scholarly journals Situating the Left-Lateralized Language Network in the Broader Organization of Multiple Specialized Large-Scale Distributed Networks

2019 ◽  
Author(s):  
Rodrigo M. Braga ◽  
Lauren M. DiNicola ◽  
Randy L. Buckner
Keyword(s):  
Large Scale ◽  
Temporal Cortex ◽  
Distributed Networks ◽  
Inferior Temporal Cortex ◽  
Association Cortex ◽  
Candidate Network ◽  
Motor Region ◽  
Classical Language ◽  
The Individual ◽  
Language Network

Using procedures optimized to explore network organization within the individual, the topography of a candidate language network was characterized and situated within the broader context of adjacent networks. The candidate network was first identified using functional connectivity and replicated across individuals, datasets, acquisition tasks, and analytic methods. In addition to classical language regions near to perisylvian cortex and temporal pole, additional regions were observed in dorsal posterior cingulate, midcingulate, anterior superior frontal and inferior temporal cortex. The candidate network was selectively activated when processing meaningful (as contrast to non-word) sentences, while spatially adjacent networks showed minimal or even decreased activity. Examined in relation to adjacent networks, the topography of the language network was found to parallel the motif of other association networks including the transmodal association networks linked to theory of mind and episodic remembering (often collectively called the default network). The several networks contained juxtaposed regions in multiple association zones. Outside of these juxtaposed higher-order networks, we further noted a distinct frontotemporal network situated between language regions and a frontal orofacial motor region and a temporal auditory region. A possibility is that these functionally-related sensorimotor regions might anchor specialization of neighboring association regions that develop into the language network. What is most striking is that the canonical language network appears to be just one of multiple similarly organized, differentially specialized distributed networks that populate the evolutionarily expanded zones of human association cortex.

scholarly journals Parallel Distributed Networks Resolved at High Resolution Reveal Close Juxtaposition of Distinct Regions

2018 ◽  
Author(s):  
Rodrigo M. Braga ◽  
Koene R. A. Van Dijk ◽  
Jonathan R. Polimeni ◽  
Mark C. Eldaief ◽  
Randy L. Buckner
Keyword(s):  
High Resolution ◽  
Large Scale ◽  
Spatial Separation ◽  
Distributed Networks ◽  
Association Cortex ◽  
Default Network ◽  
Network Organization ◽  
Parahippocampal Cortex ◽  
The Individual ◽  
Insight Into

Examination of large-scale distributed networks within the individual reveals details of cortical network organization that are absent in group-averaged studies. One recent discovery is that a distributed transmodal network, often referred to as the ‘default network’, is comprised of two separate but closely interdigitated networks, only one of which is coupled to posterior parahippocampal cortex. Not all studies of individuals have identified the same networks and questions remain about the degree to which the two networks are separate, particularly within regions hypothesized to be interconnected hubs. Here we replicate the observation of network separation across analytical (seed-based connectivity and parcellation) and data projection (volume and surface) methods in 2 individuals each scanned 31 times. Additionally, 3 individuals were examined with high-resolution fMRI to gain further insight into the anatomical details. The two networks were identified with separate regions localized to adjacent portions of the cortical ribbon, sometimes inside the same sulcus. Midline regions previously implicated as hubs revealed near complete spatial separation of the two networks, displaying a complex spatial topography in the posterior cingulate and precuneus. The network coupled to parahippocampal cortex also revealed a separate region directly within the hippocampus at or near the subiculum. These collective results support that the default network is composed of at least two spatially juxtaposed networks. Fine spatial details and juxta-positions of the two networks can be identified within individuals at high resolution, providing insight into the network organization of association cortex and placing further constraints on interpretation of group-averaged neuroimaging data.

scholarly journals Situating the left-lateralized language network in the broader organization of multiple specialized large-scale distributed networks

2020 ◽  
Vol 124 (5) ◽  
pp. 1415-1448 ◽  
Author(s):  
Rodrigo M. Braga ◽  
Lauren M. DiNicola ◽  
Hannah C. Becker ◽  
Randy L. Buckner
Keyword(s):  
Functional Connectivity ◽  
Large Scale ◽  
Distributed Networks ◽  
Task Demands ◽  
Association Cortex ◽  
Control Networks ◽  
Language Network

This research shows that a language network can be identified within individuals using functional connectivity. Organizational details reveal that the language network shares a common spatial motif with other association networks, including default and frontoparietal control networks. The language network is activated by language task demands, whereas closely juxtaposed networks are not, suggesting that similarly organized but differentially specialized distributed networks populate association cortex.

scholarly journals Parallel distributed networks dissociate episodic and social functions within the individual

2020 ◽  
Vol 123 (3) ◽  
pp. 1144-1179 ◽  
Author(s):  
Lauren M. DiNicola ◽  
Rodrigo M. Braga ◽  
Randy L. Buckner
Keyword(s):  
Theory Of Mind ◽  
Large Scale ◽  
Distributed Networks ◽  
Association Cortex ◽  
Default Network ◽  
Multiple Forms ◽  
Social Functions ◽  
Temporoparietal Junction ◽  
Parahippocampal Cortex ◽  
The Individual

Association cortex is organized into large-scale distributed networks. One such network, the default network (DN), is linked to diverse forms of internal mentation, opening debate about whether shared or distinct anatomy supports multiple forms of cognition. Using within-individual analysis procedures that preserve idiosyncratic anatomical details, we probed whether multiple tasks from two domains, episodic projection and theory of mind (ToM), rely on the same or distinct networks. In an initial experiment (6 subjects, each scanned 4 times), we found evidence that episodic projection and ToM tasks activate separate regions distributed throughout the cortex, with adjacent regions in parietal, temporal, prefrontal, and midline zones. These distinctions were predicted by the hypothesis that the DN comprises two parallel, interdigitated networks. One network, linked to parahippocampal cortex (PHC), is preferentially recruited during episodic projection, including both remembering and imagining the future. A second juxtaposed network, which includes the temporoparietal junction (TPJ), is differentially engaged during multiple forms of ToM. In two prospectively acquired independent experiments, we replicated and triplicated the dissociation (each with 6 subjects scanned 4 times). Furthermore, the dissociation was found in all zones when analyzed independently, including robustly in midline regions previously described as hubs. The TPJ-linked network is interwoven with the PHC-linked network across the cortex, making clear why it is difficult to fully resolve the two networks in group-averaged or lower-resolution data. These results refine our understanding of the functional-anatomical organization of association cortex and raise fundamental questions about how specialization might arise in parallel, juxtaposed association networks. NEW & NOTEWORTHY Two distributed, interdigitated networks exist within the bounds of the canonical default network. Here we used repeated scanning of individuals, across three independent samples, to provide evidence that tasks requiring episodic projection or theory of mind differentially recruit the two networks across multiple cortical zones. The two distributed networks thus appear to preferentially subserve distinct functions.

scholarly journals Parallel Distributed Networks Dissociate Episodic and Social Functions Within the Individual

2019 ◽  
Author(s):  
Lauren M. DiNicola ◽  
Rodrigo M. Braga ◽  
Randy L. Buckner
Keyword(s):  
Large Scale ◽  
Distributed Networks ◽  
Association Cortex ◽  
Multiple Forms ◽  
Social Functions ◽  
Temporoparietal Junction ◽  
Cortical Organization ◽  
Functional Regions ◽  
Parahippocampal Cortex ◽  
The Individual

Association cortex is organized into large-scale distributed networks. One such network, the default network (DN), is linked to diverse forms of internal mentation, opening debate about whether shared anatomy supports multiple forms of cognition. Alternatively, subtle distinctions in cortical organization could remain to be resolved. Using within-individual analysis procedures that preserve idiosyncratic details of cortical anatomy, we probed whether multiple tasks from two domains - Episodic Projection and Theory of Mind (ToM) - rely upon the same or distinct networks. In an initial experiment (n=6, subjects scanned 4 times each), we found evidence that Episodic Projection and ToM tasks activate distinct functional regions distributed throughout cortex, with adjacent regions in parietal, temporal, prefrontal and midline zones. These distinctions were predicted by the hypothesis that the DN comprises two parallel, interdigitated networks. One network, linked to parahippocampal cortex (PHC), is preferentially recruited during Episodic Projection, including both remembering the past and imagining the future. A second juxtaposed network, which includes the temporoparietal junction (TPJ), is differentially engaged during multiple forms of ToM tasks. The TPJ-linked network is interwoven with the PHC-linked network in multiple zones, including the posterior and anterior midline, making clear why it is difficult to fully resolve the two networks in group-averaged or lower-resolution data. We replicated all aspects of this network dissociation in a second, prospectively acquired dataset (n=6). These results refine our understanding of the functional-anatomical organization of association cortex as well as raise questions about how functional specialization might arise in parallel, juxtaposed association networks.

scholarly journals Attractor dynamics in networks with learning rules inferred from in vivo data

2017 ◽  
Author(s):  
Ulises Pereira ◽  
Nicolas Brunel
Keyword(s):  
Neural Network ◽  
Hebbian Learning ◽  
Temporal Cortex ◽  
Memory Storage ◽  
Inferior Temporal Cortex ◽  
Association Cortex ◽  
Visual Responses ◽  
Neural Network Models ◽  
Firing Rates ◽  
Learning Rules

AbstractThe attractor neural network scenario is a popular scenario for memory storage in association cortex, but there is still a large gap between models based on this scenario and experimental data. We study a recurrent network model in which both learning rules and distribution of stored patterns are inferred from distributions of visual responses for novel and familiar images in inferior temporal cortex (ITC). Unlike classical attractor neural network models, our model exhibits graded activity in retrieval states, with distributions of firing rates that are close to lognormal. Inferred learning rules are close to maximizing the number of stored patterns within a family of unsupervised Hebbian learning rules, suggesting learning rules in ITC are optimized to store a large number of attractor states. Finally, we show that there exists two types of retrieval states: one in which firing rates are constant in time, another in which firing rates fluctuate chaotically.

scholarly journals Large-scale hyperparameter search for predicting human brain responses in the Algonauts challenge

2019 ◽  
Author(s):  
Kamila M. Jozwik ◽  
Michael Lee ◽  
Tiago Marques ◽  
Martin Schrimpf ◽  
Pouya Bashivan
Keyword(s):  
Large Scale ◽  
Visual Stimuli ◽  
Temporal Cortex ◽  
Image Features ◽  
Inferior Temporal Cortex ◽  
Network Architectures ◽  
Image Preprocessing ◽  
Neural Representations ◽  
Brain Responses ◽  
Early Visual Cortex

Image features computed by specific convolutional artificial neural networks (ANNs) can be used to make state-of-the-art predictions of primate ventral stream responses to visual stimuli.However, in addition to selecting the specific ANN and layer that is used, the modeler makes other choices in preprocessing the stimulus image and generating brain predictions from ANN features. The effect of these choices on brain predictivity is currently underexplored.Here, we directly evaluated many of these choices by performing a grid search over network architectures, layers, image preprocessing strategies, feature pooling mechanisms, and the use of dimensionality reduction. Our goal was to identify model configurations that produce responses to visual stimuli that are most similar to the human neural representations, as measured by human fMRI and MEG responses. In total, we evaluated more than 140,338 model configurations. We found that specific configurations of CORnet-S best predicted fMRI responses in early visual cortex, and CORnet-R and SqueezeNet models best predicted fMRI responses in inferior temporal cortex. We found specific configurations of VGG-16 and CORnet-S models that best predicted the MEG responses.We also observed that downsizing input images to ~50-75% of the input tensor size lead to better performing models compared to no downsizing (the default choice in most brain models for vision). Taken together, we present evidence that brain predictivity is sensitive not only to which ANN architecture and layer is used, but choices in image preprocessing and feature postprocessing, and these choices should be further explored.

scholarly journals Tomita H, Ohbayashi M, Nakahara K, Hasegawa I, Miyashita Y

1999 ◽  
Vol 7 (6) ◽  
pp. E14
Author(s):  
William T. Couldwell
Keyword(s):  
Prefrontal Cortex ◽  
Executive Control ◽  
Temporal Cortex ◽  
Inferior Temporal Cortex ◽  
Association Cortex ◽  
Long Term Memory ◽  
Top Down ◽  
Visual Inputs ◽  
Voluntary Recall ◽  
Show Evidence

Knowledge or experience is voluntarily recalled from memory by reactivation of the neural representations in the cerebral association cortex. In inferior temporal cortex, which serves as the storehouse of visual long-term memory, activation of mnemonic engrams through electric stimulation results in imagery recall in humans, and neurons can be dynamically activated by the necessity for memory recall in monkeys. Neuropsychological studies and previous split-brain experiments predicted that prefrontal cortex exerts executive control upon inferior temporal cortex in memory retrieval; however, no neuronal correlate of this process has ever been detected. Here we show evidence of the top-down signal from prefrontal cortex. In the absence of bottom-up visual inputs, single inferior temporal neurons were activated by the top-down signal, which conveyed information on semantic categorization imposed by visual stimulus-stimulus association. Behavioural performance was severely impaired with loss of the top-down signal. Control experiments confirmed that the signal was transmitted not through a subcortical but through a fronto-temporal cortical pathway. Thus, feedback projections from prefrontal cortex to the posterior association cortex appear to serve the executive control of voluntary recall.

rCBF-activatie en neuronale circuits gerelateerd aan visuele waarneming

1996 ◽  
Vol 8 (2) ◽  
pp. 35-39
Author(s):  
B.M. de Jong
Keyword(s):  
Information Processing ◽  
Temporal Cortex ◽  
Distributed Networks ◽  
Association Cortex ◽  
Top Down ◽  
Primary Sensory Cortex ◽  
Visual Areas ◽  
Simultaneous Activation ◽  
Remote Association

SummaryThree principles of neuronal interaction within cortically distributed networks are discussed. PET-rCBF activation methods provide an opportunity to acquire insight in the distribution of functionally related areas of the human brain in vivo. The distinction of visual areas, activated by either motion or color within an observed scenery, points at a segregation in neuronal information processing. Such a segregation extends into both a dorsal and a ventral route towards consequently the parietal and temporal cortex.Simultaneous activation over the dorsal and ventral route, which for example occurs in relation to the perception of complex motion (optic flow), or motion perception after lesion of V5, suggests integration by means of cross-connectivity. The third principle, i.e. “top-down” integration, appears by analysis of V5-V1 interaction, attentional effects on V4, frontal activation in prosopagnosia, and by analysis of hallucinations. Such “top-down” integration indicates the presence of momentaneous effect on cortical areas, intimately related to the primary sensory cortex, by neuronal activity of remote “association” cortex, the latter being connected by direct (synaps-restricted) bypass from early stations of information processing.

Neurocognitive stages of spatial cognitive mapping measured during free exploration of a large-scale virtual environment

2015 ◽  
Vol 113 (3) ◽  
pp. 740-753 ◽  
Author(s):  
Markus Plank ◽  
Joseph Snider ◽  
Erik Kaestner ◽  
Eric Halgren ◽  
Howard Poizner
Keyword(s):  
Template Matching ◽  
Large Scale ◽  
Cognitive Mapping ◽  
Temporal Cortex ◽  
Cognitive Maps ◽  
Inferior Temporal Cortex ◽  
Processing Negativity ◽  
Scalp Topography ◽  
Free Exploration ◽  
Frontal Midline Theta

Using a novel, fully mobile virtual reality paradigm, we investigated the EEG correlates of spatial representations formed during unsupervised exploration. On day 1, subjects implicitly learned the location of 39 objects by exploring a room and popping bubbles that hid the objects. On day 2, they again popped bubbles in the same environment. In most cases, the objects hidden underneath the bubbles were in the same place as on day 1. However, a varying third of them were misplaced in each block. Subjects indicated their certainty that the object was in the same location as the day before. Compared with bubble pops revealing correctly placed objects, bubble pops revealing misplaced objects evoked a decreased negativity starting at 145 ms, with scalp topography consistent with generation in medial parietal cortex. There was also an increased negativity starting at 515 ms to misplaced objects, with scalp topography consistent with generation in inferior temporal cortex. Additionally, misplaced objects elicited an increase in frontal midline theta power. These findings suggest that the successive neurocognitive stages of processing allocentric space may include an initial template matching, integration of the object within its spatial cognitive map, and memory recall, analogous to the processing negativity N400 and theta that support verbal cognitive maps in humans.

Sign in / Sign up

Export Citation Format

Share Document