The Journal of Neuroscience, January 28, 2015 • 35(4):1763–1772 • 1763

Behavioral/Cognitive

Recollection-Related Increases in Functional Connectivity

Predict Individual Differences in Memory Accuracy
X Danielle R. King, Marianne de Chastelaine, Rachael L. Elward, Tracy H. Wang, and Michael D. Rugg
Center for Vital Longevity and School of Behavioral and Brain Sciences, University of Texas at Dallas, Dallas, Texas 75235

Recollection involves retrieving specific contextual details about a prior event. Functional neuroimaging studies have identified several
brain regions that are consistently more active during successful versus failed recollection—the “core recollection network.” In the
present study, we investigated whether these regions demonstrate recollection-related increases not only in activity but also in functional
connectivity in healthy human adults. We used fMRI to compare time-series correlations during successful versus unsuccessful recol-
lection in three separate experiments, each using a different operational definition of recollection. Across experiments, a broadly dis-
tributed set of regions consistently exhibited recollection-related increases in connectivity with different members of the core
recollection network. Regions that demonstrated this effect included both recollection-sensitive regions and areas where activity did not
vary as a function of recollection success. In addition, in all three experiments the magnitude of connectivity increases correlated across
individuals with recollection accuracy in areas diffusely distributed throughout the brain. These findings suggest that enhanced func-
tional interactions between distributed brain regions are a signature of successful recollection. In addition, these findings demonstrate
that examining dynamic modulations in functional connectivity during episodic retrieval will likely provide valuable insight into neural
mechanisms underlying individual differences in memory performance.
Key words: angular gyrus; episodic memory; fMRI; functional connectivity; hippocampus; recollection

Introduction courses of activity in different regions covary, the regions are

Recollection refers to the processes that support retrieval of considered functionally connected (Friston et al., 1993). Another
qualitative information about a prior event (Mandler, 1980; approach involves measuring variations in connectivity associ-
Yonelinas, 2002). Recollection can be contrasted with famil- ated with cognitive states or processes that fluctuate trial-by-trial
iarity; recognition without retrieval of qualitative details. (Friston et al., 1997; Rissman et al., 2004). This event-related
Findings from functional neuroimaging studies have demon- approach is useful when studying memory using recognition
strated that a set of brain regions are more active during rec- memory tests, when successful retrieval occurs on an unpredict-
ollection than familiarity-based recognition (Spaniol et al., 2009; able subset of trials. For such tests, an approach that permits
Kim, 2010; Rugg and Vilberg, 2013). Regions belonging to this connectivity to be contrasted on a trial-wise basis enables the
“core recollection network” include left angular gyrus, posterior direct comparison of connectivity during successful and unsuc-
cingulate/retrosplenial cortex, medial prefrontal cortex, hip- cessful retrieval.
pocampus, and parahippocampal cortex. Assessing dynamic modulation of connectivity during epi-
Although this “modular” approach to characterizing fMRI sodic retrieval tests might provide insight into the neural mech-
findings has advanced our understanding of the neural substrates anisms supporting recollection (Ranganath et al., 2005; Staresina
of recollection, the benefits of adopting a complementary ap- et al., 2013). Although the brain regions engaged during success-
proach have become increasingly evident. This approach involves ful recollection have been identified, recollection-sensitive inter-
investigating interactions between different brain regions. One actions between members of this putative network, or with other
method of assessing “functional connectivity” involves analyzing brain regions, are not well characterized. Additionally, investiga-
time-dependent fluctuations in blood oxygen level-dependent tion of individual differences in recollection-related modulation
(BOLD) signal time courses acquired over a sustained period of connectivity might provide insight into aspects of brain func-
(Greicius et al., 2003; Fox et al., 2005; Wig et al., 2011). If the time tion that support accurate memory. In one previous study, task-
related changes in connectivity between several large-scale brain
networks were assessed during episodic retrieval relative to a con-
Received Aug. 4, 2014; revised Dec. 1, 2014; accepted Dec. 5, 2014.
Author contributions: M.d.C., R.L.E., T.H.W., and M.D.R. designed research; M.d.C., R.L.E., T.H.W., and M.D.R. trol task (Fornito et al., 2012). Similarity between two of the
performed research; D.R.K., M.d.C., R.L.E., T.H.W., and M.D.R. analyzed data; D.R.K. and M.D.R. wrote the paper. networks in their task-related connectivity changes with the rest
This research was supported by NIMH Grant 5R01MH072966 and NIA Grant 1R01AG039103. of the brain correlated across individuals with the response laten-
The authors declare no competing financial interests. cies, but not the accuracy, of recollection judgments. In another
Correspondence should be addressed to Dr Danielle R. King, 1600 Viceroy Drive, Suite 800, Dallas, TX 75235.
E-mail: [email protected].
study (Schedlbauer et al., 2014), connectivity between a set of
DOI:10.1523/JNEUROSCI.3219-14.2015 regions was reported to be denser for successful than unsuccess-
Copyright © 2015 the authors 0270-6474/15/351763-10$15.00/0 ful contextual judgments. Additionally, connection density dur-
1764 • J. Neurosci., January 28, 2015 • 35(4):1763–1772 King et al. • Recollection-Related Increases in Connectivity

ing successful retrieval was correlated across subjects with (a Deutschmark). The requirement was to make an “indoor/outdoor”
retrieval accuracy. However, it remains to be demonstrated judgment about each object. At test, each studied picture was re-
whether changes in connectivity associated with successful rela- presented along with the 80 unstudied pictures. The task was to judge
tive to unsuccessful recollection covary with individual differ- whether each picture had been studied in association with a Lira or a
Deutschmark or whether it was unstudied; correct source judgments
ences in memory accuracy.
were differentially rewarded according to the identity of the coin. Here,
Here, we applied psychophysiological interactions (PPI) analysis
test items were collapsed across the two study contexts (sources) and
(Friston et al., 1997) to fMRI data acquired in three independent recollection-related activity was operationalized by the contrast between
studies, each using a different operationalization of recollection. We fMRI responses to studied items attracting correct versus incorrect
investigated whether core recollection regions exhibited task-related source judgments.
changes in functional connectivity with other brain regions during MRI data acquisition and preprocessing. In each experiment, MRI data
successful compared with unsuccessful recollection, and whether the were acquired on the same 3T Philips Achieva MRI scanner (Philips
magnitude of these changes covaried across individuals with recol- Medical Systems) equipped with a 32 channel receiver head coil. Func-
lection accuracy. tional images were acquired using a T2*-weighted, BOLD echoplanar
(EPI) sequence (SENSE factor 1.5, flip angle 70°, 80 ⫻ 80 matrix, FOV ⫽
24 cm, TR ⫽ 2000 ms, TE ⫽ 30 ms) and T1-weighted anatomical images
Materials and Methods were acquired with a magnetization-prepared rapid gradient echo pulse
Participants. Experiment 1 included 24 participants (13 female) age sequence (FOV ⫽ 240 ⫻ 240, 1 mm 3 isotropic voxels).
19 –29 years (M ⫽ 23.79; SD ⫽ 2.43). Experiment 2 included 36 partic- MRI data were preprocessed in SPM8 (Wellcome Department of Cog-
ipants (17 female) age 18 –29 years (M ⫽ 22.2; SD ⫽ 3.0), and Experi- nitive Neurology, London, UK). Functional scans were realigned to the
ment 3 included data from 28 participants (14 female), age 18 –29 years mean EPI image, subjected to slice timing correction, reoriented to ap-
(M ⫽ 23.14; SD ⫽ 3.34). Participants were recruited from the University proximate the MNI reference template, spatially normalized to MNI
of Texas at Dallas (UT Dallas) and surrounding communities. All were space, and smoothed using an 8 mm full-width half-maximum Gaussian
right-handed, had normal or corrected-to-normal vision, and had kernel. Anatomical images were similarly normalized to MNI space.
learned English before age five. They were compensated with $30/h for fMRI data analysis: average BOLD signal changes associated with suc-
their participation. All participants gave informed consent according to cessful recollection. Statistical Parametric Mapping (as implemented in
the procedures approved by the UT Dallas and University of Texas SPM8) based on a General Linear Model (GLM) was used to analyze the
Southwestern Institutional Review Boards. fMRI data. At the individual subject level, a number of events of interest
Materials and procedures. Recollection was assessed using a different (the exact number depending on the experiment) were modeled; these
behavioral paradigm in each of the three experiments. Experiment 1 used always included one event corresponding to successful recollection and
a Remember/Know procedure (Tulving, 1985; Gardiner, 1988). Experi- one event for unsuccessful recollection. The neural response elicited on
mental items consisted of 216 pictures and 216 corresponding object each trial was modeled as a delta function that corresponded to the onset
names. Of the 216 picture–word pairs, 144 were randomly selected to be of each trial. Delta functions were then convolved with a canonical he-
presented during the study phase. For 72 of the pairs, the picture was modynamic response function to model the predicted BOLD response.
presented, whereas for the remainder the name was presented. At test, the Other covariates entered into the first-level models included six param-
144 studied items and an additional 72 new items were presented as eters that represented the motion-related variance in the data (3 for
words only. During encoding, subjects made one of two judgments about rigid-body translation and 3 for rotation), as well as regressors represent-
the denoted object depending on whether it was presented as a picture or ing each of the separate scan sessions. An autoregressive AR(1) model
a word. The test task was to judge whether each test word corresponded was used during parameter estimation to correct for time-series correla-
to a studied item, regardless of its study format. There were three re- tions in the data. Data were high-pass filtered with a cutoff of 128 s to
sponse options. Subjects were instructed to respond “Remember” on remove any confounds due to slow signal drifts. Individual subjects’
trials where recognition was accompanied by retrieval of a specific detail parameter estimates for successful and unsuccessful recollection events
or details from the study episode. “Know” responses were to be used from these first-level analyses were entered into a group-level analysis
when an item was recognized in the absence of the retrieval of any specific that treated subjects as a random variable. Contrast images comparing
detail about the study event. “New” judgments were to be given if a test activity associated with recollection to activity associated with familiarity
item was not recognized from the study phase. For the present purposes, were constructed. These images were thresholded at p ⬍ 0.001, uncorrected,
behavioral and fMRI data were analyzed after collapsing over the format with a 23-voxel extent threshold (giving a cluster-wise corrected threshold of
of the studied items. Recollection success was operationalized by the p ⬍ 0.05 according to the Monte-Carlo simulation implemented in AFNI;
contrast between the fMRI BOLD activity elicited during retrieval by http://afni.nimh.nih.gov/pub/dist/doc/manual/AlphaSim.pdf).
studied items endorsed Remember and studied items endorsed Know. To compare results across experiments, first-level contrast images
Experiment 2 used an associative memory procedure. During the from each experiment were entered into a single second-level analysis,
study phase, subjects studied 240 visually presented pairs of concrete with experiment as a between-subjects factor. To identify regions that
words, judging which of the two denoted objects was more likely to fit showed reliable recollection success effects across experiments, the main
into the other. At test, 160 of the studied word pairs were re-presented effect of recollection success was inclusively masked by the simple effect
(intact pairs). A further 80 test pairs comprised studied words that had from each experiment ( p ⬍ 0.01) to ensure that only voxels where a
been repaired from study (rearranged pairs). There were also 80 new recollection effect was evident in all three experiments were included. To
pairs, consisting of unstudied words. The retrieval task was to judge identify differences in recollection-related activity across experiments,
whether each test pair was “Intact” (studied together), “Rearranged” we examined the main effect of experiment, thresholded at p ⬍ 0.001
(both words were studied but on different trials), or New (neither word with a 23-voxel extent threshold.
was studied). For the purpose of fMRI analyses, successful recollection fMRI data analysis: recollection-related changes in BOLD time-series
was operationalized as the contrast between activity elicited by intact test correlations. PPI was used to identify recollection-related changes in con-
pairs correctly judged as such, and that elicited by intact pairs wrongly nectivity (Friston et al., 1997). PPI identifies event-related differences in
judged as rearranged. the time-series correlations (functional connectivity) between a “seed”
Experiment 3 involved a source memory task (data from this experi- region and the rest of the brain after partialling out the main effect of
ment, albeit subjected to quite different analyses from those described event-related changes in activity and event-independent connectivity.
here, have been published previously by Elward et al., 2014). Experimen- Five PPI analyses, one for each seed region of interest, were conducted for
tal items consisted of 240 color pictures of objects. During the study each experiment. Seed regions were selected based on the results of the
phase, 160 of the objects were presented, 80 in association with the de- univariate analyses described above and previous research implicating
piction of one type of coin (a Lira), and 80 with the depiction of another these regions in successful recollection (Spaniol et al., 2009; Kim, 2010;
King et al. • Recollection-Related Increases in Connectivity J. Neurosci., January 28, 2015 • 35(4):1763–1772 • 1765

Rugg and Vilberg, 2013). Seed regions included left angular gyrus (AnG), bled both the behavioral measures and the parameter estimates for each
medial prefrontal cortex (mPFC), hippocampus (Hipp), middle tempo- seed–target pair and then constructed correlation matrices in the same
ral gyrus (MTG), and posterior cingulate cortex (PCC). To create the PPI way that was done with the actual data. We repeated this process 1000
interaction term of interest, we created a task vector that coded successful times per seed region for each experiment, constructing two separate null
recollection trials as 1, unsuccessful recollection trials as ⫺1, and all other distributions for each seed region/experiment. The first null distribution
trials as 0. We then extracted a time course from each seed region in a reflected the average of the correlations between a given seed region and
subject-specific manner. We did this by creating functionally defined all target regions that should be expected by chance. From this null dis-
masks for each seed region, based on the results of the group-level uni- tribution, we calculated the chance probability of obtaining the observed
variate recollection success contrast. Masks comprised 10-mm-radius average correlation value for each seed region by computing the propor-
spheres centered on the peak of the group-level recollection effect within tion of iterations within the null distribution that exceeded the obtained
each seed region. The hippocampal mask was additionally constrained by value. The second null distribution was with respect to the proportion of
an anatomical mask of the hippocampus, which was defined according to significant correlations for each seed region that would be expected by
the AAL atlas (Tzourio-Mazoyer et al., 2002). The anatomical constraint chance. We then calculated the number of iterations within the null
on the hippocampal mask ensured that voxels that were within a 10 mm distribution for which the total number of significant r values exceeded
radius sphere of the peak but were outside of the hippocampus were the observed number of significant values.
excluded from analysis, which was not a concern for any of the other seed
regions. After these masks were created, we determined the peak recol- Results
lection effect within these masks for each subject. We drew 3-mm-radius Behavioral results
spherical regions-of-interest (ROIs) centered on these peaks, and ex- Different behavioral indices were used to estimate the probability
tracted the representative time course (first eigenvariate) of the included of recollection ( pR) in each experiment. In Experiment 1, pR was
voxels. The time courses were then deconvolved with the canonical he- defined as the difference between the proportions of studied and
modynamic response function (HRF) in SPM8, multiplied by the event new items given a Remember response (Jacoby, 1991). The mean
structure vector, and then reconvolved to provide the PPI term that was score was 0.59 (SD ⫽ 0.11), significantly above the chance value
entered into the first-level regression model as the regressor of interest. of zero (t(23) ⫽ 25.76, p ⬍ 0.001). In Experiment 2, pR was mea-
The event structure vector was convolved with the HRF and entered into sured as the difference between the proportion of intact pairs
the regression model as the psychological (task) regressor of no interest,
correctly judged intact and the proportion of rearranged pairs
along with the original time course from the seed region of interest.
Additional nuisance covariates included the same motion and session wrongly judged as intact. Mean pR was 0.48 (SD ⫽ 0.19), again
parameters that were also entered into the mean signal analyses, and data significantly greater than chance (t(35) ⫽ 15.09, p ⬍ 0.001). pR in
were again subjected to a high-pass filter with a cutoff of 128 s. Experiment 3 was estimated as the proportion of studied items
Parameter estimates of the PPI regressor from the first-level PPI GLM attributed to the correct study context relative to all studied items
analyses were brought to the second level where separate group-level, recognized as old. Mean source accuracy (i.e., pR) across subjects
random effects analyses were performed for each seed in each experi- was 0.65 (SD ⫽ 0.11), also significantly greater than chance (t(27) ⫽
ment. One-sample t tests identified voxels where connectivity with a 7.01, p ⬍ 0.001).
given seed was higher during successful versus unsuccessful recollection.
The resulting contrast images from each of these analyses were thresh- Average BOLD signal changes associated with
olded at p ⬍ 0.001, uncorrected, with a voxel extent threshold of 23 (see successful recollection
previous section). To identify brain regions where recollection-related To identify regions demonstrating reliably greater neural activity
increases in connectivity were reliable across experiments, first-level con- (as indexed by fMRI BOLD signal) for trials associated with suc-
trast images from each experiment were entered into a single, second-
cessful relative to unsuccessful recollection we examined the
level model, with experiment as a between-subjects factor. The main
effect across experiments was thresholded at p ⬍ 0.001 (23 contiguous main effect of successful recollection, collapsing across experi-
voxels) and inclusively masked by the simple effect of each experiment ments (thresholded at p ⬍ 0.001, uncorrected, voxel extent ⫽
(inclusive mask threshold of p ⬍ 0.01; see above). 23), masked by the simple effect of successful recollection from
Behavioral correlations with recollection-related changes in connectivity. each experiment ( p ⬍ 0.01). We first identified regions demon-
To determine the extent to which recollection-related increases in con- strating greater neural activity for trials associated with successful
nectivity were related to recollection accuracy, we calculated the correla- relative to unsuccessful recollection in each experiment (see Ma-
tion between recollection accuracy ( pR) and connectivity change terials and Methods). Whole-brain results of the masked main
between each of the five seed regions and 90 target regions that were effect are presented in Figure 1. The masked main effect included
defined based on the AAL atlas. We used the AAL atlas (Tzourio- a set of regions that correspond closely with those held to consti-
Mazoyer et al., 2002) to parcellate each subject’s brain into 90 cortical,
tute the core recollection network (see Introduction). The re-
striatal, and thalamic regions in a manner unbiased by the outcomes of
the fMRI analyses. Using the MarsBar toolbox in SPM8 (Brett et al.,
gions included left AnG, bilateral Hipp/parahippocamal cortex,
2002), we created mask images corresponding to each region. Then, for mPFC, and PCC. In addition, recollection effects were evident in
each seed region, we extracted the mean parameter estimate of connec- left MTG, along with right ventrolateral prefrontal cortex
tivity change across all voxels within a given ROI, for each subject. We (VLPFC), right supplementary motor area (SMA), the striatum,
constructed correlation matrices where each column represented a dif- and the cerebellum.
ferent seed and each row represented one of the 90 “target” regions We used the main effect of experiment from the foregoing
defined according to the AAL atlas. Using the parameter estimates that ANOVA model to identify regions where recollection effects dif-
were extracted for each subject, we calculated for each seed–target pair a fered according to experiment (see Materials and Methods). Dif-
between-subjects Pearson’s correlation coefficient that reflected the ferences were limited to a few small clusters, including ones in
strength and direction of the relationship between recollection-related right VLPFC, bilateral dorsolateral prefrontal cortex (DLPFC),
changes in connectivity and recollection accuracy. In each experiment,
left middle occipital gyrus, and right angular gyrus. In each
for every seed we calculated both the average size of the correlations
across all target regions, and the proportion of statistically significant case, the effect was driven by greater recollection-related ac-
correlations ( p ⬍ 0.05, two-tailed). tivity in Experiments 1 and 3 compared with Experiment 2.
Permutation analyses tested whether, collapsed across all target re- Across all experiments and seeds, no regions were identified
gions, the average size and number of these correlations were greater than where connectivity was greater during unsuccessful relative to
would be expected by chance. For each experiment, we randomly scram- successful recollection.
1766 • J. Neurosci., January 28, 2015 • 35(4):1763–1772 King et al. • Recollection-Related Increases in Connectivity

Figure 1. RegionswherethemagnitudeoftheBOLDsignalwassignificantlygreaterduringsuccessfulrelativetounsuccessfulrecollectionacrossallthreeexperiments.Themaineffectofsuccessfulretrieval
activityacrossexperimentswasthresholdedatp⬍0.001withavoxelextentof23andthenmaskedbythesimpleeffectfromeachexperimentatp⬍0.01.Effectsareprojectedontoleftandrighthemisphere
inflated surface caret brains (Caret5; Van Essen et al., 2001) and presented from medial and lateral views. The group peaks of activity within each seed region (AnG, mPFC, Hipp, MTG, PCC; see Materials and
Methods)thatwereusedasmaskstoidentifyindividualsubjectpeaksthatwerethenusedtodefineseedregionsaredepictedbyfociprojectedontothelateralandmedialsurfacesofthecaretbrains(Experiment
1, magenta; Experiment 2, green; Experiment 3, blue).

Recollection-related changes in
inter-regional connectivity
To identify brain regions that demon-
strated recollection-related increases in
functional connectivity, we ran five sepa-
rate PPI analyses, one for each of five seed
ROIs, for each experiment (for a total of
15 separate analyses). Seed regions com-
prised different components of the core
recollection network, namely, the left
AnG, mPFC, PCC, and Hipp. In light of
the consistency with which it has been re-
ported to demonstrate recollection effects
(Spaniol et al., 2009; Hayama et al., 2012;
Vilberg and Rugg, 2012, 2014), we also
included the left MTG as a seed region
(see Materials and Methods). The masked
main effects identifying regions where
there were significant and consistent
recollection-related increases in connec-
tivity across experiments (main effect
thresholded at p ⬍ 0.001, uncorrected,
voxel extent ⫽ 23 masked by the simple
effect from each experiment thresholded
at p ⬍ 0.01) are depicted for each seed
region in Figure 2. As is evident from the
figure, increased connectivity for success-
ful relative to unsuccessful recollection
was apparent for all seed regions. Connec- Figure 2. Overlap map of recollection-related activity and connectivity effects. Regions where there was a reliable recollection-
tivity was found to increase both with related change in activity are depicted in green. Regions where there was a reliable recollection-related change in connectivity with
other members of the core recollection a given seed region are depicted in blue. Regions that demonstrated both a recollection-related change in activity and a
network, such as the AnG, MTG, and recollection-related change in connectivity with a given seed region are depicted in cyan. Main effects across experiments were
PCC, and regions that are not typically thresholded at p ⬍ 0.001 with a voxel extent of 23 and then masked by the simple effect from each experiment at p ⬍ 0.01. Effects
are projected onto left and right hemisphere inflated surface caret brains (Caret5; Van Essen et al., 2001) and presented from
considered part of the network, such as medial and lateral views.
the left DLPFC, dorsal aspects of the lat-
eral PCC, dorsal ACC (dACC), and extra-
striate visual cortex.
We tested for any differences between seed regions or experi- ally responsive occipital and fusiform cortex, along with a cluster
ments in recollection-related connectivity changes by entering in the left SMA. In each case, the effect was driven by greater
the first-level contrast images into a single second-level, two-way, connectivity changes in Experiment 1 compared with Experi-
mixed-factor ANOVA model (between-subjects factor of exper- ment 3. The ANOVA also identified a main effect of seed region,
iment, and within-subjects factor of seed region). The ANOVA but again, only in a few small clusters throughout the brain. Re-
failed to identify an experiment by seed region interaction any- gions demonstrating this effect included bilateral occipitoparietal
where in the brain. A few small clusters did however demonstrate cortex, left inferior temporal cortex, left precentral gyrus, right
a main effect of experiment. These were mainly localized to visu- inferior frontal gyrus, PCC, and bilateral cerebellum. Although
King et al. • Recollection-Related Increases in Connectivity J. Neurosci., January 28, 2015 • 35(4):1763–1772 • 1767

Figure 3. Correlations between recollection-related increases in connectivity and recollection accuracy. a, The depicted matrices represent the size of the correlation between seed–target
connectivity and behavioral performance across five seeds and three experiments. Within each matrix, each column corresponds to one of the five seed regions for which a PPI analysis was conducted.
Each row of the matrix corresponds to one of the 90 target regions defined by the AAL atlas. Colors within each cell of the matrix represent the size of the correlation (Pearson’s r) between recollection
accuracy and the recollection-related increase in connectivity associated with each seed–target pair. Matrices correspond to Experiments 1 (top), 2 (middle), and 3 (bottom). Matrices on the left and
right are identical except that those on the left are coded continuously, whereas those on the right are thresholded at r ⬎ 0.41 ( p ⬍ 0.05 for the experiment with the smallest sample size). b, Bar
graphs represent the average correlation between recollection-related increases in connectivity and recollection accuracy for each seed region (left), and the proportion of brain regions where the
change in connectivity correlated with recollection accuracy with an effect size of r ⬎ 0.41 (right). Error bars reflect the SEM; †p ⬍ 0.05, *p ⬍ 0.01, **p ⬍ 0.001.

the pattern of results driving this effect differed for each cluster, in varied across subjects with recollection accuracy. For each
general, estimates of changes in connectivity tended to be greater experiment, we constructed seed–target correlation matrices,
for the AnG and MTG seed regions relative to the other seed where each column of the matrix represented one of the five seed
regions. regions, and each row represented one of the 90 target regions
defined by the AAL atlas (see Materials and Methods). For each
Relationship between recollection-related changes in seed–target pair, we extracted the mean parameter estimate of
connectivity and recollection accuracy connectivity change for every subject. We then calculated the
We next investigated whether the magnitude of recollection- across-subjects correlation between recollection accuracy and
related increases in connectivity in the different seed regions co- recollection-related connectivity change for each seed–target
1768 • J. Neurosci., January 28, 2015 • 35(4):1763–1772 King et al. • Recollection-Related Increases in Connectivity

pair. The resulting correlation matrices are depicted in Figure 3a. correlations between connectivity change and memory perfor-
To quantify the strength of the relationship between connectivity mance were re-estimated after controlling for BOLD signal
changes and recollection accuracy, for each seed, we calculated change in the both seed and target regions, the results were almost
the average correlation coefficient across all target regions (Fig. identical to those from the original analyses.
3b). Permutation analyses tested whether these values were sig-
nificantly greater than what would be expected by chance (see Correlations between recollection-related changes in connectivity
Materials and Methods). The strongest results were in Experi- and other behavioral measures
ment 2, where the average size of the correlation exceeded chance To assess the specificity of the relationship between connectivity
( p ⬍ 0.01, which corresponds to a Bonferroni-corrected ␣-level and recollection accuracy, we investigated whether recollection-
of p ⬍ 0.05, after correcting for 5 comparisons) in all five seed related changes in connectivity correlated with other perfor-
regions. In Experiment 3, the average correlation was significant mance measures, including reaction time (RT; cf. Fornito et al.,
for the mPFC, Hipp, and MTG seeds, and it was also significant in 2012), and estimates of familiarity and response bias. RT was
the AnG and PCC seeds before Bonferroni correction ( p ⬍ 0.05, defined as the average response latency to all test items. In
uncorrected). In Experiment 1, the average correlation across Experiment 1, familiarity was estimated according the procedure
target regions exceeded chance only for the Hipp seed region after recommended by Yonelinas and Jacoby (1995) [pKhit/(1-pRhit) ⫺
Bonferroni correction; however, without this correction, it ex- pKfalse alarm/(1 ⫺ pRfalse alarm)]. In Experiment 2, familiarity was
ceeded chance ( p ⬍ 0.05) for all five seed regions. defined as pRearranged 兩 Intact/(pRearranged 兩 Intact ⫹ pNew 兩
We also calculated the proportion of brain regions where the Intact) ⫺ pRearranged 兩 New/(pRearranged 兩 New ⫹ pNew 兩 New).
change in connectivity correlated with recollection accuracy with In Experiment 3, familiarity was defined as pSI/(1 ⫺ pSC), where pSI
an effect size of r ⫽ ⬎0.41 (Fig. 3b). This effect size was chosen is the proportion of incorrect source judgments and pSC is the pro-
because it represents the smallest statistically significant value portion of source correct judgments. Bias was assessed in Experi-
( p ⬍ 0.05, two-tailed) in the experiment with the smallest sample ments 1 and 2 only, as in Experiment 3, all behavioral and neural
size (Experiment 1). Permutation analyses were again used to measures were collapsed across the two possible source judgments.
determine whether these proportions exceeded chance (see Ma- In Experiment 1, bias was calculated as pRfalse alarm/(pRfalse alarm ⫹
terials and Methods). In Experiment 2, the proportions of corre- pKflase alarm). In Experiment 2, bias was defined as pIntact 兩
lations exceeding our chosen threshold were greater than chance Rearranged/(pIntact 兩 Rearranged ⫹ pRearranged 兩 Rearranged).
in all five seed regions ( p ⬍ 0.01). In Experiment 3, proportions To test whether there was a relationship between recollection-
were significant for the mPFC and MTG seed regions after cor- related changes in connectivity and the foregoing measures of
recting for multiple comparisons ( p ⬍ 0.01), and for each of the behavioral performance, we ran a series of analyses identical to
remaining seeds (AnG, Hipp, and PCC) without this correction those used for recollection accuracy, but with the recollection
( p ⬍ 0.05). In Experiment 1, proportions were significant for the accuracy data replaced by RT, familiarity, and bias respectively.
AnG and Hipp seeds only after correcting for multiple compari- Permutation analyses revealed that for each of these measures,
sons ( p ⬍ 0.01), but without this correction, the proportions neither the average size of the correlations nor the proportion of
significantly exceeded chance for all of the seed regions ( p ⬍ significant correlations exceeded chance. We also re-estimated
0.05). Together, these analyses indicate that recollection-related the recollection accuracy correlation matrices after partialling
changes in connectivity between each of the five seed regions and out the effects of RT, familiarity, and bias. In each case, the results
much of the cortex were reliably correlated with behavioral per- closely resembled those from the original analyses. Hence, these
formance. We created heat maps for each seed region to visualize findings demonstrate that the relationship between accuracy and
the target regions where the correlations between connectivity recollection-related connectivity change cannot be explained by
change and performance were most consistent (Fig. 4). As is ev- variance in RT, familiarity, or bias; rather, the relationship be-
ident from the figure, the correlation between accuracy and con- tween changes in connectivity and performance are specific to
nectivity exceeded r ⫽ ⬎0.41 in at least two experiments in the recollection accuracy.
majority of seed–target pairs, with this relationship occurring in
all three experiments in several seed–target pairs, especially for the Correlations between recollection-related changes in connectivity
AnG, mPFC, and PPC seeds. Although the regions demonstrating and pR controlling for differences in the proportion of trials
correlations between connectivity and performance were diffusely associated with successful and unsuccessful recollection
spread throughout the brain, regions where this relationship was In each experiment, individuals with better recollection perfor-
observed most consistently across experiments were mPFC, PCC, mance tended to have a higher ratio of successful to unsuccessful
left AnG, superior frontal gyrus, and middle occipital gyrus. recollection trials than did worse performers. Thus, across sub-
ject differences in the relative proportions of trials in the two
Comparison and control analyses response categories might be responsible for mediating the rela-
We ran several additional analyses to rule out possible confounds tionship between connectivity change and performance. To
with respect to the correlations between connectivity change and assess this possibility, we reanalyzed the data after randomly se-
memory performance described above, and to narrow down the lecting subsets of trials to equate the number of successful and
range of potential explanations. unsuccessful recollection trials for each subject. The relationships
between connectivity change and recollection accuracy were
Correlations between pR and recollection-related changes in largely unaffected. Hence, the relationship is not mediated by
BOLD activity across-subject variation in relative trial numbers.
We examined the relationship between recollection accuracy and
the recollection-related increase in BOLD activity in both target Correlations between pR and item-related connectivity change
and seed regions. Unlike the PPI effects, recollection-related in- In a final analysis, we investigated whether the relationship be-
creases in BOLD signal were not reliably correlated with memory tween recollection-related connectivity increases and recollec-
performance. Unsurprisingly, in light of these findings, when the tion accuracy was a manifestation of a more general relationship
King et al. • Recollection-Related Increases in Connectivity J. Neurosci., January 28, 2015 • 35(4):1763–1772 • 1769

Figure 4. Heat maps representing, for each of the 90 ROIs, the number of experiments for which there was a significant relationship between connectivity change and recollection accuracy.
Borders of the 90 cortical, striatal, and thalamic regions defined by the AAL atlas are projected onto left and right hemisphere inflated surface caret brains (Caret5; Van Essen et al., 2001) which are
presented from medial and lateral views. For each seed, target regions where the correlation between connectivity and accuracy was significant in all three experiments are presented in red; regions
where this relationship was significant in two of three experiments are presented in orange; regions where this relationship was significant in only one of three experiments are presented in yellow,
and regions where this relationship was not significant in any of the three experiments are presented in white. Correlations were considered statistically significant if the correlation coefficient
(Pearson’s r) exceeded r ⬎ 0.41. This effect size was chosen because it corresponds to the smallest statistically significant value (p ⬍ 0.05, two-tailed) in the experiment with the smallest sample size.

between brain dynamics and performance. By this hypothesis, connectivity both with other members of the core recollection
the brains of higher performing individuals are simply more ca- network and with regions extrinsic to the network, notably the
pable, in general, of switching between different states or network DLPFC and intraparietal sulcus, where BOLD activity did not
configurations than those of lower performing individuals. We vary with recollection success. These findings add to previous
tested this hypothesis by running additional PPI analyses in work pointing to the significance of increased connectivity in the
which the task regressor (and consequently the PPI regressor) context of episodic memory retrieval (Staresina et al., 2013; Wa-
was constructed by weighting every test trial as 1, thereby gener- trous et al., 2013; Schedlbauer et al., 2014). In addition, the mag-
ating a PPI that represented where connectivity increased relative nitude of recollection-related increases in connectivity covaried
to baseline each time a test item was presented. The same analysis
across subjects with recollection accuracy. Although prior studies
procedures that identified the relationship between recollection-
have reported relationships between memory performance and
related connectivity increases and performance failed to identify
low-frequency fluctuations in the BOLD signal during sustained
any such relationship for this more general measure of item-
related connectivity change. periods of task engagement or rest (Wig et al., 2008; Hampson et
al., 2010; Tambini et al., 2010; Wang et al., 2010; but see, Schedl-
Discussion bauer et al., 2014), the present findings constitute novel evidence
In three experiments using different operational definitions of that recollection-related modulation of connectivity can also
recollection, brain regions exhibiting recollection-related in- serve as a predictor of individual differences in recollection
creases in BOLD activity also demonstrated increased functional accuracy.
1770 • J. Neurosci., January 28, 2015 • 35(4):1763–1772 King et al. • Recollection-Related Increases in Connectivity

The results from the three datasets were highly consistent, the memoranda used in the three experiments (i.e., pictures and
with respect to both the regions where recollection-related in- names of concrete objects), it is perhaps unsurprising that extra-
creases in activity and connectivity were identified, and where striate visual areas, where “reinstatement” of retrieved visual in-
increases in connectivity correlated with performance. Whereas formation would be expected (for review, see Danker and
some differences in the pattern of connectivity effects across ex- Anderson, 2010), interacted with core recollection regions dur-
periments were identified, we focus below on features of the data ing successful recollection.
that converged across experiments. This is not to suggest that the Regardless of the region used as a seed, neither the mPFC
across-experiment differences that we observed are unimportant. nor the hippocampus was identified as showing enhanced
It is possible, for example, that the findings that connectivity recollection-related connectivity. However, when these two re-
increases in fusiform cortex were greater in Experiment 1 than gions were themselves used as seeds, they identified similar pat-
Experiment 3 reflects between-experiment differences in the rel- terns of recollection-related connectivity increases to those
evance of retrieved visual information for the respective memory identified for the other seed regions. Such “unidirectional” or
tests. Similarly, the finding that the hippocampal seed demon- “asymmetric” relationships are not uncommon in PPI analyses
strated the strongest correlation with recollection performance in (Gerchen et al., 2014). Although it is tempting to interpret such
Experiment 1, whereas the strongest correlations were found for asymmetries as evidence for effective connectivity (that is, a
the mPFC and MTG seeds in the other experiments, might signify causal influence) between seed and target regions, there are sev-
that the three regions were engaged differentially across experi- eral alternative explanations that can account for this pattern of
ments. At present, however, there is no means to determine results (e.g., differential colinearity between task and PPI regres-
which of the numerous variables that differed between the exper- sors in different seed regions). Therefore, it would be premature
iments were responsible for these or other differences in the pat- to draw inferences about the directionality of the flow of infor-
tern of the results. mation based on such asymmetries.
Consistent with prior studies (for review, see Spaniol et al., The current findings not only demonstrated that a number of
2009; Kim, 2010; Rugg and Vilberg, 2013) several regions that brain regions showed enhanced recollection-related connectivity
showed enhanced recollection-related activity in the current ex- with members of the core recollection network, but also that the
periments corresponded with those held to constitute a core rec- magnitude of connectivity change throughout much of the cortex
ollection network, including left AnG, mPFC, PCC, and Hipp, as correlated across individuals with recollection accuracy. Al-
well as the left MTG. When each of these regions was used as a though recollection-related increases in connectivity, as identi-
seed, recollection-related connectivity increases were consis- fied at the group level, were relatively restricted anatomically,
tently identified both with other members of this network (al- regions where the magnitude of connectivity increases correlated
though see below), and regions outside of the network, including with performance were more broadly distributed. Particularly com-
DLPFC, dorsal aspects of the lateral parietal cortex, dACC, and pelling evidence for the anatomically diffuse relationship between
extrastriate visual cortex. These findings are consistent with the connectivity changes and performance comes from the finding that,
results of two previous studies (Watrous et al., 2013; Schedlbauer in each of the 90 anatomically defined ROIs, a reliable correlation
et al., 2014), both of which identified more connections between between connectivity and performance was evident for at least three
a broadly distributed set of regions during successful than unsuc- of the seed regions across the different experiments (Fig. 4). Thus,
cessful recall of contextual information. Here, we demonstrate the relationship between recollection-related modulation of connec-
not only that the number of inter-regional connections, but also tivity and recollection accuracy might depend on a mechanism dis-
the strength of those connections, varies with recollection suc- tinct from that responsible for the group-level, recollection-related
cess. Together, the present and prior findings point to the impor- connectivity enhancements evident in Figure 2.
tance of both global and regionally specific connectivity changes It seems reasonable to suppose that the present recollection-related
during episodic memory retrieval. enhancements in connectivity reflect increases in inter-regional infor-
Among the regions extrinsic to the core recollection network mation exchange. However, what might be responsible for the ana-
that demonstrated enhanced connectivity with members of the tomically diffuse nature of the relationship between connectivity
network were components of what has been termed the “execu- change and memory performance? One possible explanation is re-
tive metasystem” (Cocchi et al., 2013), which encompasses the lated to the fact that an increase in functional connectivity between
“frontoparietal” and “cingulo-opercular” networks (Seeley et al., two brain regions can reflect not only increased inter-regional com-
2007; Dosenbach et al., 2008; Power et al., 2011; Power and Pe- munication, but also the influence on multiple regions of a com-
tersen, 2013). Regions belonging to these networks are held to act mon, “driving” input (or inputs; Friston, 2011; Smith et al., 2011).
as flexible hubs that couple with different functional networks From this perspective, the changes in connectivity that covary with
depending on task demands (Chadick and Gazzaley, 2011; Cole behavioral performance might reflect a generic mechanism that
et al., 2013). Hence, recollection-related modulation of func- modulates inter-regional synchrony across much of the brain. One
tional connectivity between these control networks and the core possibility is that these effects are caused by event-related variations
recollection network might reflect the allocation of control pro- in the influence of one or more of the ascending neuromodulatory
cesses to the support of postretrieval operations, such as, evalua- systems that innervate much of the cortex and other forebrain
tion of the products of recollection and the selection of the structures (Schölvinck et al., 2010; Lee and Dan, 2012). Accord-
appropriate response (Rugg, 2004; Achim and Lepage, 2005; ing to this hypothesis, the strength of any such neuromodulatory
Ranganath et al., 2007). Thus, brain regions where recollection- influence covaries with the strength of the recollection “signal,”
related changes in signal magnitude are not detected might none- and hence with across-trial estimates of recollection perfor-
theless contribute to retrieval processing by interacting with mance. Importantly, the present data do not allow for an assess-
regions that do demonstrate a mean signal change. ment of the causal relationship between these two variables; thus,
In addition to regions belonging to control networks, it is unclear whether the strength of the putative modulatory
recollection-related connectivity increases were also consistently input contributes to recollection accuracy, or whether strength of
identified in extrastriate visual cortex. Given the visual nature of recollection influences cortical synchrony.
King et al. • Recollection-Related Increases in Connectivity J. Neurosci., January 28, 2015 • 35(4):1763–1772 • 1771

In summary, the present findings suggest that coordinated physiological and modulatory interactions in neuroimaging. Neuroimage
functional interactions between distributed brain regions form 6:218 –229. CrossRef Medline
part of the neural mechanism supporting successful recollection. Gardiner JM (1988) Functional aspects of recollective experience. Mem
Cognit 16:309 –313. CrossRef Medline
Regions where activity varied as a function of recollection success Gerchen MF, Bernal-Casas D, Kirsch P (2014) Analyzing task-dependent
demonstrated enhanced recollection-related connectivity with a brain network changes by whole-brain psychophysiological interactions:
consistent set of brain regions, including both regions belonging a comparison to conventional analysis. Hum Brain Mapp 35:5071–5082.
to and regions extrinsic to the core recollection network. This CrossRef Medline
finding suggests that in addition to the core recollection network, Greicius MD, Krasnow B, Reiss AL, Menon V (2003) Functional connectiv-
other regions might contribute to recollection despite their fail- ity in the resting brain: A network analysis of the default mode hypothesis.
ure to demonstrate changes in signal magnitude. The findings Proc Natl Acad Sci U S A 100:253–258. CrossRef Medline
Hampson M, Driesen N, Roth JK, Gore JC, Constable RT (2010) Functional
also demonstrated that the magnitude of connectivity change in
connectivity between task-positive and task-negative brain areas and its
widely distributed brain areas correlated across individuals with relation to working memory performance. Magn Reson Imaging 28:
recollection accuracy, leading to the conjecture that these 1051–1057. CrossRef Medline
changes reflect a transient neuromodulatory input, which influ- Hayama HR, Vilberg KL, Rugg MD (2012) Overlap between the neural cor-
ences synchronicity throughout much of the brain. The finding relates of cued recall and source memory: evidence for a generic recollec-
that connectivity, unlike BOLD signal amplitude, covaried across tion network? J Cogn Neurosci 24:1127–1137. CrossRef Medline
subjects with recollection accuracy suggests that it may have Jacoby LL (1991) A process dissociation framework: separating automatic
promise for the understanding of individual differences in mem- from intentional uses of memory. J Mem Lang 30:513–541. CrossRef
Kim H (2010) Dissociating the roles of the default-mode, dorsal, and ventral
ory accuracy. Future research will be necessary to determine
networks in episodic memory retrieval. Neuroimage 50:1648 –1657.
whether these dynamic changes in connectivity, and their rela- CrossRef Medline
tion with behavioral performance, are specific to the domain of Lee SH, Dan Y (2012) Neuromodulation of brain states. Neuron 76:209 –
episodic memory. Regardless, the current results highlight the 222. CrossRef Medline
value of examining event-related modulations of functional con- Mandler G (1980) Recognizing: the judgment of previous occurrence. Psy-
nectivity, in addition to modulation of local signal amplitude, to chol Rev 87:252–271. CrossRef
gain a more complete understanding of the neural correlates of Power JD, Petersen SE (2013) Control-related systems in the human brain.
successful memory retrieval. Curr Opin Neurobiol 23:223–228. CrossRef Medline
Power JD, Cohen AL, Nelson SM, Wig GS, Barnes KA, Church JA, Vogel AC,
Laumann TO, Miezin FM, Schlaggar BL, Petersen SE (2011) Functional
References network organization of the human brain. Neuron 72:665– 678. CrossRef
Achim AM, Lepage M (2005) Dorsolateral prefrontal cortex involvement in Medline
memory post-retrieval monitoring revealed in both item and associative Ranganath C, Heller A, Cohen MX, Brozinsky CJ, Rissman J (2005) Func-
recognition tests. Neuroimage 24:1113–1121. CrossRef Medline tional connectivity with the hippocampus during successful memory for-
Brett M, Anton JL, Valabregue R, Poline JB (2002) Region of interest anal- mation. Hippocampus 15:997–1005. CrossRef Medline
ysis using an SPM toolbox. Presentation at the 8th International Confer- Ranganath C, Heller AS, Wilding EL (2007) Dissociable correlates of two
ence on Functional Mapping of the Human Brain, June 2– 6, Sendai, classes of retrieval processing in prefrontal cortex. Neuroimage 35:1663–
Japan. 1673. CrossRef Medline
Chadick JZ, Gazzaley A (2011) Differential coupling of visual cortex with Rissman J, Gazzaley A, D’Esposito M (2004) Measuring functional connec-
default or frontal-parietal network based on goals. Nat Neurosci 14:830 – tivity during distinct stages of a cognitive task. Neuroimage 23:752–763.
832. CrossRef Medline CrossRef Medline
Cocchi L, Zalesky A, Fornito A, Mattingley JB (2013) Dynamic cooperation Rugg MD (2004) Retrieval processing in human memory: electrophysiolog-
and competition between brain systems during cognitive control. Trends ical and fMRI evidence. In: The cognitive neurosciences, Ed 3, pp 727–
Cogn Sci 17:493–501. CrossRef Medline 737. Cambridge, MA: MIT.
Cole MW, Reynolds JR, Power JD, Repovs G, Anticevic A, Braver TS (2013) Rugg MD, Vilberg KL (2013) Brain networks underlying episodic memory
Multi-task connectivity reveals flexible hubs for adaptive task control. Nat retrieval. Curr Opin Neurobiol 23:255–260. CrossRef Medline
Neurosci 16:1348 –1355. CrossRef Medline Schedlbauer AM, Copara MS, Watrous AJ, Ekstrom AD (2014) Multiple
Danker JF, Anerson JR (2010) The ghosts of brain states past: remembering interacting brain areas underlie successful spatiotemporal memory re-
reactivates the brain regions engaged during encoding. Psychol Bull 136:
trieval in humans. Sci Rep 4:6431. CrossRef Medline
87–102. CrossRef Medline
Schölvinck ML, Maier A, Ye FQ, Duyn JH, Leopold DA (2010) Neural basis
Dosenbach NU, Fair DA, Cohen AL, Schlaggar BL, Petersen SE (2008) A
of global resting-state fMRI activity. Proc Natl Acad Sci U S A 107:10238 –
dual-networks architecture of top-down control. Trends Cogn Sci 12:99 –
10243. CrossRef Medline
105. CrossRef Medline
Seeley WW, Menon V, Schatzberg AF, Keller J, Glover GH, Kenna H, Reiss
Elward RL, Vilberg KL, Rugg MD (2014) Motivated memories: effects of
AL, Greicius MD (2007) Dissociable intrinsic connectivity networks for
reward and recollection in the core recollection network and beyond.
Cereb Cortex. Advance online publication. Retrieved May 28, 2014. salience processing and executive control. J Neurosci 27:2349 –2356.
CrossRef Medline CrossRef Medline
Fornito A, Harrison BJ, Zalesky A, Simons JS (2012) Competitive and co- Smith SM, Miller KL, Salimi-Khorshidi G, Webster M, Beckmann CF, Nich-
operative dynamics of large-scale brain functional networks supporting ols TE, Ramsey JD, Woolrich MW (2011) Network modelling methods
recollection. Proc Natl Acad Sci U S A 109:12788 –12793. CrossRef for FMRI. Neuroimage 54:875– 891. CrossRef Medline
Medline Spaniol J, Davidson PS, Kim AS, Han H, Moscovitch M, Grady CL (2009)
Fox MD, Snyder AZ, Vincent JL, Corbetta M, Van Essen DC, Raichle ME Event-related fMRI studies of episodic encoding and retrieval: meta-
(2005) The human brain is intrinsically organized into dynamic, anticor- analyses using activation likelihood estimation. Neuropsychologia 47:
related functional networks. Proc Natl Acad Sci U S A 102:9673–9678. 1765–1779. CrossRef Medline
CrossRef Medline Staresina BP, Cooper E, Henson RN (2013) Reversible information flow
Friston KJ (2011) Functional and effective connectivity: a review. Brain Con- across the medial temporal lobe: the hippocampus links cortical mod-
nect 1:13–36. CrossRef Medline ules during memory retrieval. J Neurosci 33:14184 –14192. CrossRef
Friston KJ, Frith CD, Liddle PF, Frackowiak RS (1993) Functional connec- Medline
tivity: the principal-component analysis of large (PET) data sets. J Cereb Tambini A, Ketz N, Davachi L (2010) Enhanced brain correlations during
Blood Flow Metab 13:5–14. CrossRef Medline rest are related to memory for recent experiences. Neuron 65:280 –290.
Friston KJ, Buechel C, Fink GR, Morris J, Rolls E, Dolan RJ (1997) Psycho- CrossRef Medline
1772 • J. Neurosci., January 28, 2015 • 35(4):1763–1772 King et al. • Recollection-Related Increases in Connectivity

Tulving E (1985) Memory and consciousness. Can Psychol Can 26:1–12. performance in cognitively intact older individuals. Neuroimage 51:910 –
CrossRef 917. CrossRef Medline
Tzourio-Mazoyer N, Landeau B, Papathanassiou D, Crivello F, Etard O, Del- Watrous AJ, Tandon N, Conner CR, Pieters T, Ekstrom AD (2013)
croix N, Mazoyer B, Joliot M (2002) Automated anatomical labeling of Frequency-specific network connectivity increases underlie accurate spa-
activations in SPM using a macroscopic anatomical parcellation of the tiotemporal memory retrieval. Nat Neurosci 16:349 –356. CrossRef
MNI MRI single-subject brain. Neuroimage 15:273–289. CrossRef Medline
Medline Wig GS, Grafton ST, Demos KE, Wolford GL, Petersen SE, Kelley WM
Van Essen DC, Drury HA, Dickson J, Harwell J, Hanlon D, Anderson CH (2008) Medial temporal lobe BOLD activity at rest predicts individual
(2001) An integrated software suite for surface-based analyses of cerebral differences in memory ability in healthy young adults. Proc Natl Acad Sci
cortex. J Am Med Inform Assoc 8:443– 459. CrossRef Medline U S A 105:18555–18560. CrossRef Medline
Vilberg KL, Rugg MD (2012) The neural correlates of recollection: transient Wig GS, Schlaggar BL, Petersen SE (2011) Concepts and principles in the
versus sustained FMRI effects. J Neurosci 32:15679 –15687. CrossRef analysis of brain networks. Ann N Y Acad Sci 1224:126 –146. CrossRef
Medline Medline
Vilberg KL, Rugg MD (2014) Temporal dissociations within the core recol- Yonelinas AP (2002) The nature of recollection and familiarity: a review of 30
lection network. Cogn Neurosci 5:77– 84. CrossRef Medline years of research. J Mem Lang 46:441–517. CrossRef
Wang L, Laviolette P, O’Keefe K, Putcha D, Bakkour A, Van Dijk KR, Pihla- Yonelinas AP, Jacoby LL (1995) The relation between remembering and
jamäki M, Dickerson BC, Sperling RA (2010) Intrinsic connectivity be- knowing as bases for recognition: effects of size congruency. J Mem Lang
tween the hippocampus and posteromedial cortex predicts memory 34:622– 643. CrossRef