Kein Folientitel

1
Spezielle Themen der Biologischen Psychologie
Vorlesungsunterlagen, SS 2009 Univ. Prof. Dr.
Wolfgang Klimesch Teil 2 Memory
2
VO SpezBio, SS 2009 Univ Prof. Dr. Wolfgang
Klimesch THEMENÜBERSICHT Klausur
Mo Empfohlene und ergänzende Lit. zur VL
Memory Fries, P., Fernandez, G. Jensen, O.
(2003). When neurons form memories. Trends in
Neuroscience, 26 (3), 123-124. Hoffman, K.L.,
McNaughton, B.L. (2002). Coordinated reactivation
of distributed memory traces in primate
neocortex. Science, 297, 2070-2073. Horn, G.
(2004). Pathways of the past The imprint of
memory. Nature Review Neuroscience, 5,
108-121. Klimesch, W., Sauseng, P., Gerloff, C.
(2003). Enhancing cognitive performance with
repetitive transcranial magnetic stimulation at
human individual alpha frequency. European
Journal of Neuroscience, 17, 1129-1133. Knowlton,
B.J. Eldridge, L.L. (2006). Mnemonic binding in
the medial temporal lobe. In H.D. Zimmer, A.
Mecklinger, U. Lindenberger (Eds.) Handbook of
binding and memory Perspectives from cognitive
neuroscience (pp. 493-516). Oxford Oxford
Unversity Press. Nader, K. (2003). Memory traces
unbound. Trends in Neuroscience, 26 (2),
65-72. Nakazawa, K., McHugh, T.J., Wilson, M.A.
Tonegawa S. (2004). NMDA receptors, place cells
and hippocampal spatial memory. Nature Review
Neuroscience, 5, 361-372. Simons, J.S. Spiers,
H.J. (2003). Prefrontal and medial temporal lobe
interactions in long-term memory. Nature Review
Neuroscience, 4, 637-648. Varela, F.J., Lachaux,
J.P., Rodriguez, E., Martinerie, J. (2001).
The Brainweb Phase synchronization and
large-scale integration. Nature Reviews
Neuroscience, 2, 229-238. Vogt et al. (1998).
High-frequency components in the alpha band and
memory performance. Journal of Clinical
Neurophysiology, 15, 167-172. Wilson, M. A.
McNaughton, B.L. (1994). Reactivation of
hippocampal ensemble memories druing sleep.
Science, Vol. 265, 676-679
3
Memory trace and coordinated network activation
Pre-task Monkey may be thinking on a reward
(e.g. on a fruit)
In a study by Hoffman and McNaughton (2002),
electrophysiological recordings were obtained
from - the posterior parietal cortex (PPC), -
the dorsal prefrontal cortex (dPFC), - Motor
cortex, and - somatosensory cortex. In each of
these brain regions 144 electrode contacts were
used. Thus, recordings were analysed from a total
of 576 cortical sites.
Task Remember location of button to receive
reward (juice)
Location neuron/network reward neuron/network
Co-activation of neurons with preserved temporal
pattern reflects memory trace
4
Memory traces as distributed networks
A memory network connecting the
somatosensorycortex (SS), posterior parietal
cortex (PP), and motorcortex (M). Red squares
indicate the areas that showed memory trace
reactivation red dotted lines indicate regions
that reactivated together.
Fig.1. Dorsal view of electrode recording sites
registered to a preoperative magnetic resonance
image. Brain regions sampled included dorsal
prefrontalcortex (PFC), somatosensorycortex (SS),
posterior parietal cortex (PP), and motorcortex
(M). Each square represents the area covered by a
chronically implanted 12 by 12 array of
electrodes. Red squares indicate the areas that
showed memory trace reactivation red dotted
lines indicate regions that reactivated together.
APanterior, posterior.
5
(Fig. 2 A) Cell-by-time ?ring rate matrix from
somatosensory cortex within the ?rst 5 min of one
task. The task was repeated approximately every
40s, matching the response periodicity of several
cells. Periodic responses continued for the
duration of the task (not shown). Firing rates
are truncated at 10 Hz fori llustrative purposes.
Firing frequency in Hz
30 selected cells
(Fig. 2B) Peri-event time histograms of neural
?ring related to 52 juice reward events from the
alley maze task.Juice delivery was preceded by a
successful touch of the touch screen cue. After a
correct touch, the monkey retracted his arm and
consumed the juice before resuming lever pulling.
Task-related activity was present in all four
brain regions, and a variety of response types
was seen in each region.
About 20 of the cells examined had event-related
activity of similar magnitude to those shown
here, indicating that neurons were in some way
responsive to elements of the task.
6
Quanti?cation of memory trace reactivation based
on cell-pair ?ring rate correlation
distributions. (Fig. 3A) Cell-by-time ?ring rate
matrix from one recording session of 99
simultaneously recorded cells. Each row
represents the binned ?ring rate of one cell,
positioned according to brain region. Each column
represents the ?ring rate of each cell at a given
bin of time. For illustrative purposes, ?ring
rates were truncated at 20Hz. (Fig. 3B) Foreach
epoch, rate-independent cell-pair correlations
were made for all eligible pairs of cells (n
4838). (Top) Histogram of cell-pair correlations
during the Taskepoch. (Middle) Cell-pair
correlations of Task versus Rest1. (Bottom)
Cell-pairc orrelations of Task versus Rest2. The
cell-pair correlations during Task are more
similar to those of Rest2 than to those of Rest1
(p lt 0.001). (Fig. 3C)Explained variances (EV) of
cell-pair correlations from pairs within and
across brain regions after subtraction of the
control values. Error bars represent 95
con?dence limits. All red bars, and no blue bars,
are signi?cant (P lt 0.05).
7
Fig.4. Preservation of temporal bias of cell-pair
interactions after Task. (A)Example of
cross-correlograms (CCG) of one motor cortex and
one somatosensory cell during each epoch. The
number of coincidences within each 10-ms bin is
plotted. The bias (dotted blackline ) during
Rest2 (bottom) is similar to the bias evoked by
the task (middle), which was different from restb
eforehand (top). (B) Distributions of temporal
bias during Task and Rest epochs. The bias during
task is plotted against the difference in biases
of Rest2 and Rest1. Within-region CCG biases are
plotted in the ?rst row, and between-region
biases areplotted in the second row. The
proportion of CCGs with preserved temporal bias
(those falling in the white quadrants) is listed
in the upper right quadrant. For signi?cant
correlations, the majority of points fall in the
white quadrants, indicating that the correlation
is not driven strictly by the magnitude of
outliers.
8
Hippocampus and consolidation of memories
Reactivation of learned information
Experimental evidence for encoding processes in
the hippocampal formation. In an intersting study
Wilson Mc Naughton (1994 Science, Vol. 265,
676-679) trained rats to find food pellets in a
maze. Rats were implanted with microelectrodes in
the area CA1 of the hippocampus. In each rat, the
activity of 50 to 100 cells was recorded during
three conditions (i) SWS sleep before testing
( pre-behavioral sleep PRE) (ii) running (RUN)
when rats actually were searching for food (iii)
post-behavioral SWS sleep (POST) An example of
the experimental situation is shown on this
page. X-shaped four arm maze
9
Example for place fields in different types of
mazes
Rectangular maze
Four arm maze
Rectangular maze
10
Cross-correlations be-tween pairs of CA1 cells
overlapping and non-overlapping place fields.
Note that theta oscillations can be observed
during the RUN, but not during PRE and POST SWS
sleep. Thus, during RUN the firing pattern of
encoding cells is modulated by rhythmic theta
activity. However, during POST, cells with
overlapping place fields show a highly correlated
pattern, with irregular bursts possibly modulated
by delta activity.
11
PRE
Connectivity matrix of a network of 42 CA1 cells
(represented as dots) which were selected at
random. Lines indicate positive correlations,
color strength of correlation (red high, blue
low). The first column shows all correlations,
the second shows correlations above 0.05. Note
that most of the highly correlated pairs during
RUN appear also during POST.
RUN
POST
12
Memory systems in the brain An Overview
It is well established that brain structures in
the prefrontal, medial temporal and parietal lobe
are closely linked to different types of memory
processes. Within the medial temporal lobe the
hippocampus is involved in the creation and
retrieval of arbitrary conjunctions between items
and contextual information and by this it
supports the recall of specific studied details.
Conversely the medial temporal lobe (MTL) cortex
surrounding the hippocampus (i.e. in particular
the perirhinal cortex) represents sensory
information and also binds these information
units to create familiarity supporting memory
representation.
While the MTL is associated with the encoding
storage and retrieval of long-term memories,
prefrontal lobe structures have been linked to
control processes such as selection, inhibition
or monitoring that allow to adjust the output of
the MTL systems to task and situational demands.
Additionally, perceptual memory systems have been
localized in sensory specific cortices. These
structures represent sensory features of a
perceived stimulus and they provide memory of the
items appearance. Perceptual implicit memory
tasks bear on these sensory memories. Meanwhile
there is growing evidence that working memory is
realised by a similar network of prefrontal and
medial temporal areas. The prefrontal cortex
together with posterior parietal regions actively
maintains and transforms specific memory contents
that are stored in the same posterior neural
structures that were found active in long-term
memory. Hence, memory is implemented in a wide
cortical network that comprises many different
sub-systems. For reviews, cf. e.g., Simons
Spiers, 2003 and Knowlton, 2006. See illustration
on next two pages.
13
An illustration of the memory network consisting
of the - prefrontal cortex (PFC) - and
medial temporal lobe (MTL) structures
(Adapted from Simons Spiers, 2003)
14
Brain circuits for encoding and
consolidation Papez Circuit
Basolateral
Circuit
Gyrus cinguli
Cingulum Anterior Thalamus
Fornix Mammilary
body Hippocampus
Mediodorsal Thalamus
Basal Forebrain
Amygdala
15
Substructures within medial temporal lobe and
their contribution to different memory tasks
(Adapted from Knowlton Eldridge, 2006).
16
EEG, Memory Performance and the Improvement of
memory performance
Evidence for hypothesis that memory performance
is related to LARGER (normalized) UPPER ALPHA
POWER. From Vogt et al. (1998).
.50 .25 0
-.25 .50
Ongoing EEG, eyes closed P3.
r
A) Correlation spectral estimates with memory
performance
B) Normalized power
(For each frequency step of .25 Hz power is
expressed as the percentage of total power,
defined as the mean of all power estimates within
5-15 Hz)
Upper Alpha powerM gt M-
0 100 200 300 400 500
M-
M



5 6 7 8 9
10 11 12 13 14 15
60 subjects (35 females, mean age 21 25 males
mean age 25)
17
EEG alpha power (eyes closed resting condition)
correlates with intelligence (as measured by the
LGT 3 and IST 70) Large sample (N 74).
Doppelmayr, Klimesch, Stadler, Pöllhuber, Heine
(2002).
The IST-70 primarily measures kognitive skills
stored in LTM (such as mental rotation, semantic
analogies etc) semantic memory
The LGT-3 primarily measures learning ability
knowledge acquired during experiment
18
Conclusion for upper alpha A large difference
between resting (or reference) power and
decreased test (or phasic) power is related to
good performance.
Hypothesis Is it possible to generate a best
case by applicating repetitive Transcranial
Magnetic Stimulation (rTMS) at individual alpha
frequency (IAF) during a reference interval
UPPER ALPHA Best Case
Worst Case
rTMS at IAF
?
Reference power
Reference power
Power Decrease
Increase
Power Decrease
Increase
Ref. power
?
Phasic power
Phasic power
Phasic power
19
rTMS at IAF improves memory and cognitive
performance Klimesch, W., Sauseng, P. C.
Gerloff (2003)
TASK Mental rotation. On each trial a set of six
cubes (three in an upper and 3 in a lower row)
were presented on a computer monitor. The target
cube always appeared in the middle position of
the first row and was marked by a surrounding
square. Subjects had to decide which of the other
five cubes matched the mentally rotated target. 9
blocks, each with 8 trials were used. Analogous
to the IST-70, the target cube differed for a
series of trials whereas the 5 test cubes
remained the same. DESIGN rTMS stimulation
frequencies were IAF 1 Hz, IAF 3 Hz or 20 Hz
. They were delivered at P6 (above the
intraparietal sulcus), at Fz and rotated by 90
degrees over P6 for sham resulting in 9
stimulation conditions. 9 blocks of cubes were
assigned randomly to each of the 9 stimulation
conditions rTMS A MagStim Rapid stimulator
(MagStim, Whitland, UK) with a 70-mm figure-eight
coil was used. The output strength of rTMS was
set to 110 of the subjects motor threshold
defined as the intensity needed for eliciting
MEPs of at least 50 µV recorded at the thumb of
the left hand in 50 of single pulses delivered
to the contralateral motor cortex. Mean intensity
for rTMS was 57.3 of maximal stimulator output.
In order to keep the energy of rTMS pulses
constant between different frequencies, a fixed
number of 24 pulses were delivered in each
condition. Thus, the duration of pulse trains
varied between 1.2 to about 4.8 sec (depending on
IAF). Experiment 1 ( 16 subjects) investigation
of behavioral effects. Experiment 2 (6 subjects)
investigation of upper alpha power and ERD
20
rTMS
21
Design

• EEG Rating of IAF
• rTMS stimulation frequencies were IAF 1 Hz,
  IAF 3 Hz or 20 Hz . They were delivered at P6
  (above the intraparietal sulcus), at Fz and
  rotated by 90 degrees over P6 for sham.
• mental rotation following rTMS trains (9 blocks,
  each containing 8 rotations) gt
  
• Experiment 1 reaction time and number of
  correct responses.
• Experiment 2 EEG

22
Site Fz
(N15) significant (plt.01) increase of correct
answers after upper alpha stimulation over
frontal and right parietal sites compared to
sham stimulation

increase of correct answers
Site P6

increase of correct answers

plt.01
23
Upper alpha power in the reference interval prior
to sham or verum stimulation (around 30 sec after
TMS train of previous trial)
Significant (plt.05) differences between sham
stimulation and right parietal and frontal verum
stimulation over right and left frontal and
frontocentral sites.
24
Upper alpha power during a mental rotation task
following sham or verum stimulation
25
Selectively stronger desynchronization (upper
alpha ERD) after verum stimulation compared to
sham condition
Significant (plt.01) differences between sham
stimulation and right parietal and frontal verum
stimulation over right parietal sites, the vertex
and dorsolateral prefontal areas (bihemispheric)
26
Visual WM paradigm
27
Memory capacity and the lateralized increase in
alpha power
Left (contra) Right negative value Left
(ipsi) Right positive value
Contra Ipsi
28
Modulation of WM-Capacity by Alpha rTMS
29
Externale Modulation von WM-Kapazität durch rTMS
Sauseng et al. (under revision) Curr Biol