinal cord of humans during walking. We hypothesized that this unexpected reduction reflects activity-dependent selection of specific neuronal subpopulations that become essential for a patient to walk after spinal cord injury. To identify these putative neurons, we modelled the technological and therapeutic features underlying EESREHAB in mice. We applied single-nucleus RNA sequencing6,7,8,9 and spatial transcriptomics10,11 to the spinal cords of these mice to chart a spatially resolved molecular atlas of recovery from paralysis. We then employed cell type12,13 and spatial prioritization to identify the neurons involved in the recovery of walking. A single population of excitatory interneurons nested within intermediate laminae emerged. Although these neurons are not required for walking before spinal cord injury, we demonstrate that they are essential for the recovery of walking with EES following spinal cord injury. Augmenting the activity of these neurons phenocopied the recovery of walking enabled by EESREHAB, whereas ablating them prevented the recovery of walking that occurs spontaneously after moderate spinal cord injury. We thus identified a recovery-organizing neuronal subpopulation that is necessary and sufficient to regain walking after paralysis. Moreover, our methodology establishes a framework for using molecular cartography to identify the neurons that produce complex behaviours. Paper link: https://www.nature.com/articles/s41586-022-05385-7