blood oxygenation level dependent (bold) noise and ... · (bold) noise and behavior maurizio...
TRANSCRIPT
Blood Oxygenation Level Dependent (BOLD) noise and behavior
Maurizio Corbetta
Washington University St.Louis
attention and brain recovery labDepartment of Neurology, Radiology, Anatomy and Neurobiology
Cold Spring Harbor April 3-6, 2011
Tuesday, May 24, 2011
+
Blood FlowGlucoseUtilization
OxygenUtilization
OxygenAvailabilityor BOLD
Brain imaging 1985-2005
Fox and Raichle, 1986Tuesday, May 24, 2011
+
Tuesday, May 24, 2011
+
Stimulation Control
__ =
Tuesday, May 24, 2011
+
Tuesday, May 24, 2011
+
ON OFF ON OFF ON
Tuesday, May 24, 2011
BRAIN2% body weight
20% body energy
anatomy functionSibson et al. 1997, 1998; Shulman et al. 2001, 2004; Ames 2000, Attwell & Laughlin 2001, Lennie 2003, Wong-Riley 1989
Tuesday, May 24, 2011
BRAIN2% body weight
20% body energy
anatomy functionSibson et al. 1997, 1998; Shulman et al. 2001, 2004; Ames 2000, Attwell & Laughlin 2001, Lennie 2003, Wong-Riley 1989
Tuesday, May 24, 2011
BRAIN Task-evoked activityBRAIN
2% body weight20% body energy
anatomy functionSibson et al. 1997, 1998; Shulman et al. 2001, 2004; Ames 2000, Attwell & Laughlin 2001, Lennie 2003, Wong-Riley 1989
Tuesday, May 24, 2011
BRAIN Task-evoked activity
Spontaneous activity70-80% of which 90% signaling
BRAIN2% body weight
20% body energy
anatomy functionSibson et al. 1997, 1998; Shulman et al. 2001, 2004; Ames 2000, Attwell & Laughlin 2001, Lennie 2003, Wong-Riley 1989
Tuesday, May 24, 2011
Resting state functional MRI (Washington University, St. Louis)Tuesday, May 24, 2011
Hacker, Leuthardt, Corbetta unpublishedTuesday, May 24, 2011
He et al., PNAS 2008
BOLD functional connectivitymay correlate with slow cortical potentials (<0.5 Hz) and low frequency fluctuations of gamma band limited power
Tuesday, May 24, 2011
fMRI temporal correlation (functional connectivity)in the motor network
Biswal, Magn.Res.Imag.Medicine,1996Tuesday, May 24, 2011
fMRI temporal correlation (functional connectivity)in the motor network
Biswal, Magn.Res.Imag.Medicine,1996Tuesday, May 24, 2011
fMRI temporal correlation (functional connectivity)in the motor network
Biswal, Magn.Res.Imag.Medicine,1996Tuesday, May 24, 2011
Dorsal and Ventral Attention Networks
Corbetta & Shulman, Ann.Rev.Neurosci. 2011
Tuesday, May 24, 2011
Dorsal and Ventral Attention Networks
Corbetta & Shulman, Ann.Rev.Neurosci. 2011
Tuesday, May 24, 2011
Dorsal and Ventral Attention Networks
Corbetta & Shulman, Ann.Rev.Neurosci. 2011
Tuesday, May 24, 2011
Dorsal and Ventral Attention Networks
Corbetta & Shulman, Ann.Rev.Neurosci. 2011
Tuesday, May 24, 2011
AG
mPFCPCC
Perceptual search-‐related BOLD deac8va8ons
L R
Lateral
Medial
Z score-‐5 -‐3
Memory search-‐related BOLD ac8va8ons
L R
AG
mPFCPCC
Z score 3 5-‐5 -‐3
Default-mode networkEpisodic memory
Fox, et al. (2005) Proc. Natl. Acad. Sci. USA 102, 9673-9678Sestieri, et al. (2011) J.Neurosci.
Tuesday, May 24, 2011
AG
mPFCPCC
Perceptual search-‐related BOLD deac8va8ons
L R
Lateral
Medial
Z score-‐5 -‐3
Memory search-‐related BOLD ac8va8ons
L R
AG
mPFCPCC
Z score 3 5-‐5 -‐3
Default-mode networkEpisodic memory
Fox, et al. (2005) Proc. Natl. Acad. Sci. USA 102, 9673-9678Sestieri, et al. (2011) J.Neurosci.
Tuesday, May 24, 2011
About 90% cerebral cortex is organized in networks of spontaneous activity
3T ICA separation
Mantini & Corbetta, unpublished
dorsal attentionventral attentiondefault-mode
auditoryvisual
somatomotorlanguage
Tuesday, May 24, 2011
Functional connectivity in macaque and mouse
Vincent et al. Nature 2007
Culver et al. PLOS 2011dorsal attention default
Tuesday, May 24, 2011
Summary
Tuesday, May 24, 2011
Spontaneous activity is organized in large-scale spatiotemporal patterns (or functional connectivity) that resemble functional networks recruited during tasks.
Summary
Tuesday, May 24, 2011
Spontaneous activity is organized in large-scale spatiotemporal patterns (or functional connectivity) that resemble functional networks recruited during tasks.
Functional connectivity is a higher probability state to which the cortex goes back at rest.
Summary
Tuesday, May 24, 2011
Spontaneous activity is organized in large-scale spatiotemporal patterns (or functional connectivity) that resemble functional networks recruited during tasks.
Functional connectivity is a higher probability state to which the cortex goes back at rest.
Origin of spontaneous activity correlation patterns?
Summary
Tuesday, May 24, 2011
Spontaneous activity is organized in large-scale spatiotemporal patterns (or functional connectivity) that resemble functional networks recruited during tasks.
Functional connectivity is a higher probability state to which the cortex goes back at rest.
Origin of spontaneous activity correlation patterns?Anatomy vs. Function?
Summary
Tuesday, May 24, 2011
BOLD functional connectivity:a physiological index of anatomical connectivity?
Hagman et al. PLOS 2008Honey et al PNAS 2009
Tuesday, May 24, 2011
BOLD functional connectivity:a physiological index of anatomical connectivity?
Hagman et al. PLOS 2008Honey et al PNAS 2009
Gosh et al. PLOS 2008; Jirsa et al. JNeurosci 2009; Deco et al PNAS 2010; McIntosh & Deco, Nature
Rev.Neurosci. 2011
Connectome + delays generates spontaneous fluctuations
Tuesday, May 24, 2011
Honey et al. PNAS 2009
Structural connectivity (SC) does not predict resting state functional connectivity (rsFC)
Tuesday, May 24, 2011
Anatomical Connectivity
Tuesday, May 24, 2011
Anatomical Connectivity
FunctionalConnectivity
Tuesday, May 24, 2011
• New patterns of covariance reflecting prior experience
• Strength of covariance change related to behavioral performance
• Functional connectivity as a potential correlate of predisposition and individual variability
If functional connectivity networks are not simply hard-wired, but partly reflect
the history of prior activation (fire-wire), then training should ‘sculpt’ the
functional architecture at rest
Tuesday, May 24, 2011
Target Presentation
Array Presentation
Fixation Spot
TrainedVisual Quadrant
Trained Shape
Paradigm and behavior
Lewis et al. PNAS 2009
Tuesday, May 24, 2011
Training specific for the location and target shape
Tuesday, May 24, 2011
Training specific for the location and target shape
Tuesday, May 24, 2011
Training specific for the location and target shape
Tuesday, May 24, 2011
Design
fMRI localizer
visual cortexperceptual learning fMRI taskfMRI rest 1 fMRI rest 2
Tuesday, May 24, 2011
Design
fMRI localizer
visual cortexperceptual learning fMRI task
Visual cortex
fMRI rest 1 fMRI rest 2
Tuesday, May 24, 2011
Design
fMRI localizer
visual cortexperceptual learning fMRI task
Visual cortexTrained-novel shapes
fMRI rest 1 fMRI rest 2
Tuesday, May 24, 2011
Lewis et al. PNAS 2009
Task-evoked activity: learning enhances retinotopic activity for trained over novel target
trained shapesnovel shapes
Tuesday, May 24, 2011
Pre-Post
att.vis. unatt.vis. DAN DMN
Pre-learning Post-learning
DAN &attendedVis.Cx
DMN &unattended
Vis.Cx
att.vis. unatt.vis. DAN DMN att.vis. unatt.vis. DAN DMN
Tuesday, May 24, 2011
Pre-Post
att.vis. unatt.vis. DAN DMN
Pre-learning Post-learning
DAN &attendedVis.Cx
DMN &unattended
Vis.Cx
att.vis. unatt.vis. DAN DMN att.vis. unatt.vis. DAN DMN
Tuesday, May 24, 2011
%B
OL
D c
han
ge%
BO
LD
ch
an
ge
MR frames
Trained Visual Cortex
FEF
Lewis et al PNAS 2009
PRE-LEARNINGno correlation
POST-LEARNINGnegative correlation
Tuesday, May 24, 2011
%B
OL
D c
han
ge%
BO
LD
ch
an
ge
MR frames
Trained Visual Cortex
FEF
Lewis et al PNAS 2009
PRE-LEARNINGno correlation
POST-LEARNINGnegative correlation
Tuesday, May 24, 2011
Trained Visual Cortex
RFEFLFEF
RSPL
Untrained Visual CortexUntrained
Visual Cortex
Untrained Visual Cortex RAngG
RpCingLpCing
LmPFC
Learning-dependent changes in functional connectivity
Tuesday, May 24, 2011
Trained Visual Cortex
RFEFLFEF
RSPL
Untrained Visual CortexUntrained
Visual Cortex
Untrained Visual Cortex RAngG
RpCingLpCing
LmPFC
Learning-dependent changes in functional connectivity
Tuesday, May 24, 2011
post<prepost>pre
right V3A/LO-to-right FEF right V1-V3-to-right FEF
right FEF-to-right V1-V3 right FEF-to-right LO
decrease GCincrease GC
post-learningBottom-up:
decreased V1/V2>FEF increased V3A/LO>FEF
Top-dow:increased FEF>V1/V2decreaed FEF>V3/LO
decrease GCincrease GC
post-learning
Tuesday, May 24, 2011
post<prepost>pre
right V3A/LO-to-right FEF right V1-V3-to-right FEF
right FEF-to-right V1-V3 right FEF-to-right LO
decrease GCincrease GC
post-learningBottom-up:
decreased V1/V2>FEF increased V3A/LO>FEF
Top-dow:increased FEF>V1/V2decreaed FEF>V3/LO
decrease GCincrease GC
post-learning
Tuesday, May 24, 2011
Summary
Tuesday, May 24, 2011
Perceptual learning induces specific (retinotopic, regional) modulations of both task-evoked and resting functional connectivity.
Summary
Tuesday, May 24, 2011
Perceptual learning induces specific (retinotopic, regional) modulations of both task-evoked and resting functional connectivity.
Both task-evoked and functional connectivity changes correlate with stronger learning, and are consistent psychophysically with the development of a template for the target.
Summary
Tuesday, May 24, 2011
Perceptual learning induces specific (retinotopic, regional) modulations of both task-evoked and resting functional connectivity.
Both task-evoked and functional connectivity changes correlate with stronger learning, and are consistent psychophysically with the development of a template for the target.
Functional connectivity may represents a prior for recruitment of task-specific cortical circuitries (predisposition).
Summary
Tuesday, May 24, 2011
Perceptual learning induces specific (retinotopic, regional) modulations of both task-evoked and resting functional connectivity.
Both task-evoked and functional connectivity changes correlate with stronger learning, and are consistent psychophysically with the development of a template for the target.
Functional connectivity may represents a prior for recruitment of task-specific cortical circuitries (predisposition).
Learning-dependent changes in functional connectivity may underlie a systems level memory of task activation (Perception: Lewis et al, 2009; Motor: Albert et al 2009; Memory: Tambini et al 2010).
Summary
Tuesday, May 24, 2011
Functional connectivityand individual predisposition
Baldassare et al Current Biology in press
Tuesday, May 24, 2011
Functional connectivity (FC) & Early Accuracy
FC pre-learningone region
FC pre-learningall regions
Task-evoked post-learning
FC pre-learning & accuracy
Tuesday, May 24, 2011
Functional connectivity (FC) & Early Accuracy
FC pre-learningone region
FC pre-learningall regions
Task-evoked post-learning
FC pre-learning & accuracy
Tuesday, May 24, 2011
Functional connectivity (FC) & Early Accuracy
FC pre-learningone region
FC pre-learningall regions
Task-evoked post-learning
FC pre-learning & accuracy
Tuesday, May 24, 2011
Functional connectivity (FC) & Early Accuracy
FC pre-learningone region
FC pre-learningall regions
Task-evoked post-learning
FC pre-learning & accuracy
Tuesday, May 24, 2011
Functional connectivity (FC) & Early Accuracy
FC pre-learningone region
FC pre-learningall regions
Task-evoked post-learning
FC pre-learning & accuracy
Tuesday, May 24, 2011
Local vs. behaviorally relevant functional connectivity (FC)
Local FCVisual Cortex
FC-BehaviorVisual Cortex
Local FCAuditory Cortex
FC-BehaviorAuditory Cortex
Tuesday, May 24, 2011
Local vs. behaviorally relevant functional connectivity (FC)
Local FCVisual Cortex
FC-BehaviorVisual Cortex
Local FCAuditory Cortex
FC-BehaviorAuditory Cortex
Tuesday, May 24, 2011
Local vs. behaviorally relevant functional connectivity (FC)
Local FCVisual Cortex
FC-BehaviorVisual Cortex
Local FCAuditory Cortex
FC-BehaviorAuditory Cortex
Tuesday, May 24, 2011
Functional connectivity (FC) & Overall SensitivityFC pre-learning
one regionFC pre-learning
all regions
Task-evoked post-learning
FC pre-learning & sensitivity
Tuesday, May 24, 2011
Functional connectivity (FC) & Overall SensitivityFC pre-learning
one regionFC pre-learning
all regions
Task-evoked post-learning
FC pre-learning & sensitivity
Tuesday, May 24, 2011
Functional connectivity (FC) & Overall SensitivityFC pre-learning
one regionFC pre-learning
all regions
Task-evoked post-learning
FC pre-learning & sensitivity
Tuesday, May 24, 2011
Summary
Tuesday, May 24, 2011
Functional connectivity may represent a prior for recruitment of task-specific cortical circuitries (predisposition).
Summary
Tuesday, May 24, 2011
Functional connectivity may represent a prior for recruitment of task-specific cortical circuitries (predisposition).
-positive correlation between cognitive functions (language, memory, saliency, intelligence)(Hampson et al. 2006, 2007; Kelley et al 2008).
Summary
Tuesday, May 24, 2011
Functional connectivity may represent a prior for recruitment of task-specific cortical circuitries (predisposition).
-positive correlation between cognitive functions (language, memory, saliency, intelligence)(Hampson et al. 2006, 2007; Kelley et al 2008).
The relationship with learning or predisposition indicates a potential link with synaptic mechanisms of plasticity.
Summary
Tuesday, May 24, 2011
Current Biology, 2009
Tuesday, May 24, 2011
Tuesday, May 24, 2011
Tuesday, May 24, 2011
Tuesday, May 24, 2011
Tuesday, May 24, 2011
Tuesday, May 24, 2011
Tuesday, May 24, 2011
Tuesday, May 24, 2011
Tuesday, May 24, 2011
Acknowledgments
NINDS, NIMH, McDonnell FoundationEU Marie Curie Chair @ ITAB University of Chieti Italy
AttentionTony JackChad SylvesterGaurav PatelSerguei Astafiev
Gordon ShulmanMarcus RaichleAvi Snyder
FcMRI healthy-strokeMike FoxBiyu HeAlex Carter
Gian Luca RomaniCosimo Del GrattaClaudio Babiloni
fMRI-EEGDante MantiniGianni Perrucci
FcMRI LearningChris LewisAntonello Baldassare
TMS-EEGPaolo Capotosto
St.LouisChieti
Tuesday, May 24, 2011
Functional connectivity & development
Fair et al PLOS 2009Tuesday, May 24, 2011
Development of core task set and dorsal attention network(age 7-27)
Fair et al., PNAS, 2007
Tuesday, May 24, 2011
Tuesday, May 24, 2011
Hippocampal replay
Tuesday, May 24, 2011
III. Facilitation of neural communication
Tuesday, May 24, 2011
Behavioral performance follows phase of infra-slow EEG
Monto et al. JNeurosci 08Tuesday, May 24, 2011
Slow (0.1 Hz) MEG power fluctuations correlate between distant cortical regions
(De Pasquale et al., PNAS in press)
fMRI
MEGstat
MEGnon stat
Tuesday, May 24, 2011
43
Tuesday, May 24, 2011
He et al.Neuron, 2007
Tuesday, May 24, 2011
He et al.Neuron, 2007
Tuesday, May 24, 2011
He et al.Neuron, 2007
Tuesday, May 24, 2011
He et al.Neuron, 2007
Tuesday, May 24, 2011
1012 neurons
Total number of synapses in the cerebral cortex~2x1016
inputs5-10x103
outputs5-10x103
Optic nerve:5-10x109
synapses
Corticospinaltract:
5-10x109
synapsesTuesday, May 24, 2011
1012 neurons
Total number of synapses in the cerebral cortex~2x1016
inputs5-10x103
outputs5-10x103
Tuesday, May 24, 2011
1012 neurons
Total number of synapses in the cerebral cortex~2x1016
inputs5-10x103
outputs5-10x103
0.1 pixel
Cortical synapses=198,000 pixels
Optic nerve CST
Tuesday, May 24, 2011