Press Release

October 2010, Volume 36

Contents of this issue

15Contact & Imprint

Conferences 14News in brief

Workshops 15News in brief

BrainAmp MR/MR plus compatibility with Philips 3T MRI officially confirmed 2

IN THE FOCUS

BrainVision Recorder 1.20 to be released shortly 7

Product Update

eXtreme EEG on FMX/Motocross 13eXtreme EEG

New Distributor in Brazil 11Brain Products Distributors

Who is who - Dr. Barış Yeşilyurt 11EASYCAP Inside

Tin Cap by EASYCAP - a first introduction 12

EASYCAP Product Development

Workshop on EEG/TMS and EEG/fMRI at CINN Summer School 15

Brain Products User Workshops

Do you know how to install Software Sub-Licenses 8

Support Tip

Have you located the Solutions help documentation for Analyzer 2? 8

Support Tip

Brain-Computer Interfaces (Part 1) 9Did you know ...!?

Brain Oscillatory Substrates of Visual Short-Term Memory Capacity 4

User Research

Downloads, Programs and Updates 3News in brief

page 1 of 15

Dear Customers, Dear Friends of Brain Products,

this edition of our quarterly press release kicks off a series

of articles on a very hot topic in neurophysiology research: the

Brain-Computer Interface, or BCI. The author of the articles will be

Thorsten Zander, a customer of Brain Products, and – above all – a

prominent figure in the field of BCI.

Thorsten will guide us through the fascinating and amazing world

of BCI. He will describe its basics, discuss the different ways of

approaching it, and explain what a BCI research lab needs in order

to get started.

A new development is that many groups with no previous background in BCI research

are starting to work in the field. They are finding themselves overwhelmed by all sorts

of questions; we hope that Thorsten’s articles will help to answer at least some of them.

Working with Thorsten is a real pleasure, and his great commitment and willing helpfulness

deserve a big special thank-you from us!

I would also like to say a few words about some new developments in the EEG/fMRI field,

which result from a very friendly relationship between Brain Products and Philips MR team.

Several times during the past months, Philips MR neuro team and Brain Products MR team

had the chance to discuss some ideas to provide our joint customers with as many user-

friendly solutions as possible for the safe and correct use of our MRI-compatible systems

with the 3T Philips MRI scanner models, the Achieva and the Intera.

Philips‘ team showed always a sensitive awareness concerning the world of EEG/fMRI

research. Beyond accepting to work with us on a statement of compatibility – a very

important document which proves that our EEG system does not interfere at all with the

normal operation of MRI scanners during simultaneous recordings – Philips did something

more.

In fact, Philips is the first MRI manufacturer to design a 32-channel MRI coil that takes the

needs of EEG recordings specifically into account. In addition, Philips has implemented

a software application for Philips Achieva systems which automatically calculates an

optimized volume/slice timing parameter for combined EEG/fMRI.

In this connection too I would like to thank the Philips Team for their availability and

outstanding competence.

We hope you will enjoy reading our newsletter, and we look forward to meeting you again

in the pages of our next Press Release!

Pierluigi Castellone

Brain Products‘ CEO

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

page 2 of 15

IN THE FOCUS

BrainAmp MR/MR plus compatibility with Philips 3T MRI officially confirmed by Dr. Robert Störmer

Given the galvanically isolated, non-magnetic and completely

RF-proof design of the BrainAmp MR, this amp will probably

work directly in every imager bore if boundary conditions

(meaningful Tx coil configuration, GRE-EPI) are met. This has

been empirically proved and is no longer in question. If the EEG

cap is properly mounted and the imager gradient system timing

is reliable, good EEG during a BOLD sequence will always be

achieved.

A more crucial question is how the overall EEG system affects

imaging quality. Important research on this issue has been

conducted (Krakow et al 2000, Mullinger et al 2008,

Carmichael 2009, Mullinger et al 2010). However, the ultimate

test of compatibility remains the confirmations issued

by individual scanner manufacturers. Only if a scanner

manufacturer confirms compatibility with a given third-party

product can there be certainty of obtaining high-quality

imaging data during the combined EEG/fMRI session.

Siemens confirmed the BrainAmp MR’s compatibility with their

3T Magnetom series early in 2008 (see Brain Products’ press

release, vol. 28, 2/2008). Even though several dozen units

of the BrainAmp MR/MR plus are already in use on Philips

platforms ranging from 1.5 to 7T, we were curious to learn more

by observing the Philips compatibility test at f irst hand.

As the Philips MR neuro team and the Brain Products MR team

are always meeting at the same conferences, The Philips team

and Pierluigi Castellone (General Manager, Brain Products) were

soon able to arrange an on-site test at the Philips Healthcare

facilities in Best, the Netherlands.

Eventually, in

November 2009 a

complete 64-channel

BrainAmp MR plus

system, our certif ied

distributor in the

Netherlands, Dr. Harm-

Jan Wieringa (MedCat

Nl), and I arrived at

the centre of testing

operations, Philips’s

MRI R&D facility in

Best. It was certainly an

impressive spectacle

for us EEG guys to

see dozens of Philips

Achievas side by side.

The compatibility test is made up of a safety check followed by

checks for spurious signals, LF B f ields, B0 homogeneity, RF

coil inf luence, the influence of eddy currents, spurious signal

generation by protons, susceptibility to RF f ields/switching

gradient f ields, ECG lead conductivity, susceptibility to static

magnetic f ields, connectivity between the BrainAmp MR plus

system and the Philips MR system, and patient isolation.

Our contribution was to analyze the EEG data acquired during

the compatibility tests and to evaluate the impact of the various

sequences on it. The compatibility test reports went through

an appropriate review process, and and ultimately resulted

in a compatibility statement covering the Philips Intera 3.0T,

Achieva 3.0T, Achieva XR@3.0T, and Achieva 3.0T TX. The

certif icate is available for review at www.brainproducts.com/

product_approvals.php?aid=4.

Even more important for our customers are the practical

improvements which emerge from teaming up EEG/fMRI at

the working level.

The brand-new

32-channel Philips

SENSE head coil is

the very f irst coil in

which the needs of

combined EEG/fMRI

were considered.

Beyond its

outstanding imaging

capabilities, right

from the start this

32-element SENSE

coil was designed

with a cable channel in the Z-direction for the EEG cables of a

128-channel BrainAmp MR system. See http://www.healthcare.

p h i l i p s .c o m/i n/p r o d u c t s/m r i/o p t i o n s _ u p g r a d e s/c o i l s/

achieva3t/coils_neuro.wpd

This is an excellent solution which prevents loops involving the

BrainCap MR cable harness in the scanner.

Mandelkov et al (2007) have worked out that perfect scanner/

EEG synchronization requires consideration to be given to

sequence timing. To attack the root of this problem Philips

designed a console application (Fig. 1 on page 3) which

automatically calculates an optimized volume/slice timing

parameter for combined EEG/fMRI using the Achieva with

BrainAmp MR/MR plus systems.

The Achieva 3.0T X-series MRI combines simple operation with fast scanning and superb image quality

Philips SENSE Head coil (Rx) with 32 elements. The combination with

Brain Products EEG is supported by a dedicated pathway for the EEG leads in central Z-axis

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

page 3 of 15

We are very pleased to see Philips’s awareness of the needs of the fast-growing Achieva/

BrainAmp MR user community, and we look forward to a fruitful collaboration in the

future.

Fig. 1: The new Philips console application provides optimized sequence timing

parameters for combined EEG/fMRI and will be available soon.

References

Krakow K, Allen PJ, Symms MR, Lemieux L, Josephs O, Fish DR. EEG recording during fMRI experiments: image quality. Hum Brain Mapp, 2000 May: 10(1):10-5.

Mullinger K, Debener S, Coxon R and Bowtell R. Effects of simultaneous EEG recording on MRI data quality at 1.5, 3 and 7 tesla. Int J Psychophysiol, 2008. 67(3): pp. 178-88

Charmichael D. Image quality issues. In: EEG - fMRI: Physiological Basis, Technique, and Applications by Christoph Mulert and Louis Lemieux, Springer, 2009: pp. 173 -99

Mullinger K., Bowtel R. Influence of EEG recording equipment on MRI data quality. In: Simultaneous EEG and FMRI: Recording, Analysis, and Application. Edited by Markus Ullsperger, Stefan Debener, Oxford University Press, 2010: pp. 107 –17

Mandelkow H, Halder P, Boesiger P and Brandeis D, Synchronization facilitates removal of MRI artefacts from concurrent EEG recordings and increases usable bandwidth. Neuroimage, 2006. 32(3): pp. 1120-6.

News in brief: Downloads, Programs and Updates

June 29th, 2010 / New Update of BrainVision Analyzer Solution „EKG Markers“ available

Please visit our Analyzer 2 Download Section (www.brainproducts.com/downloads.php?kid=9&tab=3) to download the update. For Analyzer 1.05. the latest version of this Solution is available at www.brainproducts.com/downloads.php?kid=1&tab=3

July 6th, 2010 / New BrainVision Analyzer 2.0.1 Update available (ICA)

Please visit our Analyzer 2 Download Section (www.brainproducts.com/downloads.php?kid=9&tab=2) to download the update.

August 17th, 2010 / New actiCAP Manual available

Please visit our Manuals‘ Download Section (www.brainproducts.com/downloads.php?kid=5&tab=3) to download the latest manual for the actiCAP system and actiCAP ControlSoftware 1.2.1.0.

September 2nd, 2010 / New Update of BrainVision Video Recorder available

A new update of BrainVision Video Recorder (version 1.00.0005) is available in the Recorder Download Section (www.brainproducts.com/downloads.php?kid=2&tab=3).

September 24th, 2010 / New RDA Client for C# available

The new RDA Client for C# can be downloaded in the Recorder Download Section (www.brainproducts.com/downloads.php?kid=2&tab=5).

September 29th, 2010 / New BrainVision Analyzer 2.0.1 Update available (CB Correction)

Please visit our Analyzer 2 Download Section (www.brainproducts.com/downloads.php?kid=9&tab=2) to download the update.

All Updates and New Modules can be downloaded on our website at www.brainproducts.com/downloads.php. If you‘d like us to keep you posted on any new Update for BrainVision Analyzer 2, please register for our Analyzer 2 Newsflash at www.brainproducts.com/a2_newsflash.php

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

User Research

Brain Oscillatory Substrates of Visual Short-Term Memory Capacity by Paul Sauseng, Wolfgang Klimesch, Kirstin F. Heise, Walter R. Gruber, Elisa Holz, Ahmed A. Karim, Mark Glennon,

Christian Gerloff, Niels Birbaumer and Friedhelm C. Hummel

Current Biology 19, 1846-1852.

We are constantly bombarded with millions of sensory stimuli

entering the visual system. Only a very small fraction of this

input, however, can be held in memory for a short period of time.

Nowadays it is well established that the number of visual items

that can be stored in human short-term memory lies in a range

of about four, i.e. our short-term memory’s capacity is limited to

transiently maintain only up to four visual stimuli at the same

time. But why is human visual short-term memory capacity so

staggering small?

A neurocomputational model by Lisman and Idiart [1] suggests

that storage of multi-item memories is dependent on the nesting

of high-frequency gamma brain oscillations – with single cycles

representing single items – into slow theta waves which bind

these elements to multi-item memory representations. The

number of visual items which can be transiently held in memory

would therefore be limited by the number of fast oscillatory

cycles that can be nested into a theta wave. However, recent

EEG research suggests that short-term memory capacity also

depends on the ability to select relevant from distracting visual

information [2] in order not to overload short-term memory with

irrelevant input. In any case, it is highly plausible that human

short-term memory capacity does not rely on only one cognitive

function but that different capacity-related functions are involved

in parallel, i.e. the nesting of brain oscillations as substrate of

target retention might be found at the same time as for instance

information selection mechanisms, and both functions could be

responsible for memory capacity. This was recently addressed

by a study in which EEG and transcranial magnetic stimulation

(TMS) experiments were used to find and dissociate different

brain correlates of visual short-term memory capacity in healthy

humans [3]. In the following this study published in Current

Biology (19, 1846-1852) will be described in detail.

In EEG experiment 1 a delayed match-to-sample paradigm as

suggested by Vogel and Machizawa [2] was used. Visual scenes

comprising coloured squares left and right from a central fixation

cross were shortly presented. Subjects were told to retain only

the colour and spatial position of the items presented in the

visual hemifield which was previously cued by an arrow. After

a one-second retention interval colour and spatial position

of the cued (and therefore maintained) visual items had to

be compared to a probe stimulus. Since targets which were

required to retain and distracters which should be completely

ignored by the subjects were always presented in different

visual hemifields, this experimental setup has the advantage

that brain correlates associated with active maintenance of

visual information can be expected contralateral to the targets,

whereas one would expect to find

correlates of distracter suppression

rather at ipsilateral parietal recording

sites to target presentation. Based

on neurocomputational models as

mentioned above we hypothesized increased nesting of brain

oscillations as substrate of target maintenance at contralateral

posterior electrode sites during the delay interval. At ipsilateral

posterior sites, on the other hand, we were looking for an EEG

marker of cortical suppression, indicating attenuated processing

of irrelevant information. From previous research we know that

oscillatory brain activity at alpha frequency (around 10 Hz) is

associated with cortical deactivation and top-down inhibition.

Therefore, ipsilateral (to targets) alpha amplitude should be a

correlate of memory capacity related filtering processes.

Using a Brain Products BrainAmp system EEG data were recorded

from 28 scalp channels, and EOG was acquired for horizontal

and vertical eye movements. The latter was of great importance

since subjects were required to fixate the central cross on the

screen throughout the whole experiment. Trials containing

any horizontal or vertical eye movements were discarded from

further analysis. To attenuate effects of volume conduction and

to make data quasi reference-free raw data were transformed

into Laplacian current source density* as implemented in

BrainVision Analyzer. We then segmented data focusing on the

one second retention interval. Since the analyzed time interval

was immediately preceded by visual presentation of the memory

array, ongoing EEG activity might have been overlaid by evoked

potentials. This could be problematic when analyzing cross-

frequency interaction (nesting of oscillations), as a filter artifact

due to an evoked response affecting multiple frequency bands

could lead to spurious cross-frequency coupling. Therefore, the

BrainVision Solution for subtracting averaged responses from

single trial data was applied and only induced ongoing EEG

activity was further analyzed.

To investigate nesting of theta and gamma frequencies, as

a first step theta phase to gamma amplitude coupling was

analyzed. Therefore, for each single trial data were submitted to

Gabor expansion to obtain instantaneous phase and amplitude

values. Amplitude values for frequencies between 20 and 70

Hz and theta (4 to 8 Hz) phase values from single trials were

concatenated. Then gamma instantaneous amplitude values

were sorted according to their respective theta phase value (see

[3]). Last, gamma amplitude values were averaged over defined

theta phase segments. Thus, a nesting of gamma burst into theta

waves should be indicated by increased gamma amplitude for

Paul Sauseng

page 4 of 15

* number of splines: 4; max. degr. of Legendre polynomials: 10; lamda: 1 e-5

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

certain segments of a theta cycle.

Indeed, we found that there was increased gamma amplitude

following the negative peak of the theta cycle. This effect was

significant for a gamma frequency range between 50 and 70

Hz and could be found bilaterally at posterior recording sites

independent of how many targets and distractors had to be

retained/suppressed (Fig. 1a).

This implies that locking of gamma amplitude bursts to

theta phase seems to be a more general, short-term memory

unspecific mechanism since there was no difference between

posterior recording sites retaining targets (contralateral) and

sites processing distracters (ipsilateral). Thus we analyzed a

parameter of which we thought might be more specific: theta

phase to gamma phase synchronization, i.e. the very precise

locking of two frequencies’ instantaneous phases.

To do so phase values for frequencies between 50 and 70 Hz were

sorted in respect to instantaneous theta phase. Then for each

theta phase segment a synchronization estimate was calculated

which indicates the consistency of gamma phase values that

can be observed at a specific theta phase value. If there is little

circular variance of gamma phase angles that are obtained at a

certain instantaneous theta phase this synchronization index

is high and would imply a very precise nesting of gamma into

theta frequency. In contrast to theta phase to gamma amplitude

locking, phase to phase synchronization was stronger at posterior

sites processing targets than sites processing distracters.

Moreover, this contralaterally enhanced theta-gamma phase

synchronization exhibited a memory load dependent increase,

i.e. the more visual items that had to be retained the higher

contralateral theta-gamma phase synchronization. This pattern,

however, was only observable up to a memory load of 4 items.

When memory load was further increased, contralateral theta-

gamma synchronization was attenuated (Fig. 1b). This suggests

that when the number of gamma cycles, which have to be nested

into a theta wave, becomes too large, no stable cross-frequency

phase synchronization can be achieved anymore leading to a

capacity limit of short-term memory.

As a consequence, individual subjects with a very low short-term

memory capacity should exhibit such breakdown of stable cross-

frequency synchronization already when subjects with high

memory capacity still show an increase of contralateral theta

to gamma phase coupling. Indeed, we could demonstrate that

contralateral theta-gamma phase synchronization enhancement

from memory load 2 to 4 items was a significant predictor of

individual short-term memory capacity. Thus, one can conclude

that during the delay interval cross-frequency phase locking at

brain areas processing targets are a neuronal correlate of active

visual short-term memory retention.

Filtering information, i.e. preventing irrelevant information

to be retained in short-term memory should be another

mechanism influencing memory capacity. Hence, we analyzed

an EEG parameter which is thought to reflect an index of cortical

deactivation: EEG alpha band activity. Single trial EEG traces were

submitted to Fast Fourier Transformation. Amplitude estimates

between 8 and 12 Hz were averaged over left and right posterior

recording sites separately. In contrast to cross-frequency phase

synchronization alpha amplitude showed an ipsilateral memory

load dependent increase (Fig. 2a). This means that alpha amplitude

at brain sites involved in distracter processing was enhanced

depending on the amount of irrelevant visual information. This

load dependent increase, however, saturated when average

short-term memory capacity limit was reached. The amount of

memory load dependent alpha amplitude increase was a highly

significant predictor for individual memory capacity. It is of note,

however, that in contrast to contralateral cross-frequency phase

synchronization ipsilateral alpha amplitude was a correlate

of short-term memory capacity based on the suppression of

irrelevant information retention.

In two TMS experiments we addressed the question whether

the relation between ipsilateral alpha amplitude and memory

page 5 of 15

Fig. 1: Contralateral theta-gamma phase synchronization and ipsilateral alpha amplitude as substrates of short-term memory capacity. (a) Theta phase to gamma amplitude coupling indicates that posterior gamma amplitude is increased around the negative peak of a theta wave. This effect, however, is not short-term memory capacity specific. Precise synchronization between theta and gamma phase (independent of amplitude) exhibits a memory load related increase over posterior recording sites processing relevant visual information (b) and predicts short-term memory capacity. Figure modified from [3].

Fig. 1a Fig. 1b

Thet

a-ga

mm

a p

hase

sync

hro

niza

tion

dif

fere

nce

cont

rala

t. <

ipsi

lat.

cont

rala

t. >

ipsi

lat.

-0.005

0.005

0.01

0

load 2 load 3

Memory Load

load 4 load 6

Freq

uenc

y [H

z]

Theta phase-locked gamma amplitude

Theta Phase pi/-pi

Am

plit

ude

[a.u

.]

Amplitude [a.u.]

70 0.06

- 0.06

- 0.03 0.030.0

60

50

40

-

30

0+

0.0

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

capacity was only correlative or of causal nature. We used

rhythmical stimulation during the retention interval of the

above described task with the aim of entraining alpha activity

at the stimulation site. In a first experiment parietal rTMS at 10

Hz was either applied ipsilateral or contralateral to the visual

hemifield containing targets for retention in memory. Either real

or placebo stimulation with tilted coil was delivered. As another

control condition real TMS at 10 Hz was applied at the vertex.

We could show that ipsilateral stimulation at alpha frequency,

mimicking increased alpha amplitude, led to enhanced short-term

memory capacity compared to placebo and control stimulation

(Fig. 2b). Less surprising, TMS contralateral to relevant

information was associated with detrimental effects of short-

term memory retention. In a second TMS experiment we could

demonstrate that the described effect was only observable after

stimulation at alpha frequency but not when rTMS was applied

at 15 Hz. This frequency-specific effect is indirect evidence that

ipsilateral alpha amplitude had been entrained by 10 Hz TMS and

shows that gearing into ongoing alpha activity has causal impact

on short-term memory capacity.

The last question to be answered in this study [3] was whether

the two correlates of short-term memory capacity, contralateral

theta-gamma phase synchronization and ipsilateral alpha

amplitude were independent from each other or reflected similar

substrates of short-term memory retention. Hence, we carried

out another EEG experiment in which the number of targets and

distracters was systematically varied. It could be shown that

contralateral theta-gamma phase synchronization was only

influenced by the number of targets which had to be encoded,

completely independent of distracters. On the other hand,

ipsilateral alpha amplitude responded exclusively to the amount

of irrelevant information, independent of the number of targets

presented.

So one can conclude that in our study [3] we were able to show

that human visual short-term memory capacity relies on two

independent cognitive functions which are implemented by

brain oscillatory mechanisms: (i) active retention of relevant

information reflected by theta-gamma phase synchronization at

posterior brain sites processing targets, and (ii) deactivation of

distracter items to prevent maintenance of irrelevant information

which was indicated by increased EEG alpha amplitude at brain

areas processing distracting information. It would be naïve

to think that there were not a number of additional cognitive

functions which might have impact on visual short-term memory

capacity, such as visual attention mechanisms or central

executive monitoring processes. It should be the goal of future

research to investigate brain correlates of such functions with

a multi-methods approach in order to disentangle them during

visual short-term memory processing and determine their

individual contribution to capacity limits of human memory.

page 6 of 15

References

[1] Lisman, J.E., and Idiart, M.A.P. (1995). Storage of 7 +/- 2 short-term memories in oscillatory subcycles. Science 267, 1512–1515.

[2] Vogel, E.K., and Machizawa, M.G. (2004). Neural activity predicts individual differences in visual working memory capacity. Nature 428, 748–751.

[3] Sauseng, P., Klimesch, W., Heise, K.F., Gruber, W.R., Holz, E., Karim, A.A., Glennon, M., Gerloff, C., Birbaumer, N., Hummel, F.C. (2009). Brain oscillatory substrates of visual short-term memory capacity. Curr Biol. 19, 1846-1852.

Fig. 2: Amplitude of alpha frequency exhibits a memory load dependent increase at posterior brain sites processing distracting information and reflects the deactivation of irrelevant memory representations (2a). 2b: Rhythmical transcranial magnetic stimulation at alpha frequency applied over distracter processing brain areas mimics increased alpha activity and leads to enhanced short-term memory capacity. Figure modified from [3].

load 2 load 3 load 4 load 6

Memory Load

Late

raliz

ed a

lpha

am

plit

ude

[µV

/m²]

ipsi

lat.

> c

ontr

alat

.

1

0

3

4

2

Mem

ory

Capa

city

[K]

**

* * p < 0.053.5

3.0

2.5

2.0

1.5Contralateral VertexIpsilateral

real rTMS sham rTMS

Fig. 2a Fig. 2b

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

Product Update

BrainVision Recorder 1.20 to be released shortly by Dr. Davide Riccobon

We just have successfully concluded user

testing of the beta version of BrainVision

Recorder 1.20. I’d like to take this opportunity

to thank all those customers who have

supported us with their valuable feedback

and crucial suggestions both currently

and in previous years. I look forward to

this much-appreciated cooperation both

continuing and increasing further.

Very few problems were reported regarding

the most recent version, but every report we

received has led to valuable improvements.

The time for BrainVision Recorder 1.20 to be

released is now very close – namely, by the

end of October.

We ask our users to be patient just a little

longer; the wait will be worth it. BrainVision

Recorder 1.20, the most stable and user-

friendly version ever, contains innovations

that will satisfy researchers from different

scientific disciplines having a variety of

needs and expectations.

Fans of speed will really appreciate our

BrainVision Recorder 1.20 running under

the 64-bit version of Windows 7. This new

combination opens up new frontiers to

all those researchers who have reached

the performance limits of their 32-bit

processors.

Customers who have already come to

appreciate the advantages of synchronous

recordings of video data together with EEG

and/or peripheral activity, either as a way

of monitoring their subjects or because it is

essential to their experimental design, will

be pleased with the improvements made

to the Video Recorder (Fig. 1). Even under

bad operational conditions, such as when

the camera cable becomes disconnected

and reconnected when saving data, the

recording is carefully safeguarded and the

loss of information is minimized from a

software perspective.

Moreover, it is possible to select different

video resolutions if the camera supports

this, and thus to fine-tune the recording in

accordance with individual requirements.

actiCAP users can control the impedance

check of active electrodes directly via

BrainVision Recorder (Fig. 2), and the

impedance values are automatically saved

in the Recorder header file.

With the new “Select Virtual Amplifiers”

function (Fig. 3), BrainAmp users can create

new workspaces even if no amplifiers are

actually connected. In this way, complex

workspaces (with up to 256 channels,

individual channel properties, different

filters etc.) can be designed and developed

offline out of the lab, for instance at the

office or even at home.

The handling of workspaces, the display of

time intervals and the selection of single or

multiple channels during EEG monitoring

are all easier and more intuitive.

The Online Average functionality has been

further enhanced with a view to its potential

use for educational purposes, for the

detailed online monitoring of data, or for

some other application. It is now possible

to add to the average view an averaged

wave that would previously either have

been recorded with BrainVision Recorder or

generated using BrainVision Analyzer, and

thus to compare the real-time wave with a

prototypical one (Fig. 4).

These are just a few of Recorder’s new

features. You will find an exhaustive list on

the BrainVision Recorder download site as

well as in the manual.

We are anxious to know your opinions.

Please do not hesitate to contact us with

your invaluable questions, suggestions and

feedback.

page 7 of 15

Fig. 1: BrainVision Video Recorder

Fig. 2: Impedance Check

Fig. 4: Static Overlay

Fig. 3: Virtual Amplifier

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

page 8 of 15

Support Tip

Do you know how to install Software Sub-Licenses? by Dr. Davide Riccobon

Some optional components of our programs will only run if you have previously purchased sub-licenses for them. A sub-license is a

file associated with your USB dongle.

If you purchased your sub-licenses and the program at the same time, the sub-license file will be included on the supplied USB data

carrier. Sub-licenses that are purchased subsequently can be downloaded from the Brain Products website. If you need to download

a sub-license file from the Web, you will f irst have to register your dongle here: www.brainproducts.com/productreg.php.

This will generate your login data; this will also be valid for any other program that you download from our home page. From now on

you must follow procedures that will differ according to which program you are using.

BrainVision Analyzer 2

Download: After receiving your access data, you can log in and download

your individual license file for Analyzer 2 at

www.brainproducts.com/downloads.php

(the license file can be found at the bottom of the page)

Installation: Independently from your operating system (Windows XP, Win-

dows Vista, Windows 7), install the license file by double-clicking

on the downloaded installation program (BrainProducts-License-

xxxxxxxx-xxxx.exe) and follow the wizard’s instructions.

BrainVision Recorder

Download: After receiving your access data, you can log in and download

the Recorder license file at

www.brainproducts.com/downloads.php

(the license file can be found at the bottom of the page).

Installation: The installation procedure to be followed will depend on your

operating system. When working with

Windows XP, install the license file by

double-clicking on the downloaded

installation program (BrainProducts-

License-xxxxxxxx-xxxx.exe) and follow

the wizard’s instructions.

When working with Windows Vista or

Windows 7, click on the license file icon

with the right mouse button and select

the [Run as administrator] option in

the drop-down menu. Then follow the

wizard’s instructions.

The Solutions are small programs which add useful functionali-

ties to BrainVision Analyzer. For example, one of them enables

you to import files saved in the BESA file format, while another

makes it possible to export the area under the curve around the

peak markers.

Since the Solutions are not part of the

main BrainVision Analyzer 2 program,

they are not described in the User

Manual. Each Solutions file contains

its own associated help documenta-

tion. In Analyzer 2, the Solutions /

Help / Solutions Help menu opens

the Solutions Help Explorer program,

where a short description appears

when you left-click on an item.

The short descriptions generally

consist of just a few sentences,

while double-clicking opens a

separate window that displays more

detailed help documentation.

Here is where you can download the

latest version of the Solutions for

BrainVision Analyzer 2:

www.brainproducts .com/down-

loads.php?kid=9&tab=3

Support Tip

Have you located the Solutions help documentation for Analyzer 2? by Dr. Roland Csuhaj

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

The following article represents the first of a series of reports

on Brain-Computer Interfaces. It presents an overview of the

field, its technologies and methods, and also its main goals.

A subsequent report will present the main achievements of

BCI research during the past two decades, followed by a report

on novel BCI-based approaches. The final report will contain

an overview of freely available tools and methods needed for

building up your own BCI laboratory.

>> BCI in a nutshell

A Brain-Computer Interface (BCI) is a direct link between a

human brain and a technical system. It detects patterns in

brain activity and translates them into input commands given

to the machine. Usually, brain activity is recorded using an EEG

system and is interpreted by a conventional personal computer

using machine learning and signal processing techniques. The

initial, principal goal of BCI-based applications has been to

provide communication and control channels for users who have

lost their ability to communicate naturally. These are mainly

patients suffering from amyotrophic lateral sclerosis (ALS) or

tetraplegia.

From the perspective of Human-Machine Systems (the science

of interaction between humans and machines), a BCI defines

a new input modality for the active or passive transmission of

commands. Active commands are intentional and focused, like

a mouse click, while passive commands are not the result of

deliberate intentions. A passive command can, for example, be

derived from the brain’s reaction to the perception of an error or

other aspects of cognitive user state.

Figure 1 depicts how a BCI could be implemented in the context

of a human-machine system feedback loop.

>> Why EEG?

Most of BCI research is based on the

electroencephalogram (EEG). Why

is this the technology of choice for

most BCI researchers? The reason

is that EEG provides high temporal

resolution, it is comparably easy to use and it is not too

expensive. Apart from these basic characteristics, the main

advantage of using EEG for BCI is that it has been thoroughly

researched. In the last century numerous studies were conducted

that were based on analyses of EEG features corresponding

to cognitive processes. This enables vast possibilities for BCI

research. The drawbacks of EEG – its limited spatial resolution

and vulnerability to artifact sources – are factors that potentially

limit current BCI research. However, these may be resolved in the

near future using powerful methods derived from engineering

and mathematics, like Independent Component Analysis (ICA) or

novel sensor designs.

Other measures too can be utilized for BCI research. These can

be categorized as invasive versus non-invasive technologies.

Other relevant non-invasive measures are functional magnetic

resonance imaging (fMRI), magnetoencephalogram (MEG) and

functional near-infrared specotrography (fNIRS). fMRI and MEG

share the major drawbacks of being complex in application and

being unsuitable for long-term use. These technologies are also

comparatively expensive. fNIRS and fMRI share the drawback

of having a low temporal resolution, since both rely on the

blood-oxygen-level-dependent (BOLD) component. The spatial

resolution of fNIRS is also low; its greatest potential therefore

lies in being a secondary measure used in conjunction with EEG

to provide additional information about the BOLD component.

Potential invasive techniques are represented by the use of

electrocorticogram (ECoG) and microelectrode arrays. Apart from

being invasive, these technologies share one major drawback:

once placed they can only be switched to other spatial areas

with great difficulty and it is not possible to cover the whole

cortex with sensors. They can therefore only be used for certain

applications.

>> Goals of BCI research

Attaining the primary goal of providing a channel for

communication and control that relies on no other human bodily

activity besides that of the brain requires the attainment of

several sub-goals.

Optimizing signal acquisitionThe first step in Brain-Computer interfacing is to acquire

information on brain activity. As a BCI is intended for long-term

use a proper sensor is needed that is quick and easy to apply, is

Did you know ...!?

Brain-Computer Interfaces (1) by Thorsten Zander

Machine

Interpretation

BCI

Inputmodalities

Outputmodalities

feedback

adaptation

Output Input

User

Figure 1: Feedback cycle in human-machine systems augmented by a BCI. BCI-based input can be either active or passive. It can be the sole input modality or it can be combined with other input modalities, such as mouse or speech control.

page 9 of 15

Thorsten Zander

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

as unobtrusive as possible, provides reliable information on brain

activity, and operates without harming the user. EEG partially

fulfills these criteria, but there is some scope for optimization.

The use of gel for impedance reduction is time consuming, limits

the length of time that a BCI can continuously be used, and

means that the user’s hair needs to be washed afterwards. Also,

as the gel is usually abrasive, uncomfortable skin irritations can

develop in long term use. Sensor technology therefore needs to

be improved. First dry electrodes are presently being developed.

These could provide a quality of data that is similar to that of the

EEG systems currently in common use, but without the need for

any liquids.

Defining features in EEGAnother important BCI research issue involves the identification

of distinctive EEG features that can provide useful information

on the user’s current cognitive state. The term ‘Features in EEG’

refers to characteristic task- or state-related EEG patterns that can

be used to infer information on

the current state of the subject’s

brain. One major criterion for

the usefulness of such features

is that the user must be able

to generate the features easily.

But, as EEG is no biunique

mapping of brain activity (mostly

because of volume conduction

and the reduction of the signal

to two spatial dimensions) it is

vital to minimize the likelihood

of interference from features

that may be generated by

other cognitive processes and

projected in a similar manner.

Last but not least, the selected

features should be easy to detect

using automated methods. Typically, it is important to ensure a

good signal-to-noise ratio; this can be achieved with signals that

display low variance and strong coherence across trials, as well

as high amplitudes in the time or frequency dimensions. In this

respect, BCI research benefits strongly from previous research

done in neuroscience, providing a lot of information on features

possibly suitable for BCI applications. Detailed examples will be

included in the second report, which will focus on BCI research.

Detecting BCI-featuresAnother main task for BCI research is the development of

efficient algorithms for translating brain activity into input

commands. From an abstract perspective, these algorithms

select and transform those portions of EEG data that best reflect

a previously-selected brain activity pattern. This results in BCI

features that do not necessarily retain the structure of EEG data

and are consequently more abstract than the standard features.

This is usually done in a three-step process. EEG produces a lot

of data that conveys no information about the process being

investigated and must be filtered out. Accordingly, a first step

is to apply restrictive filters – that are implemented retaining

causality, in contrast to the signal processing typically used in

EEG analyses. This restricts the temporal, spatial and oscillatory

bandwidth of the EEG recording. Subsequently, features will be

extracted based on knowledge derived from the neurosciences.

In this step, particular components of the data will be selected,

combined and attenuated by further processing. The final step

is to calibrate a classification algorithm based on prototypical

features that are usually generated in a separate training

session. The resulting algorithm is used in the final application

to derive information on the cognitive user state from the EEG. It

is therefore highly important that BCI research has proper tools

at its disposal that enable the optimal selection of features.

>> Ingredients of BCI research

Research in the BCI field is usually highly interdisciplinary, and

has its foundation chiefly in

the neurosciences, psychology,

mathematics and computer

science. However, in recent

years, as more BCIs are being

incorporated in non-laboratory

applications, BCI research

has also been influenced by

the fields of human factors,

cognitive science, engineering

and human-computer interaction

(HCI). The neurosciences supply

valuable information regarding

the structure and functionality

of the human brain. Proper

analysis of the results and the

setting-up of experimental

designs in which the factor

of investigation could be modulated while other factors are

controlled, as well as the drawing of accurate inferences from a

given set of results, is contributed by the domains of psychology

and cognitive science. Mathematics and Computer Science

usually are consulted for building up a system for automating

predefined steps of inference on EEG data. The knowledge and

techniques derived from the fields of human factors, engineering

and HCI are invaluable for setting up a usable and effective

application that is based on BCI input. The systems constructed

must make efficient use of the limited bandwidth available in a

BCI system. This can be done by optimizing the interface design,

incorporating as much automation as possible and including

information on the environmental, user-related and technical

state of the system being used.

>> Possible areas of application

The initial, principal purpose of BCI systems is to provide a

mechanism for communication and control that can be used

Fig. 2: BCI research is influenced by many disciplines. Neurosciences, Psychology, Computer Science and Mathematics are a substantial part of basic BCI research, while Cognitive Science, Engineering and Human Factors became

more relevant for building up BCI based applications.

page 10 of 15

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

page 11 of 15

Brain Support Brasil is a small company dedicated to the sale

of equipment and services connected with neuroscience and

behaviour. Our work in the neuroscience market began at the

beginning of 2010. Prior to this, we were focused on selling

clinical neurology equipment.

Our experience in medical equipment has been consolidated

through sales, installations and training. We have already

worked with neurology equipment and software produced by

Nihon Kohden, Polysmith, NICOLET, TECA, Medelec, Oxford,

Stellate, Magstim, Digitimer and Compumedics/Neuroscan; we

also used to represent 4D Neuroimage.

We believe it is ideas that build companies. Our original concept

was born 30 years ago, when I paused one day to ask myself,

“How does the construction of thoughts actually happen?”

This led me to select a secondary school that offered a course

in clinical pathology. I then studied electronic and electrical

engineering, and subsequently specialized in theology,

parapsychology and bioengineering.

So here I am. Uncertainty still exists in my mind concerning how

thoughts are constructed, but I now know that the best answer

can be found in the work of our neuroscientists.

I started my professional life teaching. I then spent 5 years

working as a clinical technician in a 1,000-bed general hospital,

followed by 15 years selling neurology equipment.

Brain Support Brasil was a dream that has now become a reality.

Our current work can be summed up as making good ideas

available to Brazil’s neuroscientists through offering them

equipment, software and services.

When we embarked on representing Brain Products, we were

surprised by the high quality of service and dedication of all

its employees. It is the first time that I’ve felt the presence of

such commitment and involvement in a company, and I’m really

looking forward to our cooperation.

Brain Products Distributors

New Distributor in Brazil: Brain Support Brasil by Jackson Cionek

BRAIN SUPPORT BRASILSão Paulo Capital Brasil

by severely disabled people. However, as the reliability and

usability of BCI systems has improved during the past decade,

their applicability and appeal for other applications has also

grown. With the introduction of passive and hybrid BCI systems,

this technology could also be of interest to users in a normal

state of health. In particular, users in specialized working

environments, such as astronauts, surgeons and people

interacting in augmented environments, might benefit from

being able to use additional input mechanisms. This will be

discussed in more detail in the next Press Release issue.

I would like to introduce myself as the new member of EASYCAP

Support Team.

I studied Electrical Engineering at the Technical University

of Istanbul and subsequently at the University of Siegen,

where I specialized in communications engineering. During my

academic studies, I dealt particularly with multicast-multimedia-

communication as well as cryptographic applications and

software implementation of encrypted multimedia conferencing

systems. After receiving my diploma degree from the University

of Siegen, I started my PhD at the High-Field Magnetic Resonance

Center at the Max Planck Institute for Biological Cybernetics.

The main focus of my research studies was (a) detecting and

characterizing hemodynamic response to very brief sensory

stimuli (b) exploring the relationship between hemodynamic

response and the underlying neuronal events (c) investigating

neuronal interactions using non-invasive blood-oxygenation-

level-dependent (BOLD) contrast.

Within the framework of my research

studies, additional to functional

magnetic resonance imaging (fMRI)

technique at high and ultra-high

field MRI scanners, I worked with

EASYCAP’s and Brain Products’ - MRI compatible - EEG recording

caps, hardware and recording/analysis software.

I spent a great deal of my PhD time concerning myself with the

fusion of complementary and simultaneous EEG-fMRI recordings.

After earning my PhD at the International Max Planck Research

School at the University of Tübingen, I joined the EASYCAP

Support Team in June 2010.

I am looking forward to meeting you at support@easycap.de and

discussing with you your technical and scientific questions.

EASYCAP Inside

Who is who - Dr. Barış Yeşilyurt Technical and Scientific Support

Dr. Barış Yeşilyurt

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

EASYCAP is well known for its high-quality sintered Ag/AgCl

electrodes, which can cope with any electrophysiological

demands that are placed on them. However, the requirements

of many EEG applications can also be satisfied using electrodes

with tin sensors, which are much cheaper. Therefore EASYCAP

now offers caps and single electrodes with sensors made of pure

solid tin, which also have great mechanical strength.

• Tin electrodes have no stable electrode potential,

and recordings should therefore be made using an

AC amplifier (or an amplifier in AC mode) with a

high-pass filter equal to or higher than 0.3 Hz (time

constant: 0.53 sec).

• Especially when they are quite new, they also

have a tendency to produce random but infrequent

spikes that exceed the usual EEG voltage range.

An automated artefact removal routine should

therefore reject sequences with max./min. values

more than 150 µV apart.

• Tin electrodes need a minimum surface area of

120 square millimeters to allow for low impedances

and ensure stable signals. During the mounting and

filling of the cap/electrodes, care should be taken to

ensure that the electrolytic medium wets as much of

the sensor surface as possible.

If these conditions are satisfied, tin electrodes will deliver

a clear, drift-free and noise-free signal. Their limits are only

reached when components slower than 1-2 Hz, such as Readiness

Potential (RP), Contingent Negative Variation (CNV) or DC-EEG

are being studied, or if EEG is to be recorded simultaneously

alongside other technologies like fMRI, MEG or TMS. But without

such co-registration they are as suitable as any other sensor

material for all clinical EEG and evoked potentials, and for most

event-related potentials.

Although electrodes are rather small and lightweight objects,

they are subjected to a lot of mechanical and chemical stresses

in daily practice: they are carried around, stretched over people’s

heads, submerged in water and brushed. Sometimes, they are

not properly cleaned and dried, or are left exposed to direct

sunlight on a window-sill. Disinfecting them also shortens their

lifespan. We have therefore set out to produce mechanically

strong electrodes that are absolutely waterproof, and which are

equipped with cables covered in insulation that is as tough as

possible.

These tin electrodes are now available, either as single

electrodes or mounted together in complete EEG-recording caps.

Individual electrodes can of course be attached to bare skin

with an adhesive bandage, or secured in the hair with sticky

electrolyte. But the same electrodes can also be snapped

into head caps with electrode holders, thus permitting any

combination of individual external electrodes with cap-mounted

electrodes at any location and in any quantity. This method

allows for many degrees of freedom, which for instance is useful

for education environments or rapidly changing research tasks.

In other environments, whether these are clinical environments

or research environments involving many subjects over a

short time, it is useful to have a more time-saving set-up.

The complete EEG-recording caps are intended for such

applications: all the electrodes are positioned in the

right locations, and have a universal quick-connect interface

that is compatible with almost all brands of EEG amplifier. These

EEG-recording caps come in standard layouts with 21, 32, 64, or

128 channels.

We intend to offer both single electrodes and complete EEG caps

in standard versions at competitive price points. As a bonus,

we are naturally also able to utilize our EASYCAP experience to

produce individualized, customized caps that use tin electrodes,

but understandably we cannot offer this service free of charge

as we do with the sintered Ag/AgCl electrodes, but have to apply

a reasonable surcharge.

The tin electrodes will be presented at forthcoming exhibitions

that feature the participation of EASYCAP, Brain Products and

MES, such as SfN in San Diego or EEG-Tage in Munich.

We are now also able to respond to requests for samples. And

please keep an eye out for additional information here, in future

editions of the Brain Products Press Release.

EASYCAP

Tin Cap by EASYCAP: A first introduction by Falk Minow

page 12 of 15

EASYCAP‘s Tin Electrodes are available with either large or small center openings. Large openings can accomodate both, cotton swabs for painless and effective impedance

minimization, or syringes with or without blunted needle. Small openings can be used only with blunted needles.

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

page 13 of 15

eXtreme EEG

eXtreme EEG on FMX/Motocross by Saül Martínez-Horta

Ever since I started working with EEG and event-related brain

potentials, one of the most challenging problems we have had

to solve in every experimental session with our clinical and

movement-disorders-related population

(e.g. those with Parkinson’s disease) has

involved the typical movement-related

artifacts. But the first time I read about

the possibility of performing an EEG

acquisition under out-of-lab conditions I

needed only five minutes to think up more

than a thousand possibilities we could try.

In the course of an exchange of emails

with some people from Brain Products,

they encouraged me to develop one of my

proposals. This was an EEG acquisition

to be performed during motocross and

FMX, a variant of motocross also known

as freestyle, which consists of executing

maneuvers like backflips and similar while

doing jumps of more than 20 meters.

After that, the professional Spanish

snowboarder Carlos Manich (http://

carlosmanich.wordpress.com) told me

that I had an opportunity to perform EEG

acquisitions with Sydney de Andres of Freestyle

Factory (www.freesportsfactory.com) and

Eugenio Zafra (www.eugeniozafra.com). Both

of them are professional international FMX and

motocross riders who were really excited about

the possibility of participating in this project.

First encounter with EEG and motocross

We decided with representatives from Brain

Products that we would set up two experimental

sessions during the course of the Human Brain

Mapping congress in Barcelona in June 2010.

One session involved circuit-based motocross

with Sydney de Andres, and the other would take

place with Eugenio Zafra in an FMX park.

Then we agreed a testing session with Sydney just to try out all the

equipment and possible setup permutations. During the testing

session it became apparent that we would not be able to work

with all the proposed channels due to the limitations imposed by

the helmet. For this reason we decided to use the actiCAP with

just FP1, FP2, Fz, Cz, F3, F4, C3 and C4.

The other components of the equipment consisted of a video

camera synchronized with the EEG acquisition and attached to

the helmet, while the computer and Extreme EEG amplifier were

mounted in a backpack.

Afterwards, looking at the EEG recorded under these very difficult

conditions, it was clear that a new era in EEG acquisition had

been kicked off.

Sydney rode in a 15-minute motocross

session trying jumps of more than 15

meters in which all the equipment worked

perfectly. Obviously, we found a lot of

muscular and vibration-related artifacts,

but none in any of the EEG segments,

nor, most importantly, during the jumps

and landings. So after well-founded

initial doubts about this issue we are now

absolutely convinced that EEG acquisition

can be successfully achieved during FMX.

The experimental session: Neuroscience meets freestyle

During the Human Brain Mapping congress,

Barcelona endured wet, rainy weather that

excluded the possibility of conducting

the experiment with the motocross riders

in optimal conditions. Landing on a wet

surface after a jump of more than 20

meters could turn out to have absolutely

disastrous consequences, and this was the prime

reason for cancelling the experimental sessions.

After this intervention of Murphy’s Law, just

a few days later the weather was perfect.

With not much time in hand, together

with Jordi Cortes from Bionic Iberica

(www.bionic.es) we set a date with Sydney and

Eugenio to conduct both sessions in one day

– first the motorcross session, then the FMX

session.

Having set-up up the actiCAP system with the

design adapted to the helmet, Sydney rode

extremely fast during a motocross session

lasting over 20 minutes during which the system

functioned perfectly throughout. The pictures

show that even more than the jumps, the motocross circuit

consisted of really sharp bends, bumps and all the possible

surface conditions that one might suppose would be incompatible

with an EEG acquisition attempt.

Then it was time to FMX. When Jordi and I arrived in the park, our

first response was “Look at those ramps!”. A really tall structure

rose up at a distance of more than 20 meters from the landing

area. But by contrast with the motocross circuit, here the ride

consisted simply of a full-power straight line across a smooth

surface followed by the ramp, a void of more than 20 meters, and

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

page 14 of 15

the landing area.

We carefully mounted all the equipment,

placing the video camera in the handlebar

because Edgar preferred to avoid having

’unfamiliar’ weight on his head. The

pictures show the rest. Edgar started

with an enormous basic jump followed by

backflips, backflips not using his hands,

not using his feet, keeping his entire body

off the motorbike…

Again, the system worked perfectly in the course

of the 15-minute session until the cable of the

video camera became disconnected when it got

kicked during a jump.

What does this data suggest?

Tonight I have spent hours looking at the pictures

and video (it will be uploaded as soon as we

are able to do so), and obviously the data. I

discovered that some channels were creating too

much noise because of vibration and the pressure

of the helmet (FP1, FP2 and sometimes Cz), but

when this noise was subtracted, the rest of the

acquisition was amazingly good.

Subsequently, the first conclusion was that

a new era of EEG experiments has begun

in relation to studying brain activation

patterns under real-life conditions. If we

are able to acquire a good EEG signal

during a motocross flight of more than 20

meters, EEG acquisition during walking,

talking, painting, driving etc. will be child’s

play.

Secondly, BrainVision Analyzer from Brain

Products allows us to insert markers using

synchronized video information, and thus to

create distinct EEG segments that can be averaged

to obtain event-related brain potentials (ERPs).

We are now working on the segmentation of the

acquired EEG in an attempt to confirm whether

we can obtain movement, error detection and

correction-related ERPs from our data. It’s just a

matter of time, and we are absolutely convinced

that it will be possible to achieve it.

The resulting data (as well as a video and a report)

will be posted on the Brain Products website

at www.brainproducts.com/extreme_eeg.php

shortly.

Dortmund (Germany), Oct 14th to 16th, 2010 in cooperation with Easycap GmbHInternational Conference on Aging & Cognition

Munich (Germany), Oct 21st to 22nd, 2010 in cooperation with MAG & More GmbHMünchner Treffen der Arbeits-gemeinschaft TMS in der Psychiatrie

in cooperation with Brain Vision LLC.San Diego (USA), Nov 13th to 17th, 2010Annual Meeting of the Society for Neuroscience (SfN)

in cooperation with EMS srlPalermo (Italy), Nov 24th to 26th, 2010XVIII Congresso SIPF

in cooperation with Brain Vision LLCPortland (USA), Sep 29th to Oct 3rd, 201050th Annual Meeting of the SPR

in cooperation with Brain Vision LLCVancouver (Canada), Dec 6th to 9th, 2010NIPS 2010

For more information on the conferences we are about to attend, please visit our website at www.brainproducts.com/events.php

News in brief: Conferences

www.brainproducts.com

Brain Products Press Release October 2010, Volume 36

page 15 of 15

For more detailed information on upcoming workshops please visit our website at www.brainproducts.com/workshops.php To announce your interest in attending any workshop we offer, please register at www.brainproducts.com/wsrequest.php

News in brief: Workshops

Beijing (China), October 19th to 20th, 2010BrainVision Analyzer 2 User Workshop on EEG and EEG/fMRI

Portsmouth (UK), October 25th & 26th, 2010BrainVision Analyzer 2 User Workshop

San Diego (USA), November 11th & 12th, 2010BrainVision Analyzer 2 User Workshop on EEG and EEG/fMRI

This Press Release is published by Brain Products GmbH, Zeppelinstrasse 7, 82205 Gilching, Germany.

Phone +49 (0) 8105 733 84 0, www.brainproducts.com

Notice of Rights

All rights reserved in the event of the grant of a patent, utility model or design. For information on

getting permission for reprints and excerpts, contact marketing@brainproducts.com. Unauthorized

reproduction of these works is illegal, and may be subject to prosecution.

Notice of Liability

The information in this press release is distributed on as „As Is“ basis, without warranty. While every

precaution has been taken in the preparation of this press release, neither the authors nor Brain

Products GmbH, shall have any liability to any person or entity with respect to any loss or damage

caused or alleged to be caused directly or indirectly by the instructions contained in this book or by

the computer software and hardware products decribed here.

Copyright © 2010 by Brain Products GmbH

Brain Products User Workshops

Workshop on EEG/TMS and EEG/fMRI at CINN Summer School on Neurodynamics 2010 by Etienne B. Roesch

This summer saw the 2nd edition of the Summer School

on Neurodynamics organized by the Centre for Integrative

Neuroscience and Neurodynamics, at the University of Reading,

UK. This year, in addition to an exciting program gathering keynotes

by leading figures in the field, such as Prof. Walter Freeman (UC

Berkeley), the students enjoyed a two-day workshop on coupled

EEG-fMRI and coupled EEG-TMS, organized in collaboration with

Brain Products and MAG & More.

Kerstin Wendicke (MAG & More’s CEO) began the first day of the

workshop by introducing TMS, and demonstrated its effect on

herself and a couple of brave volunteers. Pierluigi Castellone

(Brain Products’ CEO) then explained the effect of TMS on the EEG

signal, and the possible ways to reduce TMS-induced artifacts. As

you can imagine, the coffee break that followed gave way to vivid

discussions of awed excitement!

We then recorded the EEG from a kind volunteer to demonstrate

some of the steps involved in acquiring an EEG signal with a

BrainAmp MR plus system during TMS stimulation performed

with the innovative PowerMAG research from the company MAG

& More.

Pierluigi then detailed the issues faced when recording EEG in the

MRI scanner, going back to MR physics to describe the impact of the

MR recording on the EEG signal. Furthermore, he described Brain

Products solutions for the combined measurement, emphasizing

what has been done to assure patient safety and data quality.

In the afternoon, we then moved to the MRI scanner, where yet

another willing volunteer was hooked up on the EEG system

in the scanner bore of our Siemens TRIO. We simultaneously

recorded the EEG and BOLD signals

elicited by the passive viewing of a

simple flickering checkerboard, whilst

the data recorded with BrainVision

Recorder was cleaned in realtime from

the gradient and pulse artifacts by

using the BrainVision RecView.

On the second day, Pierluigi explained in detail the different

steps required in the analysis of the EEG signal. Gathered in

the computer lab, the participants then followed a tutorial

on BrainVision Analyzer 2. They learned how to clean the data

from the MR artifact, and analysed the visual evoked potentials

recorded the previous day.

With an audience composed of graduate students and

postdoctoral fellows from psychology, linguistics, neuroscience,

computational neuroscience, mathematics, computer sciences

and engineering, it is needless to mention that the level of prior

exposure to neuroimaging techniques varied greatly. Kerstin and

Pierluigi nonetheless managed to convey the technicalities of

cutting-edge neuroscience with passion and enthusiasm, in a

genuine tour-de-force!

Thank you so much! ... And see you next year! :)

More information on the Centre for Integrative Neuroscience and

Neurodynamics can be found at: www.reading.ac.uk/cinn

The proceedings of the summer school can be found here:

www.reading.ac.uk/cinn/news/cinn-summerschool.aspx

Etienne B. Roesch