Somatosensory and Motor Research, March–June 2011; 28(1–2): 11–18

ORIGINAL ARTICLE

Barrels XXIII: Barrels by the shore

WANYING ZHANG

& JOSHUA C. BRUMBERG

Department of Neuroscience, Columbia University, New York, NY, USA and

Department of Psychology,

Queens College, City University of New York, Flushing, NY, USA

(Received 12 January 2010; accepted 12 January 2010)

Abstract

The 23rd annual Barrels meeting was held on the University of California, San Diego campus and highlighted the latest

advances in the whisker-to-barrel pathway and beyond. The annual meeting brought together investigators from a dozen

countries to present their data in posters and short talks. The meeting focused on several themes, first the barrel system was

used as a model to study the consequences that result from alterations in the normal pattern(s) of development. A second

session focused on what happens to whisker information once it leaves the layer IV barrel. A third session addressed issues of

coding within the barrel system and a final session highlighted the latest advances in the engineering of transgenic mouse

lines. The meeting highlighted the utility of the barrel system to study cortical circuitry in the normal and pathological state.

Keywords: Whisker, vibrissa, barrel, meeting

Forty years after the initial publication of the

discovery of the cortical barrels, the 23rd annual

Barrels meeting convened in the Calit2 Auditorium

on the campus of the University of California, San

Diego on 11 and 12 November 2010. The meeting

welcomed 90 researchers from a dozen countries to

discuss the latest advances in the rodent sensorimo-

tor system (see Appendix 1 for the complete

schedule).

The opening session focused on how the barrel

system can serve as a model to investigate what can

happen behaviorally, genomically, physiologically,

and anatomically when the system is disrupted by

developmental disorders. Fan Wang (Duke

University, USA) introduced and moderated the

morning session, reminding the audience of the

devastating consequences on neurocognitive devel-

opment in humans that result from Rett syndrome,

Fragile X, and Reelin disorders.

Sacha Nelson (Brandeis University, USA)

focused on the physiological genomics of Rett

syndrome. Rett syndrome is the second most

common genetic cause of mental retardation and is

due to the specific mutation of the MeCp2 gene,

in addition to the neurocognitive sequelae there are

disruptions of the autonomic nervous system as well

as sleep and respiratory abnormalities. He focused on

determining the expression of genes found in specific

cellular phenotypes and how their misregulation

could lead to the disruption of neural circuits

which directly leads to the resultant pathology.

Mice who have had their MeCp2 gene deleted

recapitulate the disease and it was found that cortical

neurons from these animals showed a reduction of

excitatory currents and an increase in inhibitory

currents under spontaneous conditions and display

reduced long-term potentiation and smaller excita-

tory post-synaptic potentials when synaptic inputs

were stimulated. Interestingly, genomic analyses

revealed that cell adhesion molecules that are

important in the formation and maintenance of

synapses were disrupted in the knockout animals.

Andreas Frick (University Bordeaux 2, France)

focused his investigations on the most common

Correspondence: J. C. Brumberg, PhD, Department of Psychology, Queens College, CUNY, Flushing, NY 11568, USA. Tel: þ1 718 997 3541.

Fax: þ1 718 997 3257. E-mail: joshua.brumberg@qc.cuny.edu

ISSN 0899–0220 print/ISSN 1369–1651 online ß 2011 Informa Healthcare Ltd.

DOI: 10.3109/08990220.2011.563602

inherited mental disorder, Fragile X syndrome, using

mouse models where the Fmr1 gene had been

deleted. Behavioral testing showed that knockout

animals were able to complete a gap crossing task

quicker, due to fewer whisks needed to make the

decision to jump, and they were able to cross wider

gaps compared to wild type animals. Voltage-

sensitive dye experiments revealed more excitation

following whisker stimulation in the Fragile X

animals. A hallmark of Fragile X syndrome is

dendritic alterations including increases in spine

density. The FMRP protein (the product of the

Fmr1 gene) targets many dendritic ion channels,

in addition to cytoskeleton proteins, including K

(GIRK3, Kv 3.1, BkCa), Ca

2þ

(Cav1.3), and cation

(hyperpolarization activate cation current (HCN))

channels whose misregulation was hypothesized to

impact the neurons’ computational abilities.

Experiments revealed that neurons recorded in

slices from knockout animals had less membrane

‘‘sag’’ in response to hyperpolarizing current pulses,

were more excitable in response to depolarizing

pulses, and showed lower resonance frequencies and

greater ability to summate synaptic inputs due largely

to a decrease in HCN currents. Due to a reduction in

the large conductance Ca

2þ

-dependent K

current

which normally would put the brakes on runaway

excitation, there was increased excitability in the

dendrites of knockout mice. These results confirmed

the hypothesis that the neocortical pyramidal neu-

rons of Fragile X mice have altered computational

abilities which may underlie some of the neurocog-

nitive deficits observed in afflicted individuals.

Jochen Staiger (George-August-Universitat,

Germany) concluded the session describing his

laboratories’ investigations of the reeler mouse. The

audience was reminded that in reeler animals the

cortex develops ‘‘inside out’’ with neurons normally

fated to the deep layers being found closer to the pial

surface and the normally superficial neurons locating

closer to the white mater. Although the ventral

posterior medial nucleus of the thalamus is unaf-

fected in these animals their thalamocortical projec-

tions are more widely distributed throughout the

cortical column and do not aggregate into barrel-like

units. Using in situ hybridization and indicators for

supragranular, granular, and infragranular neurons

it was seen that the thalamocortical afferents still

targeted granular fated neurons even if they were not

located in their normal position within the cortical

column.

The session concluded with the realization that

developmental disorders through their alterations in

neural circuits directly lead to the observed neuro-

cognitive and other pathological phenotypes con-

fronted by clinicians. Only through careful dissection

of the underlying neural circuits will we be able to

understand how these alterations lead to the

observed pathologies and stimulate the development

of therapeutic interventions.

Following a brief coffee break the morning

resumed with a series of short talks moderated by

Arthur Houweling (Erasmus University, the

Netherlands). Leading off, Dudi Deutsch

(Weizmann Institute of Science, Israel) focused on

the variability inherent to rat whisking. Utilizing a

head-fixed preparation with all but three whiskers

clipped (C1, C2, D1) he and his colleagues

monitored free whisking and whisking in the

presence of an object placed in the whisking path.

It was found that whiskers on the same side of the

face were very correlated and less correlated with

whiskers on the opposite side of the face. In general,

the amplitude of the whisking was more variable than

the duration of the whisking cycle. When contacting

an object there was a large increase of amplitude on

the contralateral side with no significant impact on

whisking duration. Detailed analysis of individual

whisks revealed ‘‘pumps’’ wherein periods of nega-

tive velocity were observed during protraction in

25% of all contact whisks. They concluded that there

were separate neural controllers for whisking ampli-

tude and duration. From the same group, Ehud

Ahissar (Weizmann Institute of Science, Israel)

revealed the discovery of Homo rehovotus whiskerus

a graduate student outfitted with whisker-like appen-

dages attached to their fingers as a means of

capturing all the relevant variables in regards to a

sensory–motor task—‘‘whisking’’ and aperture width

discrimination. The synthetic whiskers were outfitted

with motion and force sensors. Increasing accuracy

was correlated with more coordinated movement.

It was found that changes in motor strategies by

themselves directly led to increased perceptual

resolution. The more ‘‘whisking’’ cycles (more

sensory sampling) the smaller differences in aperture

width which could be detected. In general, percep-

tion emerges from a convergence process including

sensory and motor variables. The session concluded

with Jason Ritt (Boston University, USA) arguing

for creating standards for the measurement and

reporting of whisker behaviors and their neural

correlates. The key issue confronting researchers is

how to measure the relevant variables given that little

is known about the mechanics of the vibrissa follicle

and how these variables might influence sensory

transduction. It was pointed out that it is still unclear

as to what behavioral tests should be conducted and

in what reference frame whisker angles should be

referenced. Finally, when using different tracking

strategies to follow whisker movements it was

suggested that estimation errors/assumptions should

be highlighted in the relevant sections of the

manuscripts.

12 W. Zhang & J. C. Brumberg

After the annual business meeting (chaired by

Joshua C. Brumberg, Queens College, CUNY,

USA) and a scrumptious lunch, the afternoon

session of short talks began under the direction of

Karel Svoboda (Janelia Farms, USA). The session

was started off by Hiroshi Kawaski (The University

of Tokyo, Japan) who focused on the mechanism that

governs barrel formation. He compared the rate of

barrel formation in control animals vs those from

mothers who were induced to deliver 1 day early.

The size of the resultant litters and their body weights

and their brains’ cortical thickness were unaffected,

but the barrels of the preterm animals developed

earlier. The birthing process results in a decrease in

serotonin levels which accounted for the early barrel

development although the rate of development is the

same as in control animals. Next, Qian-Quan Sun

(University of Wyoming, USA) followed the pro-

gression of the balance of excitatory and inhibitory

(EI) inputs onto neurons throughout the barrel

column. Within the layer 4 barrel the EI ratio

decreased from postnatal day (PND) 8 to PND18

and then remained stable. In the supragranular layers

stable ratios were not obtained until PND30, layer 5

followed a similar developmental timeline to layer 4.

The decrease in the EI ratio was largely accounted

for by increases in inhibitory conductances. Mary

Ann Wilson (Johns Hopkins University, USA)

followed with a presentation on how perinatal

exposure to lead can impact spine development

within the barrel. Nursing mothers were exposed to

lead in their drinking water for the first 10 days of

their pups’ lives which resulted in a smaller overall

barrel field in the pups. Although there were no

changes in the total number of neurons or in their

dendritic architecture there was a decrease in the

branching of the thalamocortical axons and an

increase in spine density. Finally, Rodrigo Kujis

(Encephalogistics, Inc., USA) reminded the audi-

ence of the devastating consequences of Alzheimer’s

disease and the lack of therapeutic interventions.

Although primary sensory cortices had been thought

to be spared during Alzheimer’s disease, recent data

show that this is not the case and that lesions can be

found in the barrel cortex of mouse models of

Alzheimer’s disease suggesting that the barrel cortex

can serve as a model system for studying the

progression of the disease as well as for testing the

efficacy of potential treatments.

The final short talk session of the day was led by

Qian-Quan Sun (University of Wyoming, USA).

Starting off was Lu Li (Helen Wills Neuroscience

Institute, USA) who looked at rapid plasticity in the

barrel cortex using whole cell in vivo recordings. The

presentation highlighted two types of plasticity and

their distinct mechanisms; hebbian plasticity which

requires positive feedback vs homeostatic plasticity

which requires negative feedback. To probe for these

two types of plasticity in the barrel cortex, D row

whiskers were trimmed for 1–10 days starting at

PND20 and recordings were obtained in vivo from

supragranular neurons under urethane anesthesia. It

was found that there was a decrease in local field

potentials following 3, 5, or 10 days of deprivation

and following 3 days of deprivation there was an

increase in the magnitude of the long latency

response whereas after 5 days there was a decrease

in the overall number of action potentials in response

to principal whisker deflections. Whole cell record-

ings revealed that deprivation decreased overall

synaptic conductance with inhibitory conductances

decreasing more than excitatory which was consis-

tent with a model of rapid homeostatic plasticity

resulting from short-term deprivation in the supra-

granular layers. Craig Brown (University of

Victoria, Canada) focused on the role 4 nicotinic

receptors had on experience-driven cortical depres-

sion in the mouse somatosensory cortex using in vivo

voltage-sensitive dye imaging. The C1 whisker was

trimmed and then imaging commenced either 3, 7,

or 21 days post-trimming. It was found that after 3 or

7 days post-trimming there was a decrease in the

voltage-sensitive dye response to C1 deflections

which returned to normal by 21 days post-trimming.

It was determined that there was an increase in the

4 nicotinic receptor post-trimming which was

colocalized with the post-synaptic scaffolding protein

PSD95 and most GABAergic neurons had 4 puncta

in apposition to their somata. Application of 4

agonists depressed responses and if following trim-

ming 4 antagonists were continuously applied to the

barrel cortex via an osmotic pump the trimming-

induced depression was blocked. In sum, these

results suggest that 4 receptors on GABAergic

interneurons account for the majority of the trim-

ming-induced depression in the voltage-sensitive dye

response. The day ended with a stimulating poster

session and then a barbecue dinner listening to the

waves breaking on the shoreline in La Jolla, CA.

The second day of the conference opened with

a session focused on understanding the barrel cortex

as a part of a larger sensorimotor neural network.

Specifically, the question was posed, how is periph-

eral sensory information conveyed and propagated

beyond the layer IV barrel? David Kleinfeld

(University of California, San Diego, USA) intro-

duced and moderated this session, reminding the

audience that instead of an isolated computational

unit, the barrel cortex is strongly interconnected with

the nearby primary motor cortex (M1) as well as the

secondary somatosensory cortex (S2).

Carl Peterson (EPFL, Switzerland) focused on

how intrinsic brain states can modulate the whisker-

evoked responses of barrel cortex neurons in awake,

Barrels XXIII: Barrels by the shore 13

behaving mice. Voltage-sensitive dye (VSD) imaging

showed that cortical voltage responses of a passive

whisker stimulus were dramatically impacted by the

behavioral state of the animal. In awake and

quiescent animals, whisker stimulation elicited large

depolarizations in the barrel cortex that spread

almost immediately (30 ms post-stimulus) into both

M1 and S2. In contrast, in animals that were actively

whisking, passive stimulus elicited only small depo-

larizations that remained localized to the barrel

cortex. The VSD imaging results were confirmed

by intracellular patch-clamp recordings from layer

2/3 (L2/3) neurons. Peterson then investigated the

responses of cortical neurons to active-touch, or

object-contact during active whisking, using intra-

cellular patch-clamp recording. Recordings form

cortical neurons when rats where engaged in active-

touch elicited mixed excitatory and inhibitory con-

ductances which showed a large degree of variability

and action-potential firing was unreliable. However,

contact of object during active-touch increased the

correlation of sub-threshold Vm between neighbor-

ing cells. The second part of the talk focused on the

response of cortical neurons during a detection task.

Mice were trained to detect a magnetic deflection of

a single whisker for water rewards. Peterson showed

that the voltage responses of cortical L2/3 neurons

were highly dependent on the pre-stimulus EEG

state of the animal. Specifically, when the EEG is in a

slow-oscillatory state, the neurons show large stimu-

lus-elicited depolarization in Vm and when the EEG

is in a desynchronized state, neurons show very little

Vm responses. The performance of the animal was

only weakly correlated with the pre-stimulus brain

state, where the miss trials showed slightly higher

power in the low-frequency (1–5 Hz) oscillations.

Ron Frostig (University of California Irvine,

USA) focused his investigation on using the spread

of activity in barrel cortex as a general model for

mammalian sensorimotor cortex activation. He first

showed, using intrinsic imaging as well as extra-

cellular recording, that the cortical area activated by a

single-whisker periodic stimulus is about 15 mm

, far

larger than the area of a single barrel column (0.5

). The spread of cortical activation is radially

symmetrical, even when the center is near a border

between different sensory areas (e.g., S1 and primary

auditory cortex (A1)). To investigate the anatomical

basis for this spread of excitation, a tracer was

injected into L2/3 of individual barrels, and axons

local to the injection site were labeled and were

shown to radially spread several millimeters from the

center. To further study the underlying mechanism

of this non-specific local spread of excitation, Frostig

performed a series of experiments where the gray

matter between an array of eight extracellular

electrodes was transected. Local field potential

(LFP) signals in distant electrodes disappeared after

transection, demonstrating that the spread of activa-

tion is mediated by intracortical connections. Tracer

labeling was also attenuated by the transection.

Additional support for the large symmetrical spread

came from his study of cortical infarcts induced by

permanent occlusion of the middle cerebral artery

(pMCAO). When a single whisker was stimulated

immediately or 1 h after pMCAO, the cortex was

completely protected from the stroke. If whisker

stimulation commenced 2 h after pMCAO, 80% of

the animals showed complete cortical protection.

However, when a single whisker was stimulated 3 h

after pMCAO, the resulting infarct was larger than

what would have been obtained without stimulation,

and the resultant infarct matches the typical volume

of cortical activation evoked by a single whisker.

Imaging results similar to the ones obtained by

Frostig have been shown in other sensory cortical

areas such as A1 and V1 using either VSD or

intrinsic imaging. The anatomical results have also

been demonstrated in other regions of the sensor-

imotor cortex. Frostig therefore suggested that in

response to a point sensory stimulus, a large,

symmetrical sub-threshold spread of activation is a

fundamental motif of mammalian sensorimotor

cortex.

The session concluded with Tim Murphy

(University of British Columbia, Canada) discussing

his laboratory’s investigation of widespread patterns

of cortical activation under different conditions.

Using VSD imaging, he found that spontaneous

oscillation in Vm take the form of traveling waves in

both anesthetized and awake mice. The traveling

waves are highly synchronous between hemispheres

and correspond to increases in the power of the low-

frequency EEG. Areas along the midline typically

initiate the traveling cortical depolarization, and are

termed the ‘‘core of spontaneous activity’’. Recent

sensory experiences can affect the pattern of sponta-

neous activity by initiating the traveling wave near the

area recently activated by the sensory stimuli.

Activation motifs found during spontaneous travel-

ing waves were found to resemble the patterns of

activity generated by sensory stimuli. Interestingly,

sensory training stimuli can increase the amount of

match between spontaneous motifs and evoked

patterns of activity. Lastly, Murphy showed that, in

transgenic animals whose cortical layer 5b neurons

express channelrhodopsin (ChR), sensory stimuli-

evoked patterns of activity can be mimicked by local

light activation in the corresponding sensory cortical

area.

The session concluded with the realization that the

community must start viewing the barrel cortex as an

integral part of a larger sensorimotor cortical net-

work. Not only are there specific long-range

14 W. Zhang & J. C. Brumberg

reciprocal connections between barrel cortex and

motor cortex (as well as other higher sensory areas),

but there are also non-specific local connections

between primary sensory cortices of different mod-

alities. Only when carefully considering these short-

and long-range intracortical connections can the

ultimate role of the barrel cortex in performing

cortical computations be truly understood.

Following a brief coffee break the morning

resumed with a talk by Hongkui Zeng (Allen

Institute of Brain Sciences, USA) on the latest

advances in the engineering of transgenic mouse

lines. The speaker first reminded the audience of the

importance of transgenic technologies as tools to

dissect the function of specific cell types in complex

circuits. The Allen Institute has created a cell-type

selective mouse genetic toolkit system to facilitate the

study of neural circuits. The toolkit system includes

the generation of Cre-driver lines that target specific

cortical cell types, Cre-responder lines with strong

expression of fluorescent probes for neuronal tracing,

calcium imaging (gCAMP3), and membrane pro-

teins for photostimulation (Channelrhodopsin,

Halorhopsin, and Arch). A set of corresponding

AAV viral vectors will also be soon available, which

can be used in conjunction with the transgenic mice.

Experiments validating the functionality of these

probes are still ongoing, but preliminary results

suggest that investigators may soon be able to

purchase lines of mice to study specific efferent and

afferent pathways using optical means. Zeng also

described the Mouse Brain Connectivity Atlas

project, the aim of which is to create a whole-brain,

mid-to-high-resolution, three-dimensional map of

axonal projections in the mouse brain. The atlas

will be created primarily using genetic tracing

approaches that utilize a variety of cell-type specific

Cre lines and Cre-dependent viral tracers. Both the

raw data as well as the 3-D models will be presented

on a public online database to create 2-D and 3-D

searchable interactive tools and is hoped to be

completed by the end of 2012.

The morning session concluded with a quick data

blitz session moderated by Randy Bruno

(Columbia University, USA). Joshua C.

Brumberg (Queens College, CUNY, USA) pre-

sented a recent study investigating the physiological

properties of two types of corticothalamic (CT)

neurons: (1) the CT-VPM neurons and (2) the

CT-POM neurons. CT neurons are identified by

injections of retrograde tracers into either VPM or

POM. CT-VPM neurons are typically located in L6

of barrel cortex while CT-POM neurons are found

in L5b. Based on action potential (AP) half-width

measured by patch-clamp recordings of retrogradely

labeled neurons in thalamocortical slices, CT-VPM

neurons have broader APs compared to CT-POM

neurons. Both cell types receive thalamic inputs.

In response to a pair of thalamic stimulations, CT-

POM neurons showed response facilitation while

CT-VPM neurons showed neither facilitation nor

depression. Similarly, in response to a train of eight

pulses, CT-POM neurons continued to show

response facilitation, and CT-VPM neurons

showed clear response depression. Vincent Jacob

(Cardiff University, Wales) finished the data blitz

session by showing sensory deprivation can differen-

tially affect subtypes of pyramidal neurons in L5 of

barrel cortex. After single-row whisker deprivation

for 10 days in 6-week-old rats, responses to spared

whisker potentiated in L5 regular spiking (RS)

neurons but not in L5 intrinsically bursting (IB)

neurons.

After lunch, the afternoon session started with a

series of short talks moderated by Jason Ritt (Boston

University, USA). The series started off with Gaute

T. Einevoll (Norwegian University of Life Sciences,

Norway) who created a thalamocortical network

model of the barrel column using population firing

data. Multi-unit activity (MUA) and local field

potential (LFP) data were recorded with laminar

electrodes in the barrel cortex, and MUA recordings

were obtained with a single electrode in the

thalamus. Responses of both thalamic and cortical

neurons to a single-whisker stimulus with varying

velocity and amplitude were recorded. Firing rates

extracted from MUA recordings were in turn used to

identify population firing-rate models. He found that

the thalamus to L4 connection was best fitted by a

model with fast feed forward excitation from VPM

combined with a slower inhibitory process due to

feed forward and/or recurrent connections. The

model predicted that the thalamocortical circuit is

optimally stimulated by rapid changes in the thalamic

firing rate. Information transfer within the cortical

column, on the other hand, was better fitted with

a simple feed forward population firing-rate model

combined with linear or mixed linear–parabolic

activation function, which predicts that the intracor-

tical circuits are low-pass filters that respond best to

slow-varying inputs from cortical L4 neurons. Next

Robert Sachdev (Yale University, USA) discussed

the VSD response evoked in mouse barrel cortex by

single-whisker stimuli. He showed that in urethane

anesthetized mice, single-whisker stimulus elicits a

wave of depolarization. A longer whisker stimulus

caused a slower rise in the VSD signal and a

persistent depolarization because the on-response is

blending with the off-response. The persistent

depolarization cannot be mimicked by a pair of

whisker stimuli. Paired whisker stimuli at short

intervals (8 ms) do not generate a second response

but only evoke a larger spread of the initial cortical

depolarization. At longer intervals (100–200 ms), the

Barrels XXIII: Barrels by the shore 15

second stimuli produced a VSD response that is

slower to rise and slower to decay, but do not spread

as far as the initial response. Lastly, the extent of the

area of activation observed by VSD imaging is not

dependent on direction of the whisker stimuli.

Finally, Alison L. Barth (Carnegie Mellon

University, USA) presented the work of her labora-

tory developing an unbiased method to assess the

firing activity of individual neurons in the barrel

cortex. They used a fos-GFP transgenic mouse to

examine the properties of cells with a recent history

of elevated activity. Intriguingly, their investigation

has shown that neuronal responses are extremely

heterogeneous in vivo. In barrel cortex, only layer 4

neurons and a very small proportion of layer 2/3

neurons show Fos-promoted GFP expression in

response to sensory stimulation. In vitro patch-

clamp recording of fos-GFP-positive neurons were

not intrinsically more excitable than other neurons,

but receive more synaptic inputs and are highly

interconnected. She then performed dual in vivo

targeted cell-attached recording, and demonstrated

that fos-GFP-positive neurons spontaneously fire

more and earlier than fos-GFP-negative neurons.

Fos-GFP-positive neurons are also more likely to fire

in bursts. Lastly, whisker-trimming did not affect the

proportion of fos-GFP-positive neurons within the

population.

The final session of the day focused on encoding

and decoding of sensory inputs in the whisker–barrel

system and other primary sensory cortical areas.

David Golomb (Ben Gurion University, Israel)

introduced the session by reminding the audience of

the difficulties of interpreting sensory-stimulus-

evoked physiological data, and the existing debates

over the basis of cortical neural coding.

Garrett Stanley (Georgia Institute of

Technology, USA) started off the session by discuss-

ing how adaptation affects thalamocortical transfor-

mations of peripheral somatosensory information.

Using simultaneous single-unit recordings in the

VPM and the barrel cortex in anesthetized rats, they

discovered that adaptation differentially influences

the thalamus and cortex. In response to repeated

whisker stimuli (12 Hz) of various velocities, L4

cortical neurons adapt strongly in response ampli-

tude. Adaptation in turn increases the slope of the

velocity–sensitivity curve (angular velocity vs number

of APs in 30 ms window) of cortical neurons, making

it easier to discriminate different stimulus velocities

based on firing rate alone. VPM neurons, on the

other hand, were shown to adapt to repeated stimuli,

but less than cortical neurons. While the velocity–

sensitivity curve of VPM neurons does not change in

the adapted state, there is a significant decrease

in synchrony between VPM neurons. This decrease

in synchrony could account for the change in the

slope of the velocity–sensitivity curve of cortical L4

neurons. The second part of the talk involved his

more recent work involving observer analysis of VSD

imaging of cortex and anesthetized rats. In other

words, if an ideal observer was present could they

determine the nature of the sensory input based on

the observed VSD signal. The area of activation

detected by VSD imaging increases with whisker

stimulus velocity, therefore, for an ideal observer to

discriminate between stimulation of two neighboring

whiskers, there is an optimal stimulus velocity which

would still generate significant VSD signal but also

minimal amount of spread outside of the stimulated

barrel allowing for discrimination.

Next, Brent Dorion (University of Pittsburgh,

USA) presented several strategies by which a

neuronal network could be used to change output

correlations when there is shared input correlation.

Using a model of a network of neurons sharing a

common input source, and several non-shared noise

sources, he demonstrated that the background noise

level of a neuron can change the firing-rate gain

control of the cell, which in turn changes the output

noise correlation of the neurons. He further showed

that with high background noise, the output correla-

tion of the network showed higher synchrony with

a shorter window of analysis (3 ms) but lower

synchrony with longer window of analysis (over 10

ms). In vitro recording from pairs of cortical L2/3

neurons while injecting different levels of noisy

conductances using a dynamic clamp confirmed

this finding. A second strategy is the recruitment of

feed forward inhibitory circuits with higher input

amplitudes, which could create a negative image of

the input signal, which in turn decreases the amount

of output synchrony of the principal neurons. Dorion

further demonstrated that this strategy has been

utilized by the electrosensory system of the weakly

electric fish.

The meeting was concluded with a talk by Ken

Harris (Imperial College London, UK) focused on

the sensory-evoked population response of auditory

corticalneurons. Using extracellular multi-unit

recordings (MUA) from A1 of rats, he found that

the population responses of A1 neurons to a single

tune resembled spontaneous up-like events which

occur in the absence of sensory stimuli. Both

spontaneous and evoked events exhibited sparse,

spatially localized activity in L2/3 pyramidal neurons,

and densely distributed activity in larger L5 pyrami-

dal neurons and putative interneurons. The only

difference between spontaneous and evoked events

was their laminar propagation: spontaneous events

initiated in the deep layers and propagated upward

while evoked activity initiated in the thalamo-

recipient layers and then propagated throughout the

column. In both un-anesthetized and urethanized

16 W. Zhang & J. C. Brumberg

rats, global activity fluctuated between a ‘‘desyn-

chronized’’ state characterized by a low-amplitude,

high-frequency LFP and a ‘‘synchronized’’ state of

larger, lower frequency waves. Computational stu-

dies done in his lab suggested that cortical neuronal

responses could be modeled by a simple dynamical

system model fitted to the spontaneous activity

immediately preceding stimulus presentation. This

result reflected a nonlinear self-exciting system in

synchronized states and an approximately linear

system in desynchronized states.

After 2 days of stimulating talks, posters, and

discussions the 23rd annual Barrels meeting drew to

a close with a commitment to reconvene in a year’s

time in Baltimore, MD. The meeting once again

confirmed the utility of the barrel system as a model

to study not only cortical development and function-

ing, but also as a model to study neuropathology and

developmental disorders.

Acknowledgements

The Barrels organizing committee (Joshua C.

Brumberg (Chair), Randy Bruno, Mitra Hartmann,

David Kleinfeld, Jochen Staiger, and Mary Ann

Wilson) would like to thank Nancy Steinmetz and

Kathy Diekmann for their invaluable help.

Declaration of interest: The authors report no

conflicts of interest. The authors alone are respon-

sible for the content and writing of the paper.

Appendix 1. Barrels XXIII schedule

Thursday, 11 November

9:00–9:05 Welcome Joshua C. Brumberg, Queens College, CUNY

Introduction to local area David Kleinfeld, UCSD

9:05–11:15 Development Gone Haywire

9:05–9:15 Introduction/overview Fan Wang, Duke University

9:15–9:45 Sacha Nelson, Brandeis University, USA

Physiological genomics of Rett syndrome

9:45–10:15 Andreas Frick, University Bordeaux 2, France

Pathophysiology of information processing in neocortical circuits in Fragile X syndrome

10:15–10:45 Jochen Staiger, Georg-August-Universita

t, Go

ttingen, Germany

Remodeling of (thalamo-)cortical circuits in reeler mice showing severe neuronal migration defects

10:45–11:15 Discussion

11:15–11:45 Coffee Break

11:45–12:45 Short Platform Talks 1

Moderator: Arthur Houweling, Erasmus MC, the Netherlands

11:45–12:00 D. Deutsch

1

, M. Pietr

, P. M. Knutsen

, E. Ahissar

and E. Schneidman

Weizmann Institute of Science, Rehovot, Israel,

University of California, San Diego, USA

Closed loop whisking: effects of contacts

12:00–12:15 Avraham Saig



, Amos Arieli and Ehud Ahissar

Weizmann Institute of Science, Rehovot, Israel

Improving perceptual accuracy via motor and not sensory modifications

12:15–12:30 Jason Ritt

1

, Jason Wolfe

, Mitra Hartmann

, Ehud Ahissar

, Michael Brecht

and Christiaan de Kock

Boston University,

Humboldt University, Berlin,

Northwestern University,

Weizmann Institute of Science,

VU University,

Amsterdam

Establishing community standards for measurement and reporting of whisker behaviors and their neural correlates

12:30–12:45 Discussion

12:45–12:50 Business Meeting (Joshua C. Brumberg, Queens College, CUNY)

12:50–2:30 Lunch Break

2:30–3:45 Short Platform Talks 2

Moderator: Karel Svoboda, Janelia Farms

2:30–2:45 Tomohisa Toda, Itaru Hayakawa and Hiroshi Kawasaki



The University of Tokyo, Japan

The role of birth in somatosensory map formation in the mouse barrel cortex

2:45–3:00 Z. Zhang



and Q.-Q. Sun

University of Wyoming

A direct measure of E/I balance across different layers of somatosensory cortex during postnatal development

3:00–3:15 Mehdi Pedramfard, Manek S. Aulakh, Jong J. Park, Patrice D. Carr, Michael V. Johnston, Mary E. Blue and Mary Ann

Wilson



Johns Hopkins University School of Medicine

Effect of lead exposure on dendritic spine development in rodent barrel cortex

3:15–3:30 Rodrigo O. Kujis

Encephalogistics, Inc.

The barrel cortex as a novel assay for selective vulnerability and sparing in Alzheimer’s disease

(continued )

Barrels XXIII: Barrels by the shore 17

3:30–3:45 Discussion

3:45–4:00 Coffee Break

4:00–5:00 Short Platform Talks 3

Moderator: Qian-Quan Sun, University of Wyoming

4:00–4:15 Lu Li



and Daniel E. Feldman

Helen Wills Neuroscience Institute

A novel and rapid component of whisker response plasticity in rat somatosensory (S1) barrel cortex

4:15–4:30 Craig E. Brown



, Dani Sweetnam, Maddie Beange, Thomas Watson and Raad Nashmi

University of Victoria

A role for 4



nicotinic receptors in experience-driven cortical depression in adult mouse somatosensory cortex in vivo

4:30–4:45 Peter W. Hickmott

1

and Hubert Dinse

University of California, Riverside,

Ruhr University Bochum, Germany

Characteristics of thalamocortical EPSCs and IPSCs in hindpaw S1 in adult and aged rats in vitro

4:45–5:00 Discussion

5:00–5:30 Break

5:30 Poster Session

7:00 Dinner

Friday 12 November

9:00–11:00 Beyond the Barrel

9:00–9:10 Introduction/overview David Kleinfeld, UCSD

9:10–9:40 Carl Peterson, EPFL, Switzerland

Cortical sensorimotor integration

9:40–10:10 Ron Frostig, University of California Irvine

Activity spread beyond the barrel as a general model for mammalian sensory–motor cortex activation

10:10–10:40 Tim Murphy, University of British Columbia, Canada

Imaging widespread consensus patterns within spontaneous and sensory-evoked cortical depolarization in vivo,

implications for normal function and recovery after injury

10:40–11:00 Discussion

11:00–11:15 Coffee Break

11:15–12:00 Emerging Techniques

11:15–11:30 Hongkui Zeng, Allen Institute for Brain Sciences, USA

Mouse genetic toolkit for probing neural circuits and mapping connectivity

11:30–11:45 Discussion

11:45–12:30 Data Blitz

Moderator: Randy Bruno, Columbia University

12:30–2:00 Lunch

2:00–3:00 Short Platform Talks 4

Moderator: Jason Ritt, Boston University

2:00–2:15 Gaute T. Einevoll

1

, Patrick Blomquist

, Anna Devor

2,3

, Ulf G. Indahl

, Istvan Ulbert

and Anders M. Dale

Norwegian University of Life Sciences, Norway,

Massachusetts General Hospital,

University of California, San Diego,

Hungarian Academy of Sciences, Budapest, Hungary

Estimation of thalamocortical and intracortical network models for the rat barrel system from multielectrode recordings

2:15–2:30 Robert Sachdev

, Douglas Davis

1,2

and Vincent Pleribone

1,2

Yale University School of Medicine,

The John B. Pierce Laboratory

Traveling waves evoked in mouse barrel cortex by single whisker stimuli

2:30–2:45 L. Yassin



, B. L. Benedetti, J.-S. Jouhanneau, J. Wen, J. F. A. Poulet and A. L. Barth

Carnegie Mellon University

Unbiased methods to assess the firing activity of individual neurons in the neocortex have revealed that a large proportion

of cells fire at extremely low rates

2:45–3:00 Discussion

3:00–3:15 Coffee Break

3:15–5:15 Encoding and Decoding Sensory Inputs

3:15–3:25 Introduction/overview David Golomb, Ben Gurion University

3:25–3:55 Garrett Stanley Georgia Institute of Technology

Decoding the thalamocortical circuit

3:55–4:25 Brent Dorion University of Pittsburgh

Long-term plasticity and sensory response modulation

4:25–4:55 Ken Harris Imperial College London

How do neurons work together? The view from auditory cortex

4:55–5:15 Discussion

5:15 Adjourn

* indicates who is giving the presentation.

18 W. Zhang & J. C. Brumberg

be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.