Decreased Beta Power As A Predictor Of Memory Encoding

Dr. Simon Hanslmayr

University of Birmingham, School of Psychology

www.uni-konstanz.de/oscillations-and-cognition/

Why brain oscillations?

Brain Oscillations = Communication Mechanisms! (Varela et al., 2001; Nat Neurosci; Fries et al., 2005; Trends in Cogn Sci)

Large Scale

Phase:

Local Scale

Power:

The Classic Model „Fire together wire together“

Time

Brain Oscillations = Neural Synchronization = Memory

Sync Desync

-0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 3 7

13

35

100

Time [sec]

Fre

qu

en

cy [H

z]

Gamma

Beta/ Alpha

Theta

???

Theta and Gamma Sync. positively related to memory • Rutishauser et al. (2010) Nature • Fell & Axmacher (2011) Nat Rev Neurosci

How is desynchronization related to memory?

Brain Oscillations and Memory

Staudigl & Hanslmayr (2013) Curr Biol

Alpha/Beta power decreases predict memory formation

0.05 0.01

Hit vs. Miss

p-level

b – Power (13-20 Hz)

-0.5 0 0.5 1 1.5

Time (sec.)

-20

-10

0

10

Po

we

r (%

)

Hit Miss

Hanslmayr et al. (2009) Cereb Cortex Fellner et al. (2013) NeuroImage

Noh et al. (2014) NeuroImage Long et al. (2013) NeuroImage

Rem. vs. Forg.

-0.5 0 0.5 1 1.5 2

Fre

qu

en

cy [H

z]

Time [sec.]

4

6

8

10

12

14

16

18

2.5

b

t-values

1.5 2.0

b – Source b – SME

Pcorr<0.05

EEG-fMRI during encoding

SME

Beta x BOLD

SME & Beta

IFG

(~ BA 45/46)

IFG

(~ BA 9)

-0.5

-0.4

-0.3

-0.2

-0.1

0

0.1

0.2

0.3

Eig

en

va

lue

s

BA 9 BA 45/46

Rem Forg.

Hanslmayr et al. (2011) J Neurosci

• Question

– Are beta power decreases in the left IFG epiphenomenal?

– Are they causally relevant to memory formation?

– Can we entrain beta oscillations in the left IFG and would that impair memory formation?

Summary and Question

EEG-BOLD correlations

rTMS-EEG and Encoding

left IFG

(~BA 9)

2.5 sec 1.0 – 2.0 sec

18.7 Hz 10.7 Hz

6.8 Hz

Sham [18.7 Hz]

0.5 sec

rTMS (18 pulses)

1 sec

Hypothesis: Entrainment of beta oscillations at left IFG specifically impairs memory encoding

rTMS-EEG - Behavioral Data

Hanslmayr, Matuschek, Fellner (2014) Curr Biol

*

% R

ecal

led

30

35

40

45

50

18.7 Hz 11.7 Hz 6.8 Hz Sham

Memory Performance

0

10.7 Hz 18.7 Hz 6.8 Hz Sham

Single Trials

Entrainment Echo

clean EEG

ERP

ERP Pow.

EEG contaminated by artefacts

rTMS-EEG – Evidence for Entrainment

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

0.5

0.0

-0.5

Am

plit

ude

[m

v]

Time after last TMS pulse [sec.]

ERPs (17.5 – 19.5 Hz)

18.7 Hz

Sham

rTMS-EEG – Evidence for Entrainement

Hanslmayr, Matuschek, Fellner (2014) Curr Biol

rTMS-EEG – Evidence for Entrainement

3.6 3.9

t-val

TMS

pcorr<0.005

Hanslmayr, Matuschek, Fellner (2014) Curr Biol

rTMS-EEG – Evidence for Entrainement

-0.04

-0.02

0.00

0.02

0.04

0.06

0.08

0.10

0.4 - 0.9 0.9 - 1.4

low Entr

high Entr

*

n.s.

0

2

4

6

8

10

-2

-4

Me

mo

ry I

mp

airm

ent

(Sh

am

- 1

8.7

)

High vs. Low Entrainment Group

0.4 – 0.9 sec 0.9 – 1.4 sec .0000

.1000

.2000

.3000

.4000

.5000

.6000

.7000

.8000

18.7 Hz 10.7 Hz 6.8 Hz

Low Entr.

High Entr.

80

0

10

20

30

40

50

60

70

ER

P P

ow

er

[% s

ig.

ch

an

ge

]

Beta Echo (TMS – Sham) (0.4 – 0.9 sec; 17.5 – 19.5 Hz)

*

Hanslmayr, Matuschek, Fellner (2014) Curr Biol

Open Question

• Higher frequencies? tACS

Verena Braun

The information via desynchronization hypothesis

No Synchrony Low Synchrony High Synchrony Spikes

0.1 0.2 0.3 0.4 0.5 0.6 0.1 0.2 0.3 0.4 0.5 0.6 0.1 0.2 0.3 0.4 0.5 0.6

0.5

0.0 EEG

0 5 10 15 20 25 30

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8 EEG - Power

Frequency [Hz]

0

1

2

3

4

5

6

No Sync. Low Sync. High Sync.

Entr

opy [

H]

Information

0.1 0.2 0.3 0.4 0.5 0.6 0.7 1

2

3

4

5

0

Entr

opy [

H]

EEG - Power

Information vs. Synchrony

Hanslmayr, Staudigl, Fellner (2012) Front Hum Neurosci

Take home messages

• Alpha/Beta desynchronization plays a very active and causal role for memory formation and may represent information.

• Brain oscillations can be entrained by external stimulation and observed in terms of echoes.

Wednesday, 4pm-6pm Poster Session 3 Rooms P9-P11

• Cognition and Oscillations Lab:

Tobias Staudigl (Konstanz Univ.)

Thank you for your attention …

Gerd Waldhauser (Konstanz Univ.)

Marie-Christin Fellner (Konstanz Univ.)

Jonas Matuschek (Konstanz Univ.)

Verena Braun (Birmingham Univ.)

Rolandas Stonkus (Birmingham Univ.)

Sebastian Michelmann (Birmingham Univ.)

• Funding:

rTMS-EEG – Evidence for Entrainement

SME Sham (0.0 – 0.5 sec)

8

0.0

- 8

Rem

< F

org

R

em

> F

org

5 10 15 20 25

*

Pow

er

[% s

ignal change]

Frequency [Hz]

18.5 Hz

Relation between entrained frequency and spontaneous frequency

Hanslmayr, Matuschek, Fellner (2014) Curr Biol

rTMS-EEG – Evidence for Entrainement

14 15 16 17 18 19 20 21 22

1

2

3

4

5

6

Low Dev. High Dev. High Dev.

Num

ber

of subje

cts

SME Peak Frequency

-0.2

-0.1

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.4 - 0.9 0.9 - 1.4

Low Dev

High Dev

70

60

50

40

30

20

10

0

-10

-20

ER

P P

ow

er

[% s

ig. change]

Beta Echo (18.7 Hz – Sham) (17.5 – 19.5 Hz)

0.4 – 0.9 sec 0.9 – 1.4 sec

Low Dev.

High Dev.

*** **

Relation between entrained frequency and spontaneous frequency

Hanslmayr, Matuschek, Fellner (2014) Curr Biol