Please select what you would like included for printing:
Copy the text below and then paste that into your favorite email application
Enter your phone number above to have directions sent via text
This site is protected by reCAPTCHA and the  Google <a href="https://policies.google.com/privacy" target="_blank">Privacy Policy</a> and <a href="https://policies.google.com/terms" target="_blank">Terms of Service</a> apply
Service map data &copy; <a href="https://www.openstreetmap.org/copyright">OpenStreetMap</a> contributors
<a href="/articles/s41467-023-44632-x/metrics" data-track="click" data-track-action="view metrics" data-track-label="link" rel="nofollow">Metrics details</a>
Actions are biased by the outcomes they can produce: Humans are more likely to show action under reward prospect
Such motivational biases derive not only from biased response selection
but also from biased learning: humans tend to attribute rewards to their own actions
but are reluctant to attribute punishments to having held back
The neural origin of these biases is unclear
it remains open whether motivational biases arise primarily from the architecture of subcortical regions or also reflect cortical influences
the latter being typically associated with increased behavioral flexibility and control beyond stereotyped behaviors
Simultaneous EEG-fMRI allowed us to track which regions encoded biased prediction errors in which order
Biased prediction errors occurred in cortical regions (dorsal anterior and posterior cingulate cortices) before subcortical regions (striatum)
These results highlight that biased learning is not a mere feature of the basal ganglia
but arises through prefrontal cortical contributions
revealing motivational biases to be a potentially flexible
also credit assignment is subject to motivational biases
with enhanced credit for rewards given to actions
but diminished credit for punishments given to inactions
Both candidate models predict that BOLD signal in striatum should be better described by biased compared with “standard” prediction errors
the model proposing a prefrontal influence on striatal processing makes a notable prediction about the timing of signals: information about the selected action and the obtained outcome should be present first in prefrontal circuits to then later affect processes in the striatum
While fMRI BOLD recordings allow for unequivocal access to striatal activity
the sluggish nature of the BOLD signal prevents clear inferences about temporal precedence of signals from different regions
We thus combined BOLD with simultaneous EEG recordings which allowed us to precisely characterize learning signals in both space and time
which predicts better learning of actions following reward and failure to unlearn inaction following a loss
This finding provides a putative computational mechanism for how motivational action biases can arise through learning and aggravate with increased experience
We further show that BOLD signal in a range of cortical and subcortical regions is better explained by biased than by standard prediction errors
electrophysiological correlates of cortical prediction errors arise earlier than correlates of subcortical prediction errors
consistent with an influence of cortical over subcortical regions in biasing the learning of actions and inactions
this work provides evidence for a cortico-striatal basis of biased learning of action-outcome contingencies that may drive the formation of motivational action biases
We performed regression analyses to test whether (a) responses were biased by the valence of prospective outcomes (Win/Avoid)
reflecting biased responding and/or learning
and (b) whether response repetition after positive vs
negative outcomes was biased by whether a Go vs
these results suggested that behavioral adaptation following rewards and punishments was biased by the type of action that led to this outcome (Go or NoGo)
this analysis only considered behavioral adaptation on the next trial
and could not pinpoint the precise algorithmic nature of this learning bias
it did not provide trial-by-trial estimates of action values as required for model-based fMRI and EEG analyses to test for regions or time points that reflected biased learning
We thus analyzed the impact of past outcomes on participants’ choices using computational RL models
in line with the asymmetric pathways model
we added both the response bias and the learning bias
For the full model space (M1-M5) and model definitions
which relied on only a single mechanism (i.e.
biased learning from rewarded Go and punishment NoGo actions)
B BOLD signals correlated positively to “standard” RL prediction errors in several regions
C Left panel: Regions encoding both the standard PE term and the difference term to biased PEs (conjunction) at different cluster-forming thresholds (1 &lt;z &lt; 5
Clusters significant at a threshold of z &gt; 3.1 are surrounded by black edges
BOLD was significantly better explained by biased learning than by standard learning
Right panel: 3D representation with all seven regions encoding biased learning (and used in fMRI-informed EEG analyses)
there was higher BOLD signal in contralateral motor cortex and operculum as well as ipsilateral cerebellum when contrasting hand responses against each other (see below)
These results are in line with previous results on outcome processing and response selection and thus assure the general data quality
Significant encoding of both components (with the same sign) provides strong evidence for encoding of biased prediction errors PEBIAS
The PEDIF term itself has no substantive neural interpretation; it is merely an implicit model comparison of a null model (PESTD) against a full model (PEBIAS)
for voxels for which both PESTD and PEDIF are significant
one can conclude that the BOLD signal correlates with the full biased prediction error term PEBIAS
and that this correlation is significantly stronger than for the baseline prediction error term PESTD
BOLD signal in these regions was better described (i.e.
more variance explained) by biased learning than by standard prediction error learning
When testing for parametric correlates of PE magnitude
thereby effectively testing for correlations with the absolute PE magnitude (i.e.
Note that PE valence was identical for standard and biased PEs
only PE magnitude could distinguish both learning models
both midfrontal theta power (negatively) and beta power (positively) encoded PE valence
delta power encoded PE magnitude (positively)
This encoding was only significant for biased PEs
midfrontal EEG power also reflected biased learning
we tested whether the identified EEG phenomena were correlated with trial-by-trial BOLD signal in identified regions
this allowed us to test whether EEG correlates of cortical learning precede EEG correlates of subcortical learning
We performed analyses with and without PEs included in the model
which yielded identical results and suggested that EEG-fMRI correlations did not merely result from PE processing as a “common cause” driving signals in both modalities
EEG-fMRI correlations reflected incremental variance explained in EEG power by the BOLD signal in selected regions (even beyond variance explained by the model-based PE estimates)
providing the strongest test for the hypothesis that BOLD and EEG signal reflect the same neural phenomenon
As the timeseries of all seven regions were included in one single regression
their regression weights reflected each region’s unique contribution
BOLD signal from prefrontal cortical regions correlated with midfrontal EEG power earlier after outcome onset than did striatal BOLD signal:
First, dACC BOLD was significantly negatively correlated with alpha/ theta power early after outcome onset (100–575 ms, 2–17 Hz, p = 0.016, two-tailed; Fig. <a data-track="click" data-track-label="link" data-track-action="figure anchor" href="/articles/s41467-023-44632-x#Fig5">5A</a>). This cluster started in the alpha/ theta range and then spread into the theta/delta range (henceforth called “lower alpha band power”). It was not observed in the EEG-only analyses reported above.
normalized to the peak of the time course of each region
dACC-lower alpha band correlations emerged first
followed by (negative) PCC-theta correlations and finally positive striatum-beta correlations
The reverse approach using lower alpha (E)
theta (F) and beta (G) power as trial-by-trial regressors in fMRI GLMs corroborated the fMRI-informed EEG analyses: Lower alpha band power correlated negatively with the dACC BOLD
theta power negatively with vmPFC and PCC BOLD
and beta power positively with striatal BOLD
H Schematic overview of the main EEG-fMRI results: dACC encoded the previously performed response and correlated with early midfrontal lower alpha band power
vmPFC/ PCC (correlated with theta power) and striatum (correlated with beta power) both encoded outcome valence
but had opposite effects on subsequent behavior
Note that activity in these regions temporally overlaps; boxes are ordered in temporal precedence of peak activity
We thus discuss vmPFC and PCC together in the following
Note that time-frequency power is estimated over temporally extended windows (400 ms in our case)
which renders any interpretation of the “onset” or “offset” of such correlations more difficult
these results are consistent with an “external model” of motivational biases arising from early cortical processes biasing later learning processes in the striatum
While the clusters of EEG-fMRI correlation in the theta/delta and beta range matched the clusters identified in EEG-only analyses
the cluster of negative correlations between dACC BOLD and early midfrontal lower alpha band power was novel and did not match our expectations
Given that these correlations arose very soon after outcome onset
we hypothesized that dACC BOLD and midfrontal lower alpha band power might reflect a process occurring even before outcome onset
such as the maintenance (“memory trace”) of the previously performed response to which credit may later be assigned
We therefore assessed whether information of the previous response was present in dACC BOLD and in the lower alpha band around the time of outcome onset
C BOLD signal differences between Go and NoGo actions (activation by either left or right Go actions compared to the implicit baseline in the GLM
which contains the NoGo actions; left panel) and left vs
right hand responses (right panel) at the time or responses
Response-locked dACC BOLD signal was significantly higher for Go than NoGo actions
D BOLD signal differences between Go and NoGo actions at the time of outcomes
Outcome-locked dACC BOLD signal (and BOLD signal in other parts of cortex) was significantly lower on trials with Go than on trials with NoGo actions
suggesting that it did not reflect typical fatigue/ time-on task effects often observed in the alpha band
both dACC BOLD signal and midfrontal lower alpha band power contained information about the previously performed response
consistent with the idea that both signals reflect a “memory trace” of the response to which credit is assigned once an outcome is obtained
high pgACC BOLD predicted a higher likelihood of switching
likening it with the circuits formed by vmPFC and PCC (χ2(1) = 15.559
We also inspected the raw upsampled HRF shapes per region per condition
confirming that differential relationships were not driven by differences in HRF shapes across regions
participants’ behavior was best described by a computational model featuring faster learning from rewarded Go responses and slower learning from punished NoGo responses
Neural correlates of biased PEs were present in BOLD signals in several regions
These regions exhibited distinct midfrontal EEG power correlates
correlates of prefrontal cortical BOLD preceded correlates of striatal BOLD: Trial-by-trial dACC BOLD correlated with lower alpha band power immediately after outcome onset
followed by PCC (and vmPFC) BOLD correlated with theta power
These results suggest that the architecture of the asymmetric striatal pathways might not be the only neural structure that gives rise to motivational learning biases; instead
the PFC might critically contribute to these biases
corroborating that PFC outcome processing occurs before the time of EEG correlates of striatal BOLD
The only model consistent with our data assumes recurrent connections between PFC and striatum
these results are in line with a model of PFC biasing striatal outcome processing
giving rise to motivational learning biases in behavior
Prefrontal influences on striatal processes might thus be a common signature of both motivational response and learning biases
dACC exhibited stronger BOLD signal for Go than NoGo responses at the time of participants’ response
but this pattern reversed at the time of outcomes
This reversal rules out the possibility that response-locked BOLD signal simply spilled over into the time of outcomes
Future research will be necessary to corroborate such a motor “memory trace” in dACC
the dACC might be in a designated position to inform subsequent outcome processing in downstream regions by modulating the learning rate as a function of the previously performed response and the obtained outcome
Rather than striatal circuits being sufficient for the emergence of motivational biases
the more “flexible” PFC seems to play an important role in instructing downstream striatal learning processes
This dissociation suggests important differences in the role of the striatum in these two processes
The frequency-specific nature of these EEG-fMRI correlations further suggests that they are signatures of task-induced events that are specific to the trial phase and unlikely to reflect general anatomical connectivity
while these EEG-fMRI findings on outcome processing resemble our previous EEG-fMRI findings on action selection in that prefrontal signals precede striatal signals
they are dissociated in terms of the frequency specificity
highlighting the distinct roles of the striatum in these processes
this EEG surprise signal was only significantly correlated with the biased (but not the standard) PE term
corroborating that the surprise attributed to outcomes depends on the previously performed response in line with motivational learning biases
though with different consequences for future action policies
our analysis controlled for BOLD signal in motor cortex
an alternative candidate source for beta power
suggesting that late midfrontal beta power did not merely reflect motor cortex beta
Even if the striatum is not the generator of the beta oscillations over the scalp
their true (cortical) generator might be tightly coupled to the striatum and thus act as a “transmitter” of striatal beta oscillations
the analyses using trial-by-trial beta power to predict BOLD yielded significant clusters in dlPFC and SMG
two candidate regions for such a “transmitter”
our main conclusion would hold: Biased learning is present in cortical regions early after outcome onset
which cannot be a consequence of striatal input
but must constitute an independent origin of motivational learning biases
but failed to capture crucial aspects of the biased learning under investigation
Future studies comprising larger samples of participants should explore alternative implementations to reliably quantify individual differences in these learning biases
The PFC is typically believed to facilitate goal-directed over instinctive processes
PFC involvement into biased learning suggests that these biases are not necessarily agents’ inescapable “fate”
but rather likely act as global “priors” that facilitate learning of more local relationships
They allow for combining “the best of both worlds”—long-term experience with consequences of actions and inactions together with flexible learning from rewards and punishments
participants saw a gem-shaped cue for 1300 ms which signaled whether they could potentially win a reward (Win cues) or avoid a punishment (Avoid cues) and whether they had to perform a Go (Go cue) or NoGo response (NoGo cue)
or no (NoGo) button while the cue was presented
Only one response option was correct per cue
Participants had to learn both cue valence and required action from trial-and-error
After a variable inter-stimulus-interval of 1400–1600 ms
Potential outcomes were a reward (symbolized by coins falling into a can) or neutral outcome (can without money) for Win cues
and a neutral outcome or punishment (symbolized by money falling out of a can) for Avoid cues
correct responses were followed by positive outcomes (rewards/ no punishments) on only 80% of trials
while incorrect responses were still followed by positive outcomes on 20% of trials
Trials ended with a jittered inter-trial interval of 1250–2000 ms
yielding total trial lengths of 4700–6650 ms
Participants gave left and right Go responses via two button boxes positioned lateral to their body
but only one button per box was required in this task
When participants accidentally pressed a non-instructed button
they received the message “Please press one of the correct keys” instead of an outcome
these responses were recoded into the instructed button on the respective button box
such trials were modeled with a separate regressor
participants were instructed that each cue could be followed by either reward or punishment
and that the rewards and punishments were converted into a monetary bonus upon completion of the study
They performed an elaborate practice session in which they got familiarized first with each condition separately (using practice stimuli) and finally practiced all conditions together
They then performed 640 trials of the main task
separated into two sessions of 320 trials with separate cue sets
Introducing a new set of cues allowed us to prevent ceiling effects in performance and investigate continuous learning throughout the task
Each session featured eight cues that were presented 40 times
participants could take a self-paced break
The assignment of the gems to cue conditions was counterbalanced across participants
and trial order was pseudo-random (preventing that the same cue occurred on more than two consecutive trials)
Sum-to-zero coding was employed for the factors
Type 3 p-values were based on likelihood ratio tests (implemented in the R package afex version 0.28.1)
We used a significance criterion of α = 0.05 for all the analyses
we used mixed-effects logistic regression to analyze “stay behavior”
whether participants repeated an action on the next encounter of the same cue
as a function of outcome valence (positive: reward or no punishment/negative: no reward or punishment)
outcome salience (salient: reward or punishment/neutral: no reward or no punishment)
We again included all possible random intercepts
To determine whether a Pavlovian response bias
or both biases jointly predicted behavior best
we fitted a series of increasing complex computational models
choice probabilities for all three response options (a) given the displayed cue (s) were computed from their action weights (modified Q-values) using a softmax function:
the deviation between the experienced outcome and expected outcome
This model contained two free parameters: the learning rate (ε) scaling the updating term and the feedback sensitivity (ρ) scaling the received outcome (i.e.
higher feedback sensitivity led to choices more strongly guided by value difference
akin to the role of the inverse temperature parameter frequency used in RL models):
choice probabilities were fully determined by action values
We initialized action values Q0 such that they reflected a “neutral” expected value for each action
Win cues could lead to reward (+1) or neutral (0) outcomes and Avoid cues to neutral (0) or punishment (−1) outcomes
A neutral expected value would assign equal probability to either possible outcome
because participants’ feedback sensitivity parameter ρ reflected how participants weighed the outcomes they received
also the initial values had to be multiplied with the feedback sensitivity to stay neutral between 0 and participants’ re-weighted positive/ negative outcome of +/−1*ρ
initial action values Q0 were set to 1/2*ρ (Win cues) and −1/2*ρ (Avoid cues)
which accounted for individual differences in participants’ overall propensity to make Go responses
scaling how positive/ negative cue valence (Pavlovian values) increased/ decreased the weights of Go responses:
Participants were instructed that a cue was either a Win cue (affording rewards or neutral outcomes) or an Avoid cue (affording neutral outcomes or punishments)
cue valence (Win/ Avoid) did not have to be learned instrumentally; instead
it could be inferred as soon participants experienced a non-neutral outcome
\(V(s)\) was set to +0.5 for Win cues and −0.5 for Avoid cues
Note that choosing different values than 0.5 would merely rescale the bias parameter π (e.g.
halving π with cue valences of +1 and −1) without any changes in the model’s predictions
The Pavlovian response bias affected left-hand and right-hand Go responses similarly and thus reflected generalized activation/inactivation by the cue valence
increasing the learning rate for rewards after Go responses and decreasing it for punishments after NoGo responses
The learning bias was specific to the response shown
thus reflecting a specific enhancement in action learning/ impairment in unlearning for that particular response
learning rates differed between response-outcome conditions in the following way:
subtracting the learning bias κ would no longer be symmetric
we first computed the difference between the baseline learning rate and the learning rates for punished NoGo responses and used this difference to compute the learning rate for rewarded Go responses:
this procedure is only guaranteed to work when ε0 &lt; 0.5
the difference term could become &gt;0.5 and the learning rate for rewarded Go responses would become &gt;1
we first computed the learning rate for rewarded Go responses and used the difference to the baseline learning rate ε0 to compute the learning rate for punished NoGo responses:
we included both the Pavlovian response bias and the learning bias
The same priors (for the same parameters) were used across different model implementations to not bias model comparison
Alternative hyperpriors did not change the results
For computing the participant-level parameters
ρ was exponentiated to constrain it to positive values
and the inverse-logit transformation was applied to ε
We selected the “winning” model based on the protected exceedance probability
After two localizer scans to position slices
we collected functional scans with a whole-brain T2*-weighted sequence (68 axial-oblique slices
interleaved multiband slice acquisition with acceleration factor 4
The first seven volumes of each run were automatically discarded
This sequence was chosen because of its balance between a short TR and relatively high spatial resolution
which was required to disentangle cue and outcome-related neural activity
Pilots using different sequences yielded that this sequence performed best in reducing signal loss in striatum
we removed the EEG cap and collected a high-resolution anatomical image using a T1-weighted MP-RAGE sequence (192 sagittal slices per slab
flip angle 8°) which was used to aid image registration
no fieldmap was collected due to time constraints
an additional DTI data collection took place; results will be reported elsewhere
images were high-pass filtered at 100 s and pre-whitened
we computed and applied co-registration of EPI images to high-resolution images (linearly with FLIRT using boundary-based registration) and to MNI152 2 mm isotropic standard space (non-linearly with FNIRT using 12 DOF and 10 mm warp resolution)
After visual inspection of the respective clusters
we created seven anatomical masks based on the probabilistic Harvard-Oxford Atlas (included in FSL; thresholded at 10%): striatum and ACC (see above)
motor cortex (combined precentral and postcentral gyrus)
temporooccipital part) and primary visual cortex (Lingual Gyrus
We then multiplied this functional mask with each of the seven anatomical masks
returning seven masks focused on the respective significant clusters
which were then used for signal extraction
we manually excluded voxels in pgACC belonging to a distinct cluster
Masks were back-transformed to each participant’s native space
For bar plots in Fig. <a data-track="click" data-track-label="link" data-track-action="figure anchor" href="/articles/s41467-023-44632-x#Fig3">3A</a>
we multiplied the anatomical masks of vmPFC and striatum specified above with the binarized outcome valence contrast
data were modeled using two event-related GLMs
we performed a model-based GLM in which we used trial-by-trial estimates of biased PEs as regressors
we used another model-free GLM in which we modeled all possible action x outcome combinations via outcome-locked categorical regressors while at the same time modeling response-locked left- and right-hand response regressors
This model free GLM also contained the outcome valence contrast reported as an initial manipulation check
Any region displaying truly biased learning should significantly encode both the standard PE term and the difference term
The standard PE and difference term were much less correlated (mean correlation of −0.020
we furthermore added four regressors spanned by crossing cue valence and performed action (Go response to Win cue
The model-free GLM included a separate regressor for each of the eight conditions obtained when crossing performed action (Go/NoGo) and obtained outcome (reward/no reward/no punishment/punishment)
We fitted four contrasts: (1) one contrast comparing conditions with positive (reward/no punishment) and negative (no reward/punishment) outcomes
used as a quality check to identify regions that encoded outcome valence; (2) one contrast comparing Go vs
NoGo responses at the time of the outcome; (3) one contrast summing of left- and right-hand responses
NoGo responses at the time of the response; and (4) one contrast subtracting right- from left-handed responses
As this GLM resulted in empty regressors for several participants when fitted on a block level
making it impossible to use the data of the respective blocks on a higher level
we instead concatenated blocks and performed a single GLM per participant
We therefore registered the data from all blocks to the middle image of the first block (default reference volume in FSL) using MCFLIRT
The first and last 20 s of each block did not feature any task-related events
such that carry-over effects of task events in the design matrix from one block to another were not possible
we added four regressors of no interest: one for the motor response (left = +1
and one for trials with invalid motor response (and no outcome respectively)
We also added nine or more nuisance regressors: the six realignment parameters from motion correction
and a separate spike regressor for each volume with a relative displacement of more than 2 mm (occurred in 10 participants; in those participants: M = 7.40
nuisance regressors were added separately for each block as well as an overall intercept per block
We convolved task regressors with double-gamma haemodynamic response function (HRF) and high-pass filtered the design matrix at 100 s
First-level contrasts were fit in native space
co-registration and reslicing was applied to participants’ contrast maps
which were then combined on a (participant and) group level using FSL’s mixed effects models tool FLAME with a cluster-forming threshold of z &gt; 3.1 and cluster-level error control at α &lt; 0.05 (i.e.
We recorded EEG data with 64 channels (BrainCap-MR-3-0 64Ch-Standard; Easycap GmbH; Herrsching
reference electrode at FCz) plus channels for electrocardiogram
and respiration (used for MR artifact correction) at a sampling rate of 1000 Hz
We placed MRI-compatible EEG amplifiers (BrainAmp MR plus; Brain Products GmbH
Germany) behind the MR scanner and attached cables to the participants once they were located in final position in the scanner
we fixated cables using sand-filled pillows to reduce artifacts induced through cable movement in the magnetic field
the MR helium pump was switched off to reduce EEG artifacts
we recorded the exact EEG electrode locations on participants’ heads relative to three fiducial points using a Polhemus FASTRAK device
no such data were available due to time constraints/ technical errors
in which case we used the average electrode locations of the remaining 32 participants
we had a hypothesis on outcome valence encoding in the beta range
We then applied linear baseline correction based on the 200 ms prior to cue onset and used ICA to detect and reject independent components related to eye-blinks
and residual MR artifacts (mean number of rejected components per participant: 32.694
we manually rejected trials with residual motion (for all 36 participants: M = 117.722
four participants for which more than 211 (33%) of trials were rejected were excluded from any further analyses (rejected trials after excluding those participants: M = 81.875
we computed a Laplacian filter with the spherical spline method to remove global noise (using the exact electrode positions recorded with Polhemus FASTRAK)
which we also used to interpolate previously rejected channels
This filter attenuates more global signals (e.g.
signal from deep sources or global noise) and noise (heart-beat and muscle artifacts) while accentuating more local effects (e.g.
We decomposed the trial-by-trial EEG time series into their time-frequency representations using 33 Hanning tapers between 1 and 33 Hz in steps of 1 Hz
every 25 ms from −1000 until 1300 ms relative to outcome onset
We first zero-padded trials to a length of 8 s and then performed time-frequency decomposition in steps of 1 Hz by multiplying the Fourier transform of the trial with the Fourier transform of a Hanning taper of 400 ms width
centered around the time point of interest
This procedure results in an effective resolution of 2.5 Hz (Rayleigh frequency)
which was more robust to the choice of exact frequency bins
To exclude the possibility of slow drifts in power over the time course of the experiment
we performed baseline correction across participants and trials by fitting a linear model for each channel/ frequency combination with trial number as predictor and the average power 250–50 ms before outcome onset as outcome
and subtracting the power predicted by this model from the data
This procedure is able to remove slow linear drifts in power over time from the data
it is equivalent to correcting all trials by the grand mean across trials per frequency in the selected baseline time window
we averaged power over trials within each condition spanned by performed action (Go/NoGo) and outcome (reward/no reward/no punishment/punishment)
We finally converted the average time-frequency data per condition to decibel to ensure that data across frequencies
We again performed model-free and model-based analyses
we sorted trials based on the performed action (Go/NoGo) and obtained outcome (reward/no reward/no punishment/punishment) and computed the mean TF power across trials for each of the resultant eight conditions for each participant
We tested whether theta power (average power 4–8 Hz) and beta power (average power 13–30 Hz) encoded outcome valence by contrasting positive (reward/no punishment) and negative (no reward/punishment) conditions (irrespective of the performed action)
We also tested for differences between Go and NoGo responses in the lower alpha band (6–10 Hz)
we employed two-sided cluster-based permutation tests in a window from 0 to 1000 ms relative to outcome onset
results were driven by a cluster that was at the edge of 1000 ms; to more accurately report the time span during which this cluster exceeded the threshold
we extended the time window to 1300 ms in this particular analysis
Such tests are able to reject the null hypothesis of exchangeability of two experimental conditions
but they are not suited to precisely locate clusters in time-frequency space
interpretations were mostly based on the visual inspection of plots of the signal time courses
by using the BOLD signal from different regions in a multiple linear regression
one can control for variance shared among regions (e.g.
changes in global signal; variance due to task regressors) and test which region is the best unique predictor of a certain EEG signal
any correlation between EEG and BOLD signal from a certain region reflects an association above and beyond those induced by task conditions
and Cz across participants in the range of 0–1000 ms
We first obtained a null distribution of maximal cluster mass statistics from 10000 permutations
we flipped the sign of the b-map of a random subset of participants
computed a separate t-test at each time-frequency bin (bins of 25 ms
1 Hz) across participants (results in t-map)
and finally computed the maximal cluster mask statistic (sum of all t-values) for any cluster (adjacent voxels above threshold)
we computed the same t-map for the real data
identified the cluster with the biggest cluster-mass statistic
and computed the corresponding p-value as number of permutations in the null distribution that were larger than the maximal cluster mass statistic in the real data
Note that these signals were selected based on the EEG-only results and not informed by the fMRI-informed EEG analyses
given the topography that reached far beyond midfrontal channels and over the entire frontal scalp
we used a much wider ROI (AF3/ AF4/ AF7/ AF8/ F1/ F2/ F3/ F4/ F5/F6/F7/F8/FC1/FC2/FC3/FC4/FC5/FC6/FCz/Fp1/Fp2/Fpz/Fz)
We extracted those maps and retained all voxels with t &gt; 2
These masks were applied to the trial-by-trial time-frequency data to create weighted summary measures of the average power in the identified clusters in each trial
For trials for which EEG data was rejected
we imputed the participant mean value of the respective action (Go/NoGo) x outcome (reward/no reward/no punishment/punishment) condition
Note that this approach accentuates differences between conditions
which were already captured by the task regressors in the GLM
but decreases trial-by-trial variability within each condition
This imputation approach is thus conservative
While trial-by-trial beta and theta power were largely uncorrelated
theta and alpha power were moderately correlated
warranting the use of a separate channel ROI for theta and using separate GLMs for each frequency band
We used mixed-effects logistic regression to analyze “stay behavior”
as a function of BOLD signal and EEG power in selected regions
we used the trial-by-trial HRF amplitude also used for fMRI-informed EEG analyses
we used the trial-by-trial EEG power also used in the EEG-informed fMRI analyses
Further information on research design is available in the <a data-track="click" data-track-label="link" data-track-action="supplementary material anchor" href="/articles/s41467-023-44632-x#MOESM3">Nature Portfolio Reporting Summary</a> linked to this article
The misbehavior of value and the discipline of the will
Catecholaminergic challenge uncovers distinct Pavlovian and instrumental mechanisms of motivated (in)action
Modeling avoidance in mood and anxiety disorders using reinforcement learning
The specificity of Pavlovian regulation is associated with recovery from depression
activation and valence in instrumental and Pavlovian responding
Deciding how to decide: self-control and meta-decision making
Go and no-go learning in reward and punishment: interactions between affect and effect
Frontal network dynamics reflect neurocomputational mechanisms for reducing maladaptive biases in motivated action
Dorsal striatal dopamine D1 receptor availability predicts an instrumental bias in action learning
Auto-maintenance in the pigeon: sustained pecking despite contingent non-reinforcement
Reluctance to vaccinate: omission bias and ambiguity
Attributions of responsibility and affective reactions to decision outcomes
Action dominates valence in anticipatory representations in the human striatum and dopaminergic midbrain
Action controls dopaminergic enhancement of reward representations
Striatal BOLD and midfrontal theta power express motivation for action
Dynamic dopamine modulation in the basal ganglia: a neurocomputational account of cognitive deficits in medicated and nonmedicated Parkinsonism
Opponent actor learning (OpAL): modeling interactive effects of striatal dopamine on reinforcement learning and choice incentive
Instructional control of reinforcement learning: a behavioral and neurocomputational investigation
Instructed knowledge shapes feedback-driven aversive learning in striatum and orbitofrontal cortex
Model-based influences on humans’ choices and striatal prediction errors
Neural computations underlying arbitration between model-based and model-free learning
Human choice strategy varies with anatomical projections from ventromedial prefrontal cortex to medial striatum
Learning the value of information in an uncertain world
Simultaneous representation of a spectrum of dynamically changing value estimates during decision making
Effects of dopamine on reinforcement learning in Parkinson’s disease depend on motor phenotype
Frontal oscillatory dynamics predict feedback learning and action adjustment
Cortical electrophysiological network dynamics of feedback learning
Frontal theta links prediction errors to behavioral adaptation in reinforcement learning
Cortical delta activity reflects reward prediction error and related behavioral adjustments
The feedback-related negativity signals salience prediction errors
Time-frequency theta and delta measures index separable components of feedback processing in a gambling task
The role of high-frequency oscillatory activity in reward processing and learning
A reinforcement learning mechanism responsible for the valuation of free choice
Striatal activity underlies novelty-based choice in humans
Orthogonalization of regressors in fMRI models
The valuation system: a coordinate-based meta-analysis of BOLD fMRI experiments examining neural correlates of subjective value
Methods for simultaneous EEG-fMRI: an introductory review
Single-trial EEG–fMRI reveals the dynamics of cognitive function
and tonic alertness: a simultaneous electroencephalography/functional magnetic resonance imaging study
Theta and high-beta networks for feedback processing: a simultaneous EEG–fMRI study in healthy male subjects
Beta oscillations and reward processing: coupling oscillatory activity and hemodynamic responses
Basal ganglia activation localized in MEG using a reward task
Reward feedback stimuli elicit high-beta EEG oscillations in human dorsolateral prefrontal cortex
Error-related negativity predicts reinforcement learning and conflict biases
The primate basal ganglia: parallel and integrative networks
Different time courses of learning-related activity in the prefrontal cortex and striatum
Prefrontal cortex as a meta-reinforcement learning system
Orbitofrontal cortex neurons: Role in olfactory and visual association learning
The convergence of information about rewarding and aversive stimuli in single neurons
Dopamine transients are sufficient and necessary for acquisition of model-based associations
An integrated model of action selection: distinct modes of cortical control of striatal decision making
Increases in functional connectivity between prefrontal cortex and striatum during category learning
Action selection and action value in frontal-striatal circuits
Targeted stimulation of human orbitofrontal networks disrupts outcome-guided behavior
Controlling human striatal cognitive function via the frontal cortex
Two spatiotemporally distinct value systems shape reward-based learning in the human brain
Medial prefrontal cortex as an action-outcome predictor
Frontal cortex function as derived from hierarchical predictive coding
Stable and dynamic representations of value in the prefrontal cortex
Causal role of motor preparation during error-driven learning
Human oscillatory activity associated to reward processing in a gambling task
Frontal theta EEG activity correlates negatively with the default mode network in resting state
Trial-by-trial coupling between EEG and BOLD identifies networks related to alpha and theta EEG power increases during working memory maintenance
Separate neural representations of prediction error valence and surprise: evidence from an fMRI meta-analysis
Beta-band oscillations—signalling the status quo
Bursts of beta oscillation differentiate postperformance activity in the striatum and motor cortex of monkeys performing movement tasks
The feedback-related negativity (FRN) revisited: New insights into the localization
On the globality of motor suppression: unexpected events and their influence on behavior and cognition
Polarity of uncertainty representation during exploration and exploitation in ventromedial prefrontal cortex
Neural mechanisms resolving exploitation-exploration dilemmas in the medial prefrontal cortex
Temporally dissociable contributions of human medial prefrontal subregions to reward-guided learning
Event-related potential activity in the basal ganglia differentiates rewards from nonrewards: temporospatial principal components analysis and source localization of the feedback negativity
Event-related potential activity in the basal ganglia differentiates rewards from nonrewards: temporospatial principal components analysis and source localization of the feedback negativity: commentary
Striatal microstimulation induces persistent and repetitive negative decision-making predicted by striatal beta-band oscillation
Striatal beta oscillation and neuronal activity in the primate caudate nucleus differentially represent valence and arousal under approach-avoidance conflict
focally modulated β-band oscillations characterize local field potential activity in the striatum of awake behaving monkeys
Neuroelectric signatures of reward learning and decision-making in the human nucleus accumbens
Medial prefrontal cell activity signaling prediction errors of action values
Temporal filtering of reward signals in the dorsal anterior cingulate cortex during a mixed-strategy game
Posterior cingulate cortex mediates outcome-contingent allocation of behavior
Dissociable dopamine dynamics for learning and motivation
Wave-like dopamine dynamics as a mechanism for spatiotemporal credit assignment
Computational noise in reward-guided learning drives behavioral variability in volatile environments
Frontopolar cortex and the evidence in favor of alternative courses of action
The macaque anterior cingulate cortex translates counterfactual choice value into actual behavioral change
Frontal theta overrides Pavlovian learning biases
Random effects structure for confirmatory hypothesis testing: Keep it maximal
A theory of Pavlovian conditioning: variations in the effectiveness of reinforcement and nonreinforcement
in Classical Conditioning II: Current Research and Theory (eds
Hierarchical Bayesian inference for concurrent model fitting and comparison for group studies
Bayesian model selection for group studies
Active head motion reduction in magnetic resonance imaging using tactile feedback
ICA-AROMA: a robust ICA-based strategy for removing motion artifacts from fMRI data
Is model fitting necessary for model-based fMRI
A method for removing imaging artifact from continuous EEG recorded during functional MRI
FieldTrip: open source software for advanced analysis of MEG
Trial-type dependent frames of reference for value comparison
Algermissen, J., Swart, J. C., Scheeringa, R., Cools, R. &amp; den Ouden, H. E. M. Prefrontal signals precede striatal signals for biased credit assignment in motivational learning biases. johalgermissen/Algermissen2024NatComms: v1.0.0. (2024) <a href="https://doi.org/10.5281/zenodo.10352241" data-track="click_references" data-track-action="external reference" data-track-value="external reference" data-track-label="10.5281/zenodo.10352241">https://doi.org/10.5281/zenodo.10352241</a>
<a data-track="click" data-track-action="download citation references" data-track-label="link" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41467-023-44632-x?format=refman&amp;flavour=references">Download references</a>
We thank Emma van Dijk for assistance with data collection
and the weekly Donders M/EEG meeting for discussions of these results and many helpful suggestions
was funded by Netherlands Organization for Scientific Research (NWO) research talent grant 406-14-028
was funded by Netherlands Organization for Scientific Research (NWO) VENI grant 451-12-021
was funded by Netherlands Organization for Scientific Research (NWO) VICI grant 453-14-005
Netherlands Organization for Scientific Research (NWO) Ammodo KNAW Award 2017
McDonnell Foundation James McDonnell Scholar Award
HEMDO was funded by Netherlands Organization for Scientific Research (NWO) VIDI grant 452-17-016
Hahn Institute for Magnetic Resonance Imaging
The authors declare no competing interests
reviewer(s) for their contribution to the peer review of this work
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations
<a data-test="citation-link" data-track="click" data-track-action="download article citation" data-track-label="link" data-track-external="" rel="nofollow" href="https://citation-needed.springer.com/v2/references/10.1038/s41467-023-44632-x?format=refman&amp;flavour=citation">Download citation</a>
DOI: https://doi.org/10.1038/s41467-023-44632-x
Anyone you share the following link with will be able to read this content:
a shareable link is not currently available for this article
Sign up for the Nature Briefing newsletter — what matters in science
National Collegiate Equestrian Association Academic Honor Roll
Primarily a 3-day Eventer from Albuquerque
Algermissen has also been&nbsp;competitive&nbsp;in Jumpers
In Eventing she has won the high point pony club award at Abbe Ranch
and has been very competitive at the training level
In show jumping she has won&nbsp;championships&nbsp;at the 3'3'' and 3'6'' level
As a member of the United States Pony Club for the past ten years Algermissen has volunteered many hours to maintain an Equestrian Cross Country Course on Open Space
helping her mother with breeding and veterinary work
Algermissen graduated with honors and a 3.8 GPA
She was also a member of the Spanish Honor Society
Algermissen is the daughter of Holly Dietz and Anthony Algermissen
Copyright &copy; 2025 New Mexico State University Athletics / All Rights Reserved
Thanks for visiting <a class="bh0gi3mgo-modal__text-host"></a>
The use of software that blocks ads hinders our ability to serve you the content you came here to enjoy
We ask that you consider turning off your ad blocker so we can deliver you the best experience possible while you are here
known to her loved ones as a vibrant and resilient soul
Margaret was a light in the lives of all who knew her
She will be deeply missed and forever cherished
Margaret graduated from High School in Texas
where she not only excelled professionally but also formed lifelong friendships
Margaret and Trevor left Florida and moved to Utah
They embarked on a grand adventure on an epic road trip across the United States
visiting every state and immersing herself in the wonders of this beautiful nation
Her spirit for exploration were truly inspiring
she considered herself a glamper while enjoying camping with her husband
Camping trips filled with laughter and love were among her fondest memories
Margaret&rsquo;s journey was marked by both resilience and compassion
Margaret&rsquo;s warm-hearted nature and genuine care for others made her a beacon of hope
Margaret treasured her role as a wife and mother
Becoming a mother filled her heart with joy and excitement
The love she radiated from within touched all those around her
Her kind nature knew no bounds as she dedicated herself wholeheartedly to nurturing her son Mason and creating a loving home for her husband Trevor
Margaret faced numerous challenges throughout her life but emerged stronger each time
She loved and the relationship she had found with him
and we know that she was welcomed into His loving arms
From overcoming personal obstacles to carving out a better future for herself and her family
Margaret&rsquo;s accomplishments were a testament to her unwavering determination and indomitable spirit
Her journey inspired everyone who crossed her path
She enjoyed shopping and watching Movies and TV
She loved her daily walks with the dogs and going to coffee with her friends
She is survived by her devoted loving husband Trevor Cook
Margaret also leaves behind her siblings: Ellen Carlos
She was a beloved daughter-in-law to Tammy and Shad Sill and a treasured member of the Cook family
she is survived by numerous brothers- and sisters-in-law
Margaret Rainey Cook&rsquo;s life was cut short too soon
and love will continue to shine brightly in the hearts of those who had the privilege of knowing her
May we find solace in carrying forward Margaret&rsquo;s spirit of strength
and joy as we remember and honor her remarkable life
The road ahead will be long and difficult for our family as we are working to heal from the unexpected loss of Margaret
and we appreciate all the prayers we have received in her honor
In lieu of flowers donations may be made to our family to help with the continual financial needs of raising a baby and an education fund for Mason through Trevor&rsquo;s venmo
@trevor-cook-99 &nbsp;We would like to thank Trevor&rsquo;s boss Parker with Non-Typical Excavation for their kindness and generosity during this difficult time
Services will be livestreamed and available the day of services on the following link: <a href="https://www.facebook.com/RoyChristianChurch">https://www.facebook.com/RoyChristianChurch</a>
<a href="https://lindquistorders.square.site" target="_blank" rel="noopener">ORDER VIDEOS</a>
Terri Kamen now wishes she had listened more closely to her 10-year-old cocker spaniel Samantha
It might have saved the Wellington woman a lot of grief
When Kamen and her husband Michael Schulner went to San Diego in July
they hired Barbara Ann Algermissen to house-sit and pet-sit while they were gone
Little did they know that Algermissen would steal $26,000 worth of jewelry
Algermissen came highly recommended through a professional acquaintance
An in-home interview with Algermissen had gone well and raised no red flags
When Algermissen came to Kamen’s home for the interview
The pooch snapped at Algermissen and appeared “terrified,” Kamen said
“She’s never snapped at a female before,” said Kamen
Kamen and her husband hired her to house-sit again in August and September
Kamen was so happy with Algermissen that she brought her gifts and paid her more than required
Algermissen was repaying Kamen’s gratitude by stealing from her
remained in the Palm Beach County Jail late Thursday where she’s been held since her Sept
Algermissen’s first night of house-sitting was on July 18
Algermissen was already pawning their property
according to the Palm Beach County Sheriff’s Office
Kamen said the total value is more than $26,000
when a special event caused Kamen to wear her best jewelry
“I feel like a schmuck to have been so trusting,” said Kamen
The same acquaintance who had referred the 48-year-old Algermissen to Kamen had done the same for another Wellington couple
The mutual acquaintance was not a party to Algermissen’s alleged thefts and immediately contacted the Harrises after finding out that Kamen and her husband had been victimized
Algermissen began house-sitting for the Harrises on Sept
Algermissen was pawning the couple’s jewelry
Algermissen stole more than $19,000 worth of property from the Harrises
Algermissen received around $3,300 from pawnbrokers for the stolen property
according to the report.She is accused larceny between $20,000 and $100,000; larceny between $10,000 and $20,000; eight counts of dealing in stolen property and eight counts of fraud by false statements
Algermissen is being held in lieu of $28,500 bail
“My faith in humanity has really waned,” Patricia Harris said
Both Kamen and Patricia Harris said they’re kicking themselves for not looking into Algermissen further
A background check would have revealed that Algermissen was arrested in August 2011 for allegedly stealing jewelry from a co-worker
Would Kamen have hired Algermissen knowing she had a criminal history
spokesman for the Palm Beach Police Department
said residents in his city are urged to run background review on house sitters or ask potential house sitters to pull a city identification card
which includes a check for warrants on an individual
Hess said running a background check costs around $15
if you are going to hand over the keys of your house to somebody
a little legwork can go a long way to keeping your home and yourself safe,” Hess said
Trending<a href="/la/">Los Angeles</a>AReunited: Stephen Algermissen leaves Colliers
The broker shakeup continues in Los Angeles — this time
Stephen Algermissen left Cushman &amp; Wakefield in 2011 after 22 years with the brokerage
and became an executive vice president at Colliers International
a leading broker of urban retail properties on the West Coast
will operate out of Cushman &amp; Wakefield’s offices in West L.A
He told TRD he made the move back to Cushman &amp; Wakefield
because his business is more aligned with the company
Cushman &amp; Wakefield is more of an urban firm — they prioritize it the most
It comes from being a firm that has its origin in New York.”
SIGN UPWhile he had no negative words to say about Colliers
he conceded the brokerages were quite different
describing Cushman as “a very New York firm of order and method and consistency,” tapped for many luxe dealings
“The kind of business I was doing at Colliers
but it was an anomaly — I felt like I was better aligned here,” he said
Algermissen said Cushman had gone through some changes he didn’t like
He said he thinks the recent mergers have been good for the company
“I was intrigued by having more of a mandate at Colliers at the time I left,” he said
“You don’t know how good things are until you leave them sometimes
I maintained my friendships [at Cushman] but saw that I could do more if I worked with them
While many of Algermissen’s deals have been in Los Angeles — the $60 million sale of the Santa Monica Apple store and the recent sale of the Yamashiro site in Hollywood among them — his territory ranges to San Diego
Algermissen has also been involved with the sale of office properties
the CNN Building (multiple times) and Kodak Campus in Hollywood
the William Morris Agency headquarters in Beverly Hills and the landmark 100 Wilshire office building in Santa Monica
This website is using a security service to protect itself from online attacks
The action you just performed triggered the security solution
There are several actions that could trigger this block including submitting a certain word or phrase
You can email the site owner to let them know you were blocked
Please include what you were doing when this page came up and the Cloudflare Ray ID found at the bottom of this page
you don’t even know!” (Laughs) In America people talk about it
People take off work because they have to go to therapy
I thought this was crazy but I have to say
And I think it’s actually made me a better actor
this is that… Or if you use something from your personal life for work
make better categories of where to put things
Do you need to partition things in your life
I just don’t want to bring work home anymore
“Let’s go out and party!” I’m in love with my partner not with my co-star
When you were growing up there was so much said about the method and DeNiro and that style of acting
Oh, of course! My first movie, I was trying to hide from the camera! I think it’s human to not want to expose yourself. If you have to cry, you don’t want everybody to look at you, you know? Obviously you know you’re playing for the camera, but it’s like you instinctively hold your hands in front of your face or something. But you learn through other actors. My first movie was with Dennis Hopper and he literally was like, “You’re the greenest person I’ve ever seen.” And he taught me everything.
The basics, you know? Like, don’t hide from the camera! If the camera can’t see you, it doesn’t really work does it?
Do you think that will be different for the next generation of actors who have grown up with smartphones and social media? They are constantly performing for the camera in a way.
Maybe. It’s hard to tell but I think so. Most people meet on Tinder, right? I was at a premiere recently and an actor that was in the movie brought his girlfriend. I was a little drunk and I was like, “Oh, how’d you two meet?” And they were like, “On Tinder!” And I was like, “What?!” I thought you only have sex on Tinder. I didn’t know you actually date. I’m so behind…
Most people meet on the internet these days.
Good for them! It’s so much easier, right?
I feel like I was cheated to not have that simplicity in dating.
Me too! I was like, “I’m never going to meet anyone!” You had to have friends who introduced you…
Unfortunately a lot of people don't wait for that and just hit on people in uncreative ways.
The worst kind is when I’m on location and I’m at the bar after work or whatever, and you know that a guy definitely knows who you are but he pretends he doesn’t know who you are. And he just comes over and he’s so nervous to come and talk to you, and he’s like, “So, what are you in town for?” And you’re like, “Well, you know, I’m here on work…” “Oh, for a movie – I mean, what kind of work do you do?” (Laughs)
The awkwardness of it. I personally don’t like fake people. Especially in the industry that I work in, there can be a lot of superficiality. I do think you make that choice for yourself. You choose who’s around you.
You generally seem to be making your own choices. Not many Hollywood actresses would use their Instagram to post a headline of someone defecating in the seaweed at a Chinese restaurant.
I’ve always felt free. I don’t let anyone make decisions for me. I just opened this account a year ago, and it’s actually been really interesting to share aspects that I choose to share with the public and see the reactions. Sometimes people are not agreeing with me, sometimes they are. But I think it’s about showing or sharing an aspect of my life that a regular interview or a regular photo spread can’t do.
Do you feel like you can be more bold now, also in the roles you choose as an actress?
Yeah, as I get older and have more experiences in life I’m able to bring more to a role, you know? I am unafraid to ask for the things that I need to be able to perform, too. You learn that you have to speak up for yourself. If I work, it’s because I chose it and it’s a pleasure and I believe in what I’m doing.
How did you get through times in your career where that wasn’t the case?
Do you ever choose roles because you want to comment on world affairs?
Because that’s not my role. In a personal relationship, you know, if we go out for dinner, I could tell you about that. But I think people are smart enough to judge for themselves. I hate movies that are preachy! Who needs to be told what to think?
Name: Diane HeidkrügerDOB: 15 July 1976Place of birth: Algermissen, Lower Saxony, GermanyOccupation: Actress
Please enable JS and disable any ad blocker
A joint venture made up of four entities bought the building that holds the Beverly Hills shops of Italian menswear retailer Brioni and British luxury fashion house Alexander McQueen
paying $96 million for the 11,625-square-foot property at 457-459 N
The price was among the highest in the Beverly Hills market this year
The property was acquired from Lang &amp; Lang Properties by a joint venture that includes Michael Shabani
of Crown Equity; Hakim Holdings; and GWP Real Estate – all based in Beverly Hills – and Chatsworth-based Mazal Enterprises
Cushman &amp; Wakefield brokers Carine Mamann
Kazuko Morgan and Stephen Algermissen represented the buyer in the transaction
The building is 100 percent leased to the two retailers
The property features 50 feet of frontage on Rodeo Drive
according to Cushman &amp; Wakefield research
There have been bigger sales on the street
Bijan’s yellow store on Rodeo Drive sold to LVMH Moet Hennessy Louis Vuitton for $122 million or $19,405 per square foot
<a href="https://www.w3.org/WAI/tutorials/images/decision-tree/" target="_blank">Learn how to describe the purpose of the image (opens in a new tab)</a>
Leave empty if the image is purely decorative
The House of Bijan’s home might soon come tumbling down
The ritzy clothier’s Rodeo Drive site sold in July for $122 million
and sources familiar with the Beverly Hills retail market predict that the new owner
the parent of luxury lifestyle brand Louis Vuitton
will eventually replace the famous yellow storefront with a bigger building
between South Santa Monica Boulevard and Brighton Way
would make sense for a deal that set a California retail record of $19,405 a square foot
And it would allow LVMH Moet Hennessy Louis Vuitton Inc
to customize a snazzy storefront for one of its many luxury brands
“They don’t want someone else’s cute little building
they want their own look,” said Stephen Algermissen
executive director for capital markets at Cushman &amp; Wakefield
“It’s all a part of marketing and branding
… That’s why it’s so incredibly valuable.” 
What the acquisition means for the 40-year-old House of Bijan is uncertain
The company had the property under contract for $108 million when owner Barton C
represented by Holliday Fenoglio Fowler’s Bryan Ley
according to two sources familiar with the deal
But rather than seeking to close the transaction it went to the market and flipped its contract to LVMH
declined to discuss details of the transaction
the retailer’s principal and son of founder Bijan Pakzad
said he could not discuss the transaction  or the status of its lease under a confidentiality agreement
values its spot on the world stage of luxury shopping
“This year we are celebrating our 40th anniversary on Rodeo Drive
and we will continue to maintain a major presence on Rodeo Drive for the indefinite future,” he said via email
“There is no better location in the world than Rodeo Drive.” 
Just what kind of presence remains to be seen
There are only two vacancies at present on Rodeo
senior vice president for retail at Jones Lang LaSalle
The appointment-only store could also seek a second-level space on Rodeo
substantially cheaper than a ground-floor presence
which has a flagship Louis Vuitton store on Rodeo as a tenant
6,287-square-foot site eclipsed the Rodeo record
Chanel had purchased the neighboring store in 2013
the purchase price would equate to land value of nearly $698 million an acre
The new owner can replace the existing structure with a three-story
the maximum allowed under city density regulations
Should LVMH choose to pursue a ground-up project on the Bijan site it would join other chic Rodeo retailers that have built flagships from scratch to showcase bold architecture and fancy amenities
Saint Laurent opened a four-story shop lined with white marble and outfitted with a private dressing room on the third floor in 2014
Burberry the same year constructed a three-story building with a penthouse fitting room
The fashion house is proceeding with plans to build from scratch on its Rodeo sites
which together form a 15,250-square-foot parcel
A 2014 Beverly Hills Planning Commission report said Chanel intended to build a three-story shop with a rooftop deck
A city spokeswoman said company representatives have been meeting with city planners
but have not yet submitted updated designs
An attorney for Chanel declined to comment
LVMH and Chanel are among a growing number of high-end retailers choosing to become their own landlords
the companies can instead funnel money into a solid
“If Chanel were to lose the corner of Brighton and Rodeo at the end of their lease
Chanel would be out of the market for the next five to 10 years,” said Robert Cohen
It’s a purchase that makes much more sense for retailers than investors
executive vice president at Newmark Grubb Knight Frank
“An investor that wants to put brands there and lease it is not going to get a high enough rent to pay the mortgage,” he said
Opportunities to buy along the two-and-a-half block stretch are rare
especially with many landowners holding on to property bought in the 1970s and ’80s
a nonprofit that promotes the street as a shopping destination
“People believe in in the longevity of the street,” he said
By now you've heard that residential developers are flocking to Century City, but the office market has also strengthened significantly. (It's no fun having an apartment if you don't have a job to come home and complain about.) That's why we're excited to host Bisnow's <a href="https://www.bisnow.com/los-angeles/events/neighborhood/The-Future-of-Century-City-197">Future of Century City</a>
starting at 8am tomorrow at the Century Plaza Towers
Among our panelists will be CBRE SVP Stan Gerlach
who tells us Century City was one of the last markets to improve
Century City now has seen tremendous activity over the past few months
Stan handles leasing for The Plaza (1800 and 1840 Century Park East)
Recent deals include Network Disposition Services
which took a nice chunk of creative space at 1800 Century Park East
and e-commerce services provider Conversion Systems
snapped up the Orchard Supply Hardware location in Thousand Oaks (1934 E Avenida de Los Arboles) from a private investment firm in LA for nearly $16.8M
this is the highest price ever paid per SF for a single-tenant
The property consists of 43k SF on five acres
who repped both sides with David Maling and Stephen Algermissen
says among the reasons they were able to achieve this record price was that the single-tenant OSH store is in a supply-constrained market with little to no competition
The nearest hardware and home improvement center is five miles away
60k SF former Mervyn's building at Ridgecrest Town Center from China Lake 700 LLC for $5.8M or $97/SF
says the property provided El Rey with an opportunity to buy into the dominant grocery-anchored center in Ridgecrest with a high yield potential
Ridgecrest Town Center,a 205k SF community center at the corner of N China Lake Blvd and Drummond Ave
was fully leased at the time of the building's sale
bought a brand new 108k SF building in the Nasa/Saturn Business Park in Brea for $15M
The buyer consolidated existing industrial locations and expanded into the new facility (408 Saturn St)
First Team Real Estate's Karl Heim repped EVGA.COM
NAI Capital's Tim Steuernol and Rob Zaharia closed the sale of 1425 W 225th St in Torrance
Stepp Commercial's Robert Stepp closed the sale of a mixed-use property near downtown Long Beach for just over $2.5M or $161/SF
The property (1003 Orange Ave) was built in the 1930s and includes 18 apartment units
six ground-floor retail spaces and a 21-space parking lot
the buyer has plans for a possible rooftop billboard and potential development of the parking lot
NR Real Estate Brokers' Matt Pernice represented the buyer
a 142k SF building in Chino Crossroads Business Park in Pomona
The Saywitz Co's Barry Saywitz repped the tenant
Budge's Bill Budge repped landlord Northwestern Mutual Life Insurance and Financial Service
now that John Moran Auctioneers Inc has leased the 17k SF industrial warehouse for its new HQ
The move means John Moran will no longer have to hold auctions off-site
The facility will consolidate the firm's offices
NAI Capital's Marie Taylor repped John Moran in the 10-year lease
which makes musical and animated plush toys
leased a 36k SF building in the City of Commerce (2839 Tanager Ave)
Heger Industrial’s Stephan Ktorza and Philip Dray repped the tenant
and Colliers' Steve Calhoun and Phil Norton repped the landlord
Instrument &amp; Value Services Co leased a 35k SF building on 1.3 acres in Paramount (6851-6853 Walthall Way)
Heger Industrial’s Jack Whalen along with Patrick Fitzgerald and Curt Stanton with Paine Wetzel Associates repped both sides in the $1.6M deal
Michael Pakravan of Kennedy Wilson's brokerage group represented Jersey Mike’s Subs on three leases throughout Greater LA
They include a 1,200 SF store within a mixed-use multifamily and retail complex in DTLA (801 S Hope St)
where Kennedy Wilson's Lee Shapiro and Justin Weiss repped landlord Wood Partners
The sandwich shop also leased 1,200 SF at 1603 Wilshire Blvd in MacArthur Park from Wilshire Union LP
and 1,700 SF in the Mountaingate Shopping Center at 1197 E Los Angeles Ave in Simi Valley from Mountaingate TT
Fred Leeds Properties' Jordan Wheeler and CBRE's Scott Siegel and Lisa Engel
The United Farm Workers Foundation leased 500 SF at 3002 Whittier Blvd in LA
Todd Nathanson and Dimas Rangel of illi Commercial Real Estate repped both sides
George Smith Partners arranged $30.7M in construction financing for three separate multifamily developments in Koreatown
including a $14.5M construction loan to Elite Real Estate for the development of a 65-unit apartment community at 833 Harvard Blvd (above)
including $1.2M in preferred equity from a crowdsourcing firm
for Index Realty's development of a 40-unit condo project slated to break ground this month at 1101 S Harvard
and $3.2M for Elite to build a 12-unit condo building at 305 S Ardmore Ave
who arranged the loans with assistance from Adam Candler
Hunt Mortgage Group refinanced an eight-property portfolio in the South Bay using Fannie Mae Choice Refinance facilities
The early rate-lock deals include a $39M loan for the 160-unit Crystal Cove Beach Resort in Hermosa Beach; $30M for the 149-unit Copa Pacific Apartments in Torrance; and $21M each for the 100-unit Peppertree Apartments in Hermosa Beach and the 95-unit Beachbrook Village Apartments in Redondo Beach
Matt Olrich and Richard Olrich from Hunt's San Rafael office sourced and processed the loans
Mission Capital Advisors' Debt &amp; Equity Finance Group secured 23 separate loans totaling $309M to refi a diverse portfolio of real estate assets owned by The Seligman Group
The long-term loans are all IO for the entire term
The 23 assets include 11 multifamily properties in LA totaling 817 units
which received an $80M Freddie Mac loan originated by Walker &amp; Dunlop
The portfolio also includes 12 commercial properties (retail
flex and office) in Orange County and San Francisco
The Mission Capital team included Jordan Ray
Jamie Matheny and Eugene Shevaldin in the New York and Newport Beach offices
You are subscribed to the Bisnow Los Angeles Newsletter
or click <a href="#" class="copy-to-clipboard" style="color: #fff; text-decoration: underline;">here</a> to copy link to clipboard
We will email you a link to reset your password
Upcoming regulations in the European Union require us to show this pop-up and ask you to agree to keep using Bisnow.com
We want to take 15 seconds to tell you what's going on:
Legacy Luncheon: PCF President and CEO Jennifer DeVoll with event guest speaker Sherry Dewane
Photo courtesy of Pasadena Community Foundation Pasadena Community Foundation Legacy Luncheon
Photos courtesy of Pasadena Community Foundation
Legacy Luncheon: Jennifer DeVoll and PCF Board Member George Dulgeryan
Photo courtesy of Pasadena Community Foundation
Legacy Luncheon: Lorraine and Lonnie Schield
Legacy Luncheon: Robert Floe and Matthew DeVoll
Legacy Luncheon: Kenji Funahashi and Yasi Ostovar Funahashi
Legacy Luncheon: John and Margaret McAustin
Legacy Luncheon: Matthew DeVoll PCF Board Member Sandra Ell
Legacy Luncheon: PCF Board Member Gloria Pitzer with Ray and Charmayne Ealy
Legacy Luncheon: Jim Gamb  PCF Board Chair Priscilla Gamb
Legacy Luncheon: Judy McDonald Judy Gain Jim Graunke
Legacy Luncheon: Dave Kronberg and Peter and Anne Kennedy
The Pasadena Community Foundation (PCF) recently hosted its annual Legacy Luncheon on Wednesday
to recognize and honor its donors and professional advisors who have contributed significantly to its growth and sustainability over the decades
The event was held in-person and saw over 100 guests in attendance who enjoyed a social gathering
and a program with guest speakers and a video
expressed her joy at the event being held in person this year
as last year&#8217;s luncheon was held virtually due to the pandemic
She noted the excitement and enthusiasm in the room as guests caught up with each other
and the event was filled with many hugs and smiles
PCF is a nonprofit organization that aims to build and sustain a thriving community in the greater Pasadena area and beyond
the foundation awarded $2.2 million to 116 Pasadena-area nonprofit agencies
and this robust grant-making was made possible by the Endowment Builders &amp; Legacy Society Members
These donors create permanent support for causes and organizations they care about in the community
and their contributions are crucial to the foundation&#8217;s current and future grant-making
and $100 million of that total is permanently endowed
providing perpetual support for the foundation&#8217;s grant-making
etc.; Hall Capital; Hahn &amp; Hahn; and IDS Real Estate
and guests enjoyed a presentation by Sherry Dewane
a financial advisor with UBS Financial Services
She leveraged her 25 years of experience in financial services to present &#8220;Fraud
Forgery and Other Funny Business,&#8221; an illuminating talk about protecting one&#8217;s assets from nefarious actors
For more information about Pasadena Community Foundation and its programs visit <a href="http://www.pasadenacf.org" target="_blank" rel="noopener">www.pasadenacf.org</a>
 <a href="https://pasadenanow.com/main/author/clark-kent" class="more-link">More &raquo;</a>
<a href="https://pasadenanow.com/weekendr/main/the-humming-byrd-house-a-mid-century-storybook-ranch-surrounded-by-nature" target="_blank">The &#8216;Humming-Byrd&#8217; House: A Mid-Century Storybook Ranch Surrounded by Nature</a>
Pasadena Now has been published daily since April
2004 and is among the very oldest continuously operated community news websites in the U.S
Pasadena Now strives to publish a full spectrum of news and information articles in service to the entire community
The publication will remain free to readers and will not erect paywalls
Copyright © 2025. <a class="sitettl" href="https://pasadenanow.com/main">Pasadena Now</a> | Terms of - Service | Privacy Policy | CA Notice of Collection
RSS (Really Simple Syndication) feeds offer another way to get Pasadena Now content
Subscribe to our feeds to get the latest headlines
summaries and links back to full articles &#8211; formatted for your favorite feed reader and updated throughout the day
You are using an outdated browser. <a href="https://browsehappy.com/">Upgrade your browser today</a> or <a href="https://www.google.com/chromeframe/?redirect=true">install Google Chrome Frame</a> to better experience this site
<a href="https://richardcrouse.ca/wp-content/uploads/2013/10/Bridget-von-Hammersmark-inglourious-basterds-17086498-1280-1024.jpg"><a href="https://richardcrouse.ca/diane-kruger-i-can-be-really-obnoxious-when-i-want-something-badly-b-y-richard-crouse/"></a></a>Diane Kruger almost didn’t win the part of German movie-star-turned- Mata-Hari Bridget von Hammersmark in Inglourious Basterds because Quentin Tarantino’s didn’t think she was German enough
He was familiar with her from National Treasure
but assumed she was American and would have trouble with the German dialogue and accent
“It&#8217;s a testament to my dialect coaches throughout the years,” says the actress
born Diane Heidkrüger thirty-four years ago in Algermissen
“They eventually convinced him I was actually German,” she says
but unfortunately Tarantino wanted Natasha Kinski for the role
When Kinski backed out Kruger saw her chance
“I can be really obnoxious when I want something badly,” she says
For the audition she flew herself to Berlin and learned thirty pages of dialogue in German and in English
“I knew if I got my chance he couldn’t hire anyone else.”
The Berlin set of Inglourious Basterds was a long way from Algermissen
“I come from the middle of nowhere,” she says
“No one in my family knows anyone in the business.”
Her first taste of “the business” was at the Royal Ballet in London before an injury sidelined her dancing career
Returning to Germany she pursued modeling and became a top model
After starring in campaigns for Chanel and Giorgio Armani it was time for a change
“You can only care so much about free clothes and posing
At the suggestion of The Fifth Element filmmaker Luc Besson she left the runway to try acting
Working internationally—she’s fluent in three languages—she made an impression in a series of French films like Mon idole before Wolfgang Petersen cast her as Helen of Troy opposite Brad Pitt in the big budget epic Troy
Admitting to being “inexperienced and completely overwhelmed” while making the movie
it nonetheless put her “on the map” in Hollywood
Soon she was starring opposite Nicolas Cage in National Treasure and its sequel while still finding time to make challenging films like Frankie and the Oscar nominated Joyeux Noël
Inglourious Basterds is another jewel in her crown
but don’t expect her to return to that kind of role again anytime soon
“Every movie has to be a different challenge,” she says
“I don&#8217;t want to play the same part that I&#8217;ve played in a different movie
I have to be scared of it to want to do it.”
Richard Crouse shares a toast with celebrity guests and entertainment pundits every week on CTV’s exciting talk show <a href="https://www.ctvnews.ca/ctv-news-channel/pop-life" target="_blank" style="color:white;">POP LIFE</a>.Saturdays at 8:30 on CTV NewsChannel
Commercial real estate firm Cushman &amp; Wakefield announced the sale of a prime retail building along prestigious Rodeo Drive in Beverly Hills for $96 million
is home to Italian menswear retailer Brioni and British luxury fashion house Alexander McQueen
The sale represents one of the highest price-per-square-foot – $8,240 per square foot – for Beverly Hills this year
The property was acquired from Lang &amp; Lang Properties by a joint venture
Hakim Holdings and GWP Real Estate – all of which are based in Beverly Hills – and Chatsworth-based Mazal Enterprises
Rodeo Drive building were represented by Gil Dembo
The building is 100 percent leased to Kering brand luxury tenants
where rents along Rodeo Drive can run $875 to $1,000 per square foot per year
The sale is not the highest for retail property in Beverly Hills
6,200-square-foot retail property at 456 N
Rodeo Drive in Beverly Hills for $110 million – or nearly $17,750 per square foot – from Palm Beach
Fla.-based private equity group Sterling Organization
which had acquired the property for $55 million one day earlier
Sterling Organization bought the building from The Karl B
only to make a $55 million profit the next day for itself and its institutional investor partners in its Sterling Value Add Partners II fund
The property is located in the Golden Triangle area
between Santa Monica Boulevard and Brighton Way
There have been bigger sales on this street
Bijan’s yellow store on Rodeo Drive sold to LVMH for $122 million or $19,405 per square foot
Rodeo Drive has the highest retail rents in Beverly Hills and Los Angeles
Finance reporter Pat Maio can be reached at <a href="/cdn-cgi/l/email-protection" class="__cf_email__" data-cfemail="e89885898187a884898a9d9b81868d9b9b82879d9a868984c68b8785">[email&#160;protected]</a> or (323) 556-8329
Diane Kruger is a German actress and former model who has a net worth of $24 million
She has played notable roles in some of Hollywood's biggest blockbusters and has appeared in numerous television series
most of her roles have been English-speaking or French-speaking
She has also been nominated for countless others
Raised in a Roman Catholic household by middle-class parents
Diane studied hard to learn English at a young age
She attended Catholic school as a child and went on exchange programs to improve her English
she wanted to become a ballerina and auditioned for the Royal Ballet School in London
prompting Kruger to pursue a career as a fashion model in Paris
She had been successful in the modeling world ever since 1993 when she represented Germany at the Elite Model Look event
Kruger is somewhat of a rarity in the fashion world because of her relatively short stature
Although she stands at just under 5 foot 6
she has enjoyed a thoroughly successful modeling career with no shortage of jobs
Throughout her modeling career, Kruger worked with major brands such as Chanel, <a href=https://www.celebritynetworth.com/richest-businessmen/richest-designers/marc-jacobs-net-worth/ title="Marc Jacobs Net Worth">Marc Jacobs</a>
She also appeared on the covers of magazines like Marie Claire
By the early 2000s, Diane had shifted her focus towards acting and began taking drama lessons. Her first film role came in 2002 when she appeared in "The Piano Player," and she followed up with roles in roles directed by her then-husband, <a href=https://www.celebritynetworth.com/richest-celebrities/actors/guillaume-canet-net-worth/ title="Guillaume Canet Net Worth">Guillaume Canet</a>
Further film roles came with "Michel Valliant" and "Wicker Park" before one of her first real breakthroughs in the film "Troy." She played Helen of Sparta
one of the key roles in the retelling of the ancient Iliad
The film received mixed reviews but managed to rake in over $500 million at the box office
She went on to star alongside Nicholas Cage in the film "National Treasure" before following up with "Copying Beethoven," "Joyeux Noel," and "National Treasure: Book of Secrets." In 2009
she appeared in the film "Inglourious Basterds," and she was nominated for a Screen Actors' Guild Award for her performance
Nobody." She began the 2010s with a role in "Unknown," playing an illegal immigrant from Bosnia
Diane starred in a French film called "Les Adieux à la Reine." In 2013
she appeared in "The Host" before transitioning to a television role in the series "The Bridge." By 2016
she had returned to film roles once again with "The Infiltrator"." In 2019
it was announced that Diane would appear in the spy movie "355."
Kruger has maintained a close association with the Cannes Film Festival
This began in 2007 when she acted as the hostess for the opening and closing ceremonies
she became a member of the jury for the main competition at the Cannes Film Festival for the first time
She would continue to appear on this jury for a number of subsequent years
the most notable moment for Kruger came in 2017
when she won the Cannes Film Festival Award for Best Actress
This was in recognition of her performance in the film "In the Fade," which was also the first time she had ever acted in a German-language film (despite that being her native tongue)
She also won a Satellite Award for Best Actress in recognition of her role in "In the Fade." In addition
"In the Fade" won Best Foreign Film at the Golden Globe Awards
Diane Kruger supplements her income from acting with a number of notable brand endorsements. Since 2009, she has been the global spokesmodel for L'Oreal. She is also a brand ambassador for Jaeger-LeCoultre, a Swiss watch brand. In addition, she endorsed a fragrance line for <a href=https://www.celebritynetworth.com/richest-businessmen/richest-designers/calvin-klein-net-worth/ title="Calvin Klein Net Worth">Calvin Klein</a> in 2010
it was reported that Diane Kruger and Norman Reedus had purchased a property in the West Village neighborhood of Manhattan for $11.75 million
The off-market real estate deal was done through an LLC
The home spans an impressive 3,800 square feet and features construction dating all the way back to the 1800s
Federal-style architecture also makes this property stand out
the couple dropped $8.5 million on a property in the Hollywood Hills neighborhood of Los Angeles
and Kruger and Reedus managed to snatch it up for much lower than its original listed price of $13 million
The property has gone through some major renovations by top-class design professionals
although it was originally constructed in the 1920s
It also boasts over 7,700 square feet of living space and sits on half an acre of land
&copy; 2025 Celebrity Net Worth / All Rights Reserved