Abstract

Given the diversity of stimuli encountered in daily life, a variety of strategies must be used for learning new information. Relating and encoding visual and verbal stimuli into memory has been probed using various tasks and stimulus types. Engagement of specific subsequent memory and cortical processing regions depends on the stimulus modality of studied material; however, it remains unclear whether different encoding strategies similarly influence regional activity when stimulus type is held constant. In this study, participants encoded object pairs using a visual or verbal associative strategy during fMRI, and subsequent memory was assessed for pairs encoded under each strategy. Each strategy elicited distinct regional processing and subsequent memory effects: middle/superior frontal, lateral parietal, and lateral occipital for visually associated pairs and inferior frontal, medial frontal, and medial occipital for verbally associated pairs. This regional selectivity mimics the effects of stimulus modality, suggesting that cortical involvement in associative encoding is driven by strategy and not simply by stimulus type. The clinical relevance of these findings, probed in a patient with a recent aphasic stroke, suggest that training with strategies utilizing unaffected cortical regions might improve memory ability in patients with brain damage.

INTRODUCTION

Although certain brain regions are commonly reported to show general subsequent memory effects, some regions only show effects for specific stimulus types being encoded or tasks being used. During visual or picture-based encoding, subsequent memory effects have often been reported in regions including right prefrontal, superior parietal, lateral occipital, and fusiform cortex, whereas during verbal or word-based encoding, effects are often reported in regions including left inferior frontal, parietal, superior temporal, lingual, and medial frontal cortex (Kim, 2011).

An unanswered question remains, however, regarding whether these differences in regional memory responses for visual and verbal stimuli are due to the bottom–up processing of the stimuli or to the top–down strategy being used to form the memory. The influence that the implemented associative strategy has on regional activity is of considerable importance to the design and interpretation of memory studies. Whether instructed or uninstructed, participants will use a strategy for encoding, retrieving, and relating items. If regional activity is dependent on the strategy employed, instructing participants to perform a relational processing step during encoding could itself influence the subsequent memory effects. Given the large number of studies in which participants are instructed to perform a relational processing step during encoding, a better understanding of what is driving these subsequent memory effects is needed.

Additionally, determining whether the strategy or the stimulus type dictates which brain regions are more engaged in encoding could have clinical relevance. It is well established that regional neocortical atrophy occurs in neurodegenerative disorders (Dickerson et al., 2009; Listerud, Powers, Moore, Libon, & Grossman, 2009; Mummery et al., 2000). If the brain regions engaged during encoding are dependent on the stimulus type, there is little flexibility in the system, and patients will struggle when encountering stimuli that require the function of damaged regions for processing and encoding. However, if the brain regions engaged during encoding are driven by the strategy being employed, such findings could hold clinical significance in their implications for rehabilitation strategies for patients with regional brain damage or neurodegenerative disease.

A prior study elegantly examined individuals' use of different encoding strategies when no specific strategy instructions were provided (Kirchhoff & Buckner, 2006). Participants' strategies were identified using a retrospective self-report questionnaire following testing. The authors found that participants with greater use of “visual inspection” or “verbal elaboration” had better memory performance. Additionally, they found that the degree to which participants used the visual or verbal strategy was positively correlated with activation within brain regions commonly involved in visual and verbal processing, respectively (i.e., extrastriate and prefrontal regions). The authors noted the limitation, however, of relying on retrospective self-report of participants' strategy use. To examine and directly compare trials encoded using a visual or verbal strategy across participants, the instruction to use a visual versus verbal strategy would ideally be directly manipulated.

Currently, it remains unclear whether regional encoding activity is fixed to the stimulus type or driven by the employed strategy. Therefore, the present studies examine whether visual and verbal strategies influence regional subsequent memory effects if the stimulus type is held constant. In Experiment 1, regional brain responses were examined using rapid-event-related fMRI while participants associated pairs of visually presented nameable objects using either a visual imagery or verbal rehearsal strategy. Despite the robust differences in regional subsequent memory effects seen when visual versus verbal stimuli are encoded, the prediction was that similar regional distinctions would be seen when the stimulus type is held constant and participants are instructed to use a visual versus a verbal strategy. Experiment 2, a modified behavioral version of Experiment 1, was conducted in a patient recovering from a recent aphasic stroke. By examining the impact of using an instructed visual or verbal associative encoding strategy relative to the patient's own natural strategy, the theoretical implications of Experiment 1 were applied to demonstrate potential clinical relevance.

METHODS

Experiment 1

Participants

Seventeen healthy volunteers (mean age = 24.8 ± 3.2 years, four men) were recruited from the University of California, San Diego (UCSD) community and the surrounding area. All participants gave informed consent approved by the UCSD Institutional Review Board and had normal or corrected vision. Participants received monetary compensation for their time. Two participants were excluded from analyses due to poor behavioral performance.

Stimuli

Two hundred sixty-eight (268) color images of everyday objects were used as stimuli in this experiment. Twelve images were used for practice trials before scanning to ensure that participants understood the instructions. While in the scanner, participants were shown 256 images presented in concurrent pairs for encoding; each item was shown once resulting in 128 image pairs. One object from each pair was presented to the participant during the postscan recognition test. Images were acquired from Rossion and Pourtois color Snodgrass images (Rossion & Pourtois, 2004) and Hemera object library (Hemera Technologies, Inc.).

Experimental Procedure

During the associative encoding task in the scanner, participants were shown pairs of concurrently presented images, with each pair presented for 2 sec followed by a blank screen for 3 sec (Figure 1A). All items were paired pseudorandomly to remove obvious semantic, verbal, or visual relationships within pairs. The order of the runs and the pairs assigned to each task type were counterbalanced to prevent systematic confounding effects due to the nameablity of each object. A fixation cross was presented during jittered ISIs ranging between 1.5 and 10.5 sec. The ISIs were calculated to optimize the study design for modeling the hemodynamic response to trials (Dale, 1999; Dale & Buckner, 1997). Participants were instructed to encode and associate 64 pairs, split evenly among two functional runs, using the visual strategy of imagining the two items visually merged during the 3-sec blank screen following the presentation of the pair without verbalizing the object names (e.g., Figure 1, “Visual”). Participants were instructed to encode and associate the other 64 pairs, split evenly among two additional functional runs, using the verbal strategy of creating a compound word made up of the names of the two objects and rehearsing it during the 3-sec blank screen following the presentation of the pair without visualizing the objects (e.g., Figure 1, “Verbal”). Participants were given a button box and were asked to press a button whenever a pair of objects appeared on the screen to make sure that participants were attending to each image pair. One hundred twenty-eight (128) image pairs were presented to participants in the scanner across four 331-sec runs. Each image was presented once.

Figure 1. 

Experimental design. (A) Example of the encoding task used in the scanner illustrated with a sample pair of objects. Each stimulus pair was presented for 2 sec, followed by 3 sec of blank screen during which participants associated the objects using the instructed strategy. Under the visual associative encoding condition, participants were instructed to visualize the two objects as merged, without verbalizing their names. Under the verbal associative encoding condition, participants were instructed to combine and rehearse the names of the two objects, without visualizing the objects. Following the 3-sec blank screen and before the next pair was presented, a fixation cross appeared on the screen for a jittered 1.5- to 10.5-sec ISI. (B) Example of the postscan recognition test. Participants were shown one object from each pair previously viewed during the encoding task, and they were asked to verbally report to the experimenter their answers to three questions (a, b, and c). After the participants reported if each item was part of a visual or verbal association (a), the experimenter read the participants three choice pair items, one of which was the correct pair. Participants then reported which of those three choices they recognized as the pair (b) and their confidence in selecting the correct pair (c).

Figure 1. 

Experimental design. (A) Example of the encoding task used in the scanner illustrated with a sample pair of objects. Each stimulus pair was presented for 2 sec, followed by 3 sec of blank screen during which participants associated the objects using the instructed strategy. Under the visual associative encoding condition, participants were instructed to visualize the two objects as merged, without verbalizing their names. Under the verbal associative encoding condition, participants were instructed to combine and rehearse the names of the two objects, without visualizing the objects. Following the 3-sec blank screen and before the next pair was presented, a fixation cross appeared on the screen for a jittered 1.5- to 10.5-sec ISI. (B) Example of the postscan recognition test. Participants were shown one object from each pair previously viewed during the encoding task, and they were asked to verbally report to the experimenter their answers to three questions (a, b, and c). After the participants reported if each item was part of a visual or verbal association (a), the experimenter read the participants three choice pair items, one of which was the correct pair. Participants then reported which of those three choices they recognized as the pair (b) and their confidence in selecting the correct pair (c).

Following the encoding task in the scanner, participants completed a self-paced post-scan recognition test examining subsequent memory for the associative pair and strategy used (Figure 1B). Participants were shown one item from each pair previously viewed during the encoding task, that is, the “cue” item. For each of the 128 stimuli, participants were asked to verbally report to the examiner their answers to three questions. First, participants were asked if they visually or verbally associated the cue item with its pair. Second, the examiner read aloud the names of three items, and participants were asked to indicate which of those three choices was the pair of the cue item. Third, participants were asked to report their confidence in identifying the correct pair: “Not Very,” “Somewhat,” or “Very” confident. This recognition test lasted approximately 30 min.

Functional MRI Parameters

Participants were scanned using a 3T GE scanner at the Keck Center for Functional MRI at UCSD. Functional images were acquired using gradient-echo, echo-planar, T2*-weighted pulse sequence (repetition time = 2.5 sec, one shot per repetition, echo time = 30, flip angle = 90°, bandwidth = 31.25 MHz). Forty slices covering the brain were obtained perpendicular to the long axis of the hippocampus with 3.4 × 3.4 × 4 mm voxels. Field maps were acquired to measure and correct for static field inhomogeneities (Smith et al., 2004). A T1-weighted structural scan was acquired in the same plane and with the same voxel size as the functional scans. A high-resolution structural scan was also acquired sagittally using a T1-weighted (1 × 1 × 1 mm) inversion recovery prepared fast spoiled gradient recalled sequence.

Data Analysis

Structural data were transformed into Talairach space (Talairach & Tournoux, 1998) by AFNI using nearest-neighbor interpolation (Cox, 1996) after standard landmarks, including the anterior and posterior commissures, were manually defined on the anatomical scans. Medial-temporal lobe alignment between participants was then optimized using the ROI large deformation diffeomorphic metric mapping (ROI-LDDMM) alignment technique (for more details on these methods, see Miller, Beg, Ceritoglu, & Stark, 2005; Hales & Brewer, 2001).

After functional data from each run were field-map-corrected (Smith et al., 2004), slices were temporally aligned and coregistered using a three-dimensional image alignment algorithm, voxels outside the brain were eliminated using a threshold mask of the functional data, and functional runs were corrected for motion and concatenated, all using the AFNI suite of programs (Cox, 1996). A 4.0-mm FWHM Gaussian filter was also applied to smooth the functional data from each run. A general linear model was constructed using multiple regression analysis; six motion regressors obtained from the registration process were included along with eight behavioral regressors based on subsequent memory performance. Participants' behavioral trials were sorted based on accuracy and participant ratings of associative memory confidence. Trials were first divided into those in which participants were instructed to associate the items visually versus verbally. Each of those two groups of trials were then divided into one of four outcomes based on memory performance and confidence: “high-confidence” if the correct pair item was identified by the participant with a response of “Very” confident; “medium-confidence” if the correct pair was identified by the participant with a response of “Somewhat” confident; “low-confidence” if the correct pair was identified by the participant with a response of “Not Very” confident; or “incorrect” if the correct pair item was not identified. Therefore, there were eight trial outcomes: “Visual high-confidence,” “Visual medium-confidence,” “Visual low-confidence,” “Visual incorrect,” “Verbal high-confidence,” “Verbal medium-confidence,” “Verbal low-confidence,” and “Verbal incorrect.” For each outcome, a hemodynamic response function was derived from the fMRI data using signal deconvolution with tent basis functions and a defined time window of 17.5 sec following the onset of each pair of items (Cox, 1996). Multiple linear regression analysis was used to examine activity for “high-confidence” correct items encoded using a visual versus a verbal associative strategy (i.e., “Visual high-confidence” vs. “Verbal high-confidence”). To examine subsequent memory effects within each associative encoding strategy, a weighted approach was used to identify brain regions that responded linearly to the four memory conditions (i.e., incorrect < low-confidence < medium-confidence < high-confidence) across the whole brain.

Functional data underwent the same Talairach and ROI-LDDMM transformations as were applied to the anatomical data. Whole-brain voxel-wise t tests (p < .05, two-tailed) carried out across all 15 participants were conducted to examine which brain regions showed more activity for successful associative encoding using a visual versus verbal strategy and which regions showed a significant linearly weighted subsequent memory effect within each strategy separately. To correct for multiple comparisons and yield a whole-brain significance value of p < .05 corrected for all comparisons (based on Monte Carlo simulations), functional clusters of at least 12 contiguous voxels were identified in these conditions. When a whole-brain significance value of p < .001 was used for illustrative purposes, functional clusters of least four contiguous voxels were required to correct for multiple comparisons. Average hemodynamic response functions were extracted for each condition and cluster of interest, and time courses were displayed for illustrative purposes (Kriegeskorte, Simmons, Bellgowan, & Baker, 2009). An ANOVA was conducted to examine visual and verbal incorrect and high-confidence correct responses in the functional regions identified in the “Visual high-confidence” versus “Verbal high-confidence” contrasts. This analysis was performed to examine Strategy × Memory interactions within strategy-specific regions.

Experiment 2

Participants

Participants included one female patient with a focal lesion localized to the left inferior frontal lobe due to a very recent stroke (Figure 6A) and four healthy controls (mean age = 54.3 ± 2.1 years). The patient (age = 46 years) was tested 13 days poststroke. All participants gave informed consent approved by the UCSD Institutional Review Board and had normal or corrected vision. Participants received monetary compensation for their time.

Additional Patient Information

The patient was a 46-year-old woman with no significant past medical history who awoke on the day of admission with an inability to speak, except to answer “yes” or “no.” Upon presentation to the hospital, her National Institute of Health Stroke Scale was 2 for severe Broca's nonfluent aphasia. There were no other neurologic abnormalities. She was started on aspirin at 325 mg daily immediately following her admission. An MRI was obtained confirming an acute stroke in the left frontal lobe as well as a punctuate infarct (less than 2 mm in diameter) slightly lateral to the right occipital pole. There was also an old subclinical right frontal lobe stroke noted on MRI. Magnetic resonance angiography was normal. Hypercoagulability screening was negative. Physical therapy, occupational therapy, and speech therapy were consulted and followed the patient during her stay. She was discharged with ability to create full sentences but remained with inappropriate pauses in speech; some difficulties in spelling; and occasional dropped prepositions, pronouns, and articles. She was ambulatory and had no motor deficits.

Stimuli

Two hundred fifty-six (256) of the stimuli used in Experiment 1 were used in Experiment 2.

Experimental Procedure

The task used for behavioral testing was presented to the participants on a laptop. Participants were shown eight sets of encoding and retrieval runs. All items were paired pseudorandomly to remove obvious semantic, verbal, or visual relationships within pairs, and stimulus pairs were randomly assigned to each run. During each encoding run, participants were shown 16 pairs of concurrently presented images. Each pair was presented for 2 sec followed by a 3-sec blank screen and 5-sec fixation cross before the next pair was presented. Following the 16th pair, participants were given approximately 10 sec of a distractor task, in which they were asked to count backwards by a certain interval from a particular starting number (e.g., “Count backwards aloud from 47 by fours….”). Once they counted four numbers aloud, the experimenter advanced to the retrieval run. Participants were given a retrieval run following each encoding run. During retrieval, participants were shown one item from each of the 16 pairs at the top of the screen and three choice images at the bottom of the screen. The position of the correct pair, which was one of the three choice images, varied pseudorandomly between left, center, and right locations. The lure images were two unpaired items from the prior encoding run, so that all three choice images were from the same encoding session. Participants responded by verbally identifying or pointing to their choice image, and the experimenter pressed the key that corresponded with the selected image, which advanced the run to the next question. The retrieval runs were self-paced. List presentation order allowed assessment for practice effects. For the first three encoding runs, participants were instructed to associate and remember the pairs, and they were not told to use any particular strategy. For the fourth and eighth encoding runs, participants were given the verbal strategy instructions used in Experiment 1. For the fifth, sixth, and seventh runs, participants were given the visual strategy instructions used in Experiment 1.

Data Analysis

The number of pairs correctly identified for each run was calculated, and averages were calculated for each strategy (own, verbal, and visual).

RESULTS

Experiment 1

Behavioral Analysis

Participants' memory for the strategy was accurate relative to chance performance (visual: t(14) = 7.854, p < .001; verbal: t(14) = 4.586, p < .001) and performance did not differ between strategies (p > .05). Participants successfully identified the correct associative pair on visual trials (65% ± 4% SEM; t(14) = 8.034, p < .001) and on verbal trials (58% ± 3% SEM; t(14) = 8.527, p < .001) relative to chance performance of 33%. Of these visual correct trials, participants reported that they were “Very” confident on 39% (±5% SEM), “Somewhat” confident on 31% (±4% SEM), and “Not Very” confident on 29% (±3% SEM). Of these verbal correct trials, participants reported that they were “Very” confident on 20% (±3% SEM), “Somewhat” confident on 41% (±3% SEM), and “Not Very” confident on 38% (±4% SEM).

fMRI Analysis

Controlling the type of stimulus and focusing on successfully encoded associative pairs, the effects of using a visual versus a verbal strategy could be isolated. Brain regions were identified where the size of the BOLD response was greater when participants successfully encoded pairs of items with high confidence using the visual associative strategy relative to the verbal associative strategy and vice versa. Regions identified by this contrast (p < .05, corrected for multiple comparisons) are listed in Table 1. Greater response for visually relative to verbally encoded pairs was seen in bilateral middle frontal gyrus (BA 6), inferior and superior parietal cortices (BA 40 and BA 7), and lateral occipital cortex (BA 19/BA 39; Figure 2). There were a large number of voxels showing a greater response for verbally relative to visually encoded pairs at this threshold causing single clusters to encompass multiple peaks of activation; for illustration, a threshold of p < .001, corrected for multiple comparisons, was used to highlight distinct foci of activity for this contrast. Regions identified at this threshold are also listed in Table 1. Greater response for verbally relative to visually encoded pairs was seen in the left inferior frontal (BA 45) and precentral (BA 4) gyri or adjacent sulci and bilateral frontal operculum (BA 45/BA 47/BA 13), medial frontal cortex (BA 6), and medial occipital cortex (BA 19, BA 18, BA 30/BA 18; Figure 3). Given that the type of stimulus was the same for both conditions and the only difference was the use of a visual or verbal associative strategy, these differences in regional activity for associative encoding are, therefore, driven by the utilized strategy.

Table 1. 

Regional Effects of Associative Strategy on High-confidence Encoding

Region
BAVolumea
x
y
z
t
Visual > Verbal* 
L/R middle frontal gyrus 1088 −26 −9 56 3.46 
1088 22 −9 56 3.52 
L/R inferior parietal lobule 40 1856 58 −29 40 4.56 
40 3136 −54 −29 36 4.26 
40 1472 34 −41 40 3.32 
L/R superior parietal lobule 1920 −18 −73 52 3.79 
1216 14 −65 52 3.29 
L/R lateral occipital cortex 39 6144 46 −69 24 5.80 
19 5824 −30 −81 36 4.52 
 
Verbal > Visual* 
R cingulate 30 326592 22 −57 7.01 
Subclusters at higher threshold** 
  L inferior frontal gyrus 45 1344 −42 19 6.69 
  L/R frontal operculum 47/13 576 −30 23 5.80 
45/13 320 30 27 5.39 
  B medial frontal gyrus 896 −2 −5 64 6.56 
  L precentral gyrus 832 −50 −9 48 5.45 
  L postcentral gyrus 256 −26 −37 60 4.61 
  L/R medial occipital cortex 30/18 8320 22 −57 7.01 
19 448 −26 −61 5.66 
19 320 −18 −65 5.47 
18 256 −22 −73 16 5.16 
  L/R cingulate 31 320 −2 −17 44 5.26 
30 256 22 −49 12 5.11 
  L white matter – 576 −30 −49 6.00 
– 256 −30 −41 20 5.18 
– 256 −30 −37 24 5.35 
R cerebellum – 3328 30 −65 −48 4.88 
– 896 26 −33 −24 3.29 
Brain stem – 1664 −2 −29 −32 3.31 
Region
BAVolumea
x
y
z
t
Visual > Verbal* 
L/R middle frontal gyrus 1088 −26 −9 56 3.46 
1088 22 −9 56 3.52 
L/R inferior parietal lobule 40 1856 58 −29 40 4.56 
40 3136 −54 −29 36 4.26 
40 1472 34 −41 40 3.32 
L/R superior parietal lobule 1920 −18 −73 52 3.79 
1216 14 −65 52 3.29 
L/R lateral occipital cortex 39 6144 46 −69 24 5.80 
19 5824 −30 −81 36 4.52 
 
Verbal > Visual* 
R cingulate 30 326592 22 −57 7.01 
Subclusters at higher threshold** 
  L inferior frontal gyrus 45 1344 −42 19 6.69 
  L/R frontal operculum 47/13 576 −30 23 5.80 
45/13 320 30 27 5.39 
  B medial frontal gyrus 896 −2 −5 64 6.56 
  L precentral gyrus 832 −50 −9 48 5.45 
  L postcentral gyrus 256 −26 −37 60 4.61 
  L/R medial occipital cortex 30/18 8320 22 −57 7.01 
19 448 −26 −61 5.66 
19 320 −18 −65 5.47 
18 256 −22 −73 16 5.16 
  L/R cingulate 31 320 −2 −17 44 5.26 
30 256 22 −49 12 5.11 
  L white matter – 576 −30 −49 6.00 
– 256 −30 −41 20 5.18 
– 256 −30 −37 24 5.35 
R cerebellum – 3328 30 −65 −48 4.88 
– 896 26 −33 −24 3.29 
Brain stem – 1664 −2 −29 −32 3.31 

aCluster volumes (mm3); coordinates correspond to the voxel of maximum intensity for each cluster.

*p < .05, corrected for multiple comparisons.

**p < .001, corrected for multiple comparisons.

Figure 2. 

Visual associative encoding engages bilateral middle frontal, lateral parietal, and lateral occipital regions. Statistical activation maps for regions showing increased activity during highly confident associative encoding using the visual strategy relative to the verbal strategy (p < .05, corrected for multiple comparisons) are overlaid on the pial surface of the Talairach and Tournoux N27 average brain and on a coronal slice of a mean anatomical scan image across all 15 participants. Functional clusters located in bilateral middle frontal (BA 6), inferior and superior parietal (BA 40 and BA 7), and lateral occipital (BA 19, BA 39) regions were used for time-course analyses. Graphs depict the time course of percent signal change in these regions for the visual and verbal associative encoding of object pairs subsequently recognized with high confidence. The error bars illustrate the SEM, the y axis represents the percent signal change, and the x axis represents time in seconds from stimulus pair onset.

Figure 2. 

Visual associative encoding engages bilateral middle frontal, lateral parietal, and lateral occipital regions. Statistical activation maps for regions showing increased activity during highly confident associative encoding using the visual strategy relative to the verbal strategy (p < .05, corrected for multiple comparisons) are overlaid on the pial surface of the Talairach and Tournoux N27 average brain and on a coronal slice of a mean anatomical scan image across all 15 participants. Functional clusters located in bilateral middle frontal (BA 6), inferior and superior parietal (BA 40 and BA 7), and lateral occipital (BA 19, BA 39) regions were used for time-course analyses. Graphs depict the time course of percent signal change in these regions for the visual and verbal associative encoding of object pairs subsequently recognized with high confidence. The error bars illustrate the SEM, the y axis represents the percent signal change, and the x axis represents time in seconds from stimulus pair onset.

Figure 3. 

Verbal associative encoding engages bilateral inferior frontal, medial frontal, and medial occipital regions. Statistical activation maps for regions showing increased activity during highly confident associative encoding using the verbal strategy relative to the visual strategy (p < .001, corrected for multiple comparisons) are overlaid on the pial surface of the Talairach and Tournoux N27 average brain and on a coronal slice of a mean anatomical scan image across all 15 participants. Functional clusters located in the left inferior frontal (BA 45), bilateral frontal operculum (BA 47/BA 13, BA 45/BA 13), left precentral (BA 4), bilateral medial frontal (BA 6), and bilateral medial occipital (BA 19, BA 18, BA 18/BA 30) regions were used for time course analyses. Graphs depict the time course of percent signal change in these regions for the visual and verbal associative encoding of object pairs subsequently recognized with high confidence. The error bars illustrate the SEM, the y axis represents the percent signal change, and the x axis represents time in seconds from stimulus pair onset.

Figure 3. 

Verbal associative encoding engages bilateral inferior frontal, medial frontal, and medial occipital regions. Statistical activation maps for regions showing increased activity during highly confident associative encoding using the verbal strategy relative to the visual strategy (p < .001, corrected for multiple comparisons) are overlaid on the pial surface of the Talairach and Tournoux N27 average brain and on a coronal slice of a mean anatomical scan image across all 15 participants. Functional clusters located in the left inferior frontal (BA 45), bilateral frontal operculum (BA 47/BA 13, BA 45/BA 13), left precentral (BA 4), bilateral medial frontal (BA 6), and bilateral medial occipital (BA 19, BA 18, BA 18/BA 30) regions were used for time course analyses. Graphs depict the time course of percent signal change in these regions for the visual and verbal associative encoding of object pairs subsequently recognized with high confidence. The error bars illustrate the SEM, the y axis represents the percent signal change, and the x axis represents time in seconds from stimulus pair onset.

Within-strategy subsequent memory effects were examined using a linear weighted model to isolate regions consistent with the following listing from least to greatest: incorrect, correct with low confidence, correct with medium confidence, and correct with high confidence (p < .05, corrected). Regions isolated using this model for visually and verbally encoded pairs are listed in Table 2. Under a visual associative strategy, positive subsequent memory effects were seen in the left inferior frontal (BA 9, BA 47) and middle/superior frontal (BA 6) gyri (Figure 4). Under a verbal associative strategy, positive subsequent memory effects were seen in the left medial frontal gyrus (BA 6), pFC, and medial occipital cortex (BA 18; Figure 5). Left prefrontal regions showed subsequent memory effects for both the visual and verbal encoding strategies. Whereas these regions had distinct foci of peak intensity, a conjunction analysis of the two conditions revealed three small regions of overlap (centered at coordinates −38, 8, 24; −46, 32, 6; and −50, 21, 0). Additionally, an ANOVA of the responses within the strategy-specific regions (functionally defined by the effect of strategy analysis, listed in Table 1) showed a Memory × Strategy interaction or trend within bilateral lateral occipital (left: F(1, 14) = 15.162, p < .01; right: F(1, 14) = 6.434, p < .05), bilateral medial occipital (left: F(1, 14) = 3.918, p = .068; right: F(1, 14) = 3.996, p = .065), bilateral inferior frontal (left: F(1, 14) = 3.686, p = .075; right: F(1, 14) = 7.108, p < .05), and bilateral medial frontal (F(1, 14) = 4.042, p = .064) regions. Although brain regions with positive subsequent memory effects differed between strategies, deactivations predictive of subsequent memory were similar for both strategies and were located in brain regions commonly classified as “default mode network” regions, including ventral medial pFC, posterior cingulate, inferior parietal lobule, and lateral temporal cortex (see Buckner, Andrews-Hanna, & Schacter, 2008, for thorough discussion of this brain network).

Table 2. 

Regional Subsequent Memory Effects within a Visual or Verbal Associative Strategy

Region
BAVolumea
x
y
z
t
Visual Subsequent Memory Regions* 
L inferior frontal gyrus 47 2176 −42 35 3.40 
832 −34 11 24 2.91 
L middle/superior frontal gyrus 2688 −26 11 52 4.88 
 
Verbal Subsequent Memory Regions* 
L prefrontal cortex 45 15232 −26 35 16 7.32 
L medial frontal gyrus 1216 −6 15 44 4.41 
L medial occipital cortex 18 2240 −10 −69 −4 3.35 
L superior temporal gyrus 13 17600 −42 −45 20 5.50 
R insula 13 1536 26 27 12 3.77 
– 1728 38 −65 −28 3.37 
R white matter – 20032 30 −45 5.65 
– 2048 22 11 24 3.50 
– 1152 26 −5 24 3.33 
Region
BAVolumea
x
y
z
t
Visual Subsequent Memory Regions* 
L inferior frontal gyrus 47 2176 −42 35 3.40 
832 −34 11 24 2.91 
L middle/superior frontal gyrus 2688 −26 11 52 4.88 
 
Verbal Subsequent Memory Regions* 
L prefrontal cortex 45 15232 −26 35 16 7.32 
L medial frontal gyrus 1216 −6 15 44 4.41 
L medial occipital cortex 18 2240 −10 −69 −4 3.35 
L superior temporal gyrus 13 17600 −42 −45 20 5.50 
R insula 13 1536 26 27 12 3.77 
– 1728 38 −65 −28 3.37 
R white matter – 20032 30 −45 5.65 
– 2048 22 11 24 3.50 
– 1152 26 −5 24 3.33 

aCluster volumes (mm3); coordinates correspond to the voxel of maximum intensity for each cluster; only positive activations listed.

*p < .05, corrected for multiple comparisons.

Figure 4. 

Visual strategy influences regional subsequent memory effects. Statistical activation maps for regions identified using a linear weighted model of increasing subsequent memory confidence, from subsequently forgotten to subsequently recognized with high-confidence pairs, under the visual associative encoding strategy (p < .05, corrected for multiple comparisons) are overlaid on the pial surface of the Talairach and Tournoux N27 average brain. Functional clusters located in the left inferior frontal (BA 47, 9) and left middle/superior frontal (BA 6) regions were used for time course analyses. Graphs depict the time course of percent signal change in these regions for trials in which visually encoded object pairs were identified correctly with high confidence, correctly with medium confidence, correctly with low confidence, or incorrectly; the average response for all verbal trials is also depicted. The error bars illustrate the SEM, the y axis represents the percent signal change, and the x axis represents time in seconds from stimulus pair onset.

Figure 4. 

Visual strategy influences regional subsequent memory effects. Statistical activation maps for regions identified using a linear weighted model of increasing subsequent memory confidence, from subsequently forgotten to subsequently recognized with high-confidence pairs, under the visual associative encoding strategy (p < .05, corrected for multiple comparisons) are overlaid on the pial surface of the Talairach and Tournoux N27 average brain. Functional clusters located in the left inferior frontal (BA 47, 9) and left middle/superior frontal (BA 6) regions were used for time course analyses. Graphs depict the time course of percent signal change in these regions for trials in which visually encoded object pairs were identified correctly with high confidence, correctly with medium confidence, correctly with low confidence, or incorrectly; the average response for all verbal trials is also depicted. The error bars illustrate the SEM, the y axis represents the percent signal change, and the x axis represents time in seconds from stimulus pair onset.

Figure 5. 

Verbal strategy influences regional subsequent memory effects. Statistical activation maps for regions identified using a linear weighted model of increasing subsequent memory confidence, from subsequently forgotten to subsequently recognized with high-confidence pairs, under the verbal associative encoding strategy (p < .05, corrected for multiple comparisons) are overlaid on the pial surface of the Talairach and Tournoux N27 average brain. Functional clusters located in the left prefrontal, left medial frontal (BA 6), and left medial occipital (BA 18) regions were used for time-course analyses. Graphs depict the time course of percent signal change in these regions for trials in which verbally encoded object pairs were identified correctly with high confidence, correctly with medium confidence, correctly with low confidence, or incorrectly; the average response for all visual trials is also depicted. The error bars illustrate the SEM, the y axis represents the percent signal change, and the x axis represents time in seconds from stimulus pair onset.

Figure 5. 

Verbal strategy influences regional subsequent memory effects. Statistical activation maps for regions identified using a linear weighted model of increasing subsequent memory confidence, from subsequently forgotten to subsequently recognized with high-confidence pairs, under the verbal associative encoding strategy (p < .05, corrected for multiple comparisons) are overlaid on the pial surface of the Talairach and Tournoux N27 average brain. Functional clusters located in the left prefrontal, left medial frontal (BA 6), and left medial occipital (BA 18) regions were used for time-course analyses. Graphs depict the time course of percent signal change in these regions for trials in which verbally encoded object pairs were identified correctly with high confidence, correctly with medium confidence, correctly with low confidence, or incorrectly; the average response for all visual trials is also depicted. The error bars illustrate the SEM, the y axis represents the percent signal change, and the x axis represents time in seconds from stimulus pair onset.

Experiment 2

Behavioral Analysis

The stroke patient identified the correct pair in 77% of trials using her own strategy, 59% of trials using a verbal strategy, and 100% of trials using a visual strategy. The behavioral results are represented in Figure 6B. Practice effects were not evident for either the patient or controls across the eight retrieval runs; percent correct scores and ranges for each run are listed in Table 3. The healthy control participants correctly identified the pair in 99% of trials using their own strategy, 84% of trials using a verbal strategy, and 99% of trials using a visual strategy. The control participants' performance differed from the patent's performance in the “own” and “verbal” strategy conditions, but not in the “visual” strategy condition (own: t(11) = 31.15, p < .005; verbal: t(7) = 4.03, p < .005; visual: t(11) = −1.48, p > .1). Thus, in contrast to the controls, the patient was impaired when using the verbal strategy as well as her own natural strategy.

Figure 6. 

Encoding strategy influences memory performance in patient with focal brain damage. (A) Axial diffusion-weighted brain image for a patient following a stroke localized to the left inferior frontal lobe. (B) Graph depicts the patient's recognition memory performance for pairs encoded using her own strategy (light gray), the verbal strategy (white), and the visual strategy (dark gray). Relative to her memory performance when using her own strategy, the patient showed further impairment under the verbal strategy, but improvement under the visual strategy. The error bars illustrate the standard deviation, the y axis represents the percent of trials in which she selected the correct pair, and the horizontal line marks the chance level of 33% correct.

Figure 6. 

Encoding strategy influences memory performance in patient with focal brain damage. (A) Axial diffusion-weighted brain image for a patient following a stroke localized to the left inferior frontal lobe. (B) Graph depicts the patient's recognition memory performance for pairs encoded using her own strategy (light gray), the verbal strategy (white), and the visual strategy (dark gray). Relative to her memory performance when using her own strategy, the patient showed further impairment under the verbal strategy, but improvement under the visual strategy. The error bars illustrate the standard deviation, the y axis represents the percent of trials in which she selected the correct pair, and the horizontal line marks the chance level of 33% correct.

Table 3. 

Percent Correct Performance for Patient and Controls

Run
1
2
3
4
5
6
7
8
StrategyOwnOwnOwnVerbalVisualVisualVisualVerbal
Patient 81 75 75 56 100 100 100 63 
Controlsa 98 (94–100) 100 (100) 98 (94–100) 77 (56–100) 98 (94–100) 100 (100) 98 (94–100) 91 (69–100) 
Run
1
2
3
4
5
6
7
8
StrategyOwnOwnOwnVerbalVisualVisualVisualVerbal
Patient 81 75 75 56 100 100 100 63 
Controlsa 98 (94–100) 100 (100) 98 (94–100) 77 (56–100) 98 (94–100) 100 (100) 98 (94–100) 91 (69–100) 

aAverage and range of percent correct across thefour4 control participants.

DISCUSSION

The current study examined the effects of using a visual strategy compared with a verbal strategy for associative encoding. Participants were shown the same type of stimulus under both associative encoding strategies, which allowed isolation of regional brain responses because of the utilized strategy (Experiment 1). This design also allowed within-strategy subsequent memory effects to be compared for the two associative encoding strategies. On the basis of finding greater left inferior frontal involvement under a verbal relative to a visual strategy, a patient with a focal lesion localized to this region was tested on a modified behavioral version of this study (Experiment 2). This patient showed impaired performance when using both the verbal strategy and her own strategy relative to the visual strategy.

Strategy Effects

Prior studies examining the encoding of visual or verbal stimuli have reported different patterns of cortical activity for each stimulus type (Deshpande, Hu, Lacey, Stilla, & Sathian, 2010; Gottlieb, Uncapher, & Rugg, 2010; Lacey, Flueckiger, Stilla, Lava, & Sathian, 2010; Harrison & Tong, 2009; Hocking & Price, 2009; Uncapher & Wagner, 2009; Kirwan, Wixted, & Squire, 2008; Park & Rugg, 2008; Achim, Bertrand, Montoya, Malla, & Lepage, 2007; Ferber, Humphrey, & Vilis, 2005; Fletcher, Stephenson, Carpenter, Donovan, & Bullmorel, 2003; Bernstein, Beig, Siegenthaler, & Grady, 2002; Cansino, Maquet, Dolan, & Rugg, 2002; Fletcher et al., 2002; Rugg, Otten, & Henson, 2002; Baker, Sanders, Maccotta, & Buckner, 2001; Grill-Spector, Kourtzi, & Kanwisher, 2001; Rama, Sala, Gillen, Pekar, & Courtney, 2001; Iidaka, Sadato, Yamada, & Yonekura, 2000; Kohler, Moscovitch, Winocur, & McIntosh, 2000; Lee, Robbins, Pickard, & Owen, 2000; Prabhakaran, Narayanan, Zhao, & Gabrieli, 2000; Heun et al., 1999; Brewer, Zhao, Desmond, Glover, & Gabrieli, 1998; Grady, McIntosh, Rajah, & Craik, 1998; Wagner et al., 1998; Kapur et al., 1996; Demb et al., 1995; Kapur et al., 1994). Kim (2011) recently published a meta-analysis of 74 fMRI studies looking at subsequent memory effects for item or associative encoding of verbal or pictorial material. This meta-analysis indicated that the left inferior frontal gyrus was more active for encoding verbal material, whereas bilateral fusiform, occipital, hippocampal, and posterior parietal regions were more active for encoding pictorial material.

In accordance with such findings and those of Kirchhoff and Buckner (2006) described earlier, this study using direct manipulation of encoding instructions found distinct regions recruited for encoding visual versus verbal associations, although the same type of stimulus was presented under both conditions. Furthermore, these distinct regions were the same as those often reported when visual versus verbal stimuli are encoded. The present findings and those of Kirchhoff and Buckner (2006), therefore, suggest that differences in reported activity during studies using various stimulus types may be driven by the strategies used and not by the types of stimuli encountered. It should be noted, however, that participants likely used some degree of verbal processing during the visual strategy task and vice versa, given that the stimuli were inherently conducive to both visual and verbal processing. Therefore, despite primary engagement in the instructed strategy, some use of the uninstructed strategy would also be expected.

Although across all participants more pairs were correctly identified with high confidence for visual trials than for verbal trials, this effect was not global as there was also substantial individual variability across participants as to which strategy had more high-confidence correct trials. The bases of such individual variability in preferred strategy, a topic of great interest and applicability, remains unknown. Examining the neural bases of such effects was beyond the scope of this study and might benefit from directed recruitment of participants with strong predilections in either direction; nevertheless, future research might expand upon these findings and those of Kirchhoff and Buckner (2006) by examining whether individuals with strong asymmetry in visual versus verbal performance might also show engagement of regional brain activity that can be linked to their preferred strategy, even when encoding strategy is uncontrolled. Such a study, however, would require substantial task modifications as well as greater power to explore such fine-grated relationships between predilection for encoding strategy and brain activity. In this study, the between-strategy analysis used only trials matched across the two strategies for highest confidence, and accuracy for these judgments was high for both strategies. Additionally, comparisons across strategy within lower levels of confidence (i.e., medium confidence and low confidence) yielded patterns of brain activity that were qualitatively similar to the patterns seen in the high-confidence judgments. Thus, the differences in regional activity for associative encoding were likely driven by strategy and not by other factors related to confidence.

Within-strategy Subsequent Memory Effects

Within-strategy subsequent memory effects were also seen in regions more responsive to a particular strategy. Given that both encoding conditions involved associating the same type of stimulus, distinct subsequent memory effects can be attributed to the participants' use of different instructed strategies. Whereas most effects were in regions more responsive for that strategy, some were in regions selective for the other strategy. For example, left inferior frontal regions were more active during successful verbal, than visual, encoding (Figure 3) but showed subsequent memory effects within both strategies (Figures 4 and 5).

The current findings support the claim that some cortical regions engaged during processing of an event are also important for its successful encoding (Rugg et al., 2002). Otten and Rugg (2001) helped to elucidate the influence task has on brain regions recruited during episodic memory encoding. Their study found that incidental encoding of single words engaged distinct regions when participants made animacy versus syllable judgments about the words, possibly reflecting the effects of level of processing within the verbal domain. The present findings identify additional influences on the neural correlates of episodic encoding and extend what is known about subsequent memory activity by showing how the use of different domain-based processing strategies to associatively encode visual stimuli can engage diverse subsequent memory networks.

Implications for Encoding Studies

Evidence that using different encoding strategies influences regional responses, even when the stimuli remain constant, has important implications for designing and interpreting studies of successful memory formation. Without instruction, participants likely use a strategy related to the type of stimuli they are encoding; this explanation fits with the patterns of cortical activity reported when studies use visual versus verbal stimuli. Participants' individual predilection for one strategy over another may also have an effect and is a topic for further examination, but such individual differences are expected to have minimal impact on group data as more robust group effects dominate the resulting activations. In the absence of explicit strategy instruction, strategy remains influenced by stimulus type or modality.

Examining brain activity in response to the use of different encoding strategies is an important area of research; however, the current findings have cautionary implications. Many studies have examined different levels of processing, that is, deep versus shallow encoding or relational versus item encoding. If the instructions for the two types of processing differ in their demand on using visual or verbal encoding strategies, there could be confounding effects between strategy use and level of processing. Given the current findings that explicit strategy use dominates over implicit effects of the presented stimulus type, such potential confounds are an important consideration for designing future studies.

Potential Clinical Applications

Impairments that result from neurological disease or damage can extend beyond primary cognitive findings to impact memory. Once the affected brain regions are identified, an important component of rehabilitation is targeting and retraining intact brain regions and pathways to reorganize and compensate for the impairment. The processes and treatments involved in functional reorganization of motor pathways have been extensively studied in patients after suffering a stroke (Ward, 2004). Theoretically, the same principles could be applied to rehabilitating patients with damage in non-motor brain regions that cause different functional deficits. For example, following focal damage or a stroke to the left inferior frontal lobe, a patient is likely aphasic, and although other brain regions remain unaffected, this patient can have disproportionate difficulty remembering grocery lists or verbal directions. However, results from Experiment 1 suggest that brain regions engaged during an encoding task are not locked to the stimulus type but are flexible and can be controlled by top–down influences of the employed strategy. Therefore, individuals might improve memory performance by using a strategy that avoids the damaged tissue.

This hypothesis was tested in a patient with a recent stroke localized to the left inferior frontal lobe. This patient exhibited memory impairment when using her own associative encoding strategy, and this deficit became more pronounced when she used a verbal strategy. Greater impairment with using a verbal strategy was not surprising given the finding of enhanced left inferior frontal engagement during verbal associative encoding in Experiment 1. However, when she was instructed to abandon the verbal strategy and only use a visual strategy, her performance improved to 100% for all three runs. This improvement occurred despite her reluctance to use a strategy she considered suboptimal for a “verbal person” like herself. Her impaired memory performance when using her own strategy was possibly due to continued use of a strategy that was no longer ideal given her brain damage. By abandoning the verbal strategy, she no longer relied on the integrity of this brain region and showed marked improvement in her memory encoding ability, bringing her performance to the level of controls. Regardless of the specific elements of the patient's natural associative encoding strategy, the directed visual strategy afforded improvement in her memory performance, reflecting compensation perhaps through engagement of unaffected parietal regions.

The results of the first experiment elucidate how engagement in a particular associative strategy dictates which brain regions are recruited for encoding. Such findings are relevant to both past and future investigation into regional brain involvement in encoding. In addition, the outcome of the second experiment suggests additional implications of these findings for extending beyond basic research and toward developing possible targets within clinical treatment. It, therefore, may be valuable to provide patients with adaptive memory strategies that avoid damaged brain regions and engage spared regions.

Acknowledgments

This work was supported by the National Institute of Neurological Disorders and Stroke (K23 NS050305) and the University of California, San Diego Departments of Neurosciences and Radiology. J. B. H. was supported by the National Science Foundation through the Graduate Research Fellowship Program.

Reprint requests should be sent to James B. Brewer, Human Memory Laboratory, 8950 Villa La Jolla Drive C212, La Jolla, CA 92037, or via e-mail: jbrewer@ucsd.edu.

REFERENCES

REFERENCES
Achim
,
A. M.
,
Bertrand
,
M. C.
,
Montoya
,
A.
,
Malla
,
A. K.
, &
Lepage
,
M.
(
2007
).
Medial temporal lobe activations during associative memory encoding for arbitrary and semantically related object pairs.
Brain Research
,
1161
,
46
55
.
Baker
,
J. T.
,
Sanders
,
A. L.
,
Maccotta
,
L.
, &
Buckner
,
R. L.
(
2001
).
Neural correlates of verbal memory encoding during semantic and structural processing tasks.
NeuroReport
,
12
,
1251
1256
.
Bernstein
,
L. J.
,
Beig
,
S.
,
Siegenthaler
,
A. L.
, &
Grady
,
C. L.
(
2002
).
The effect of encoding strategy on the neural correlates of memory for faces.
Neuropsychologia
,
40
,
86
98
.
Brewer
,
J. B.
,
Zhao
,
Z.
,
Desmond
,
J. E.
,
Glover
,
G. H.
, &
Gabrieli
,
J. D.
(
1998
).
Making memories: Brain activity that predicts how well visual experience will be remembered.
Science
,
281
,
1185
1187
.
Buckner
,
R. L.
,
Andrews-Hanna
,
J. R.
, &
Schacter
,
D. L.
(
2008
).
The brain's default network: Anatomy, function, and relevance to disease.
Annals of the New York Academy of Sciences
,
1124
,
1
38
.
Cansino
,
S.
,
Maquet
,
P.
,
Dolan
,
R. J.
, &
Rugg
,
M. D.
(
2002
).
Brain activity underlying encoding and retrieval of source memory.
Cerebral Cortex
,
12
,
1048
1056
.
Cox
,
R. W.
(
1996
).
AFNI: Software for analysis and visualization of functional magnetic resonance neuroimages.
Computers in Biomedical Research
,
29
,
162
173
.
Dale
,
A. M.
(
1999
).
Optimal experimental design for event-related fMRI.
Human Brain Mapping
,
8
,
109
114
.
Dale
,
A. M.
, &
Buckner
,
R. L.
(
1997
).
Selective averaging of rapidly presented individual trials using fMRI.
Human Brain Mapping
,
5
,
329
340
.
Demb
,
J. B.
,
Desmond
,
J. E.
,
Wagner
,
A. D.
,
Vaidya
,
C. J.
,
Glover
,
G. H.
, &
Gabrieli
,
J. D.
(
1995
).
Semantic encoding and retrieval in the left inferior prefrontal cortex: A functional MRI study of task difficulty and process specificity.
Journal of Neuroscience
,
15
,
5870
5878
.
Deshpande
,
G.
,
Hu
,
X.
,
Lacey
,
S.
,
Stilla
,
R.
, &
Sathian
,
K.
(
2010
).
Object familiarity modulates effective connectivity during haptic shape perception.
Neuroimage
,
49
,
1991
2000
.
Dickerson
,
B. C.
,
Bakkour
,
A.
,
Salat
,
D. H.
,
Feczko
,
E.
,
Pacheco
,
J.
,
Greve
,
D. N.
,
et al
(
2009
).
The cortical signature of Alzheimer's disease: Regionally specific cortical thinning relates to symptom severity in very mild to mild AD dementia and is detectable in asymptomatic amyloid-positive individuals.
Cerebral Cortex
,
19
,
497
510
.
Ferber
,
S.
,
Humphrey
,
G. K.
, &
Vilis
,
T.
(
2005
).
Segregation and persistence of form in the lateral occipital complex.
Neuropsychologia
,
43
,
41
51
.
Fletcher
,
P. C.
,
Palomero-Gallagher
,
N.
,
Zafiris
,
O.
,
Fink
,
G. R.
,
Tyler
,
L. K.
, &
Zilles
,
K.
(
2002
).
The influence of explicit instructions and stimulus material on lateral frontal responses to an encoding task.
Neuroimage
,
17
,
780
791
.
Fletcher
,
P. C.
,
Stephenson
,
C. M.
,
Carpenter
,
T. A.
,
Donovan
,
T.
, &
Bullmorel
,
E. T.
(
2003
).
Regional brain activations predicting subsequent memory success: An event-related fMRI study of the influence of encoding tasks.
Cortex
,
39
,
1009
1026
.
Gottlieb
,
L. J.
,
Uncapher
,
M. R.
, &
Rugg
,
M. D.
(
2010
).
Dissociation of the neural correlates of visual and auditory contextual encoding.
Neuropsychologia
,
48
,
137
144
.
Grady
,
C. L.
,
McIntosh
,
A. R.
,
Rajah
,
M. N.
, &
Craik
,
F. I.
(
1998
).
Neural correlates of the episodic encoding of pictures and words.
Proceedings of the National Academy of Sciences, U.S.A.
,
95
,
2703
2708
.
Grill-Spector
,
K.
,
Kourtzi
,
Z.
, &
Kanwisher
,
N.
(
2001
).
The lateral occipital complex and its role in object recognition.
Vision Research
,
41
,
1409
1422
.
Hales
,
J. B.
, &
Brewer
,
J. B.
(
2001
).
The timing of associative memory formation: Frontal lobe and anterior medial temporal lobe activity at associative binding predicts memory.
Journal of Neurophysiology
,
105
,
1454
1463
.
Harrison
,
S. A.
, &
Tong
,
F.
(
2009
).
Decoding reveals the contents of visual working memory in early visual areas.
Nature
,
458
,
632
635
.
Heun
,
R.
,
Klose
,
U.
,
Jessen
,
F.
,
Erb
,
M.
,
Papassotiropoulos
,
A.
,
Lotze
,
M.
,
et al
(
1999
).
Functional MRI of cerebral activation during encoding and retrieval of words.
Human Brain Mapping
,
8
,
157
169
.
Hocking
,
J.
, &
Price
,
C. J.
(
2009
).
Dissociating verbal and nonverbal audiovisual object processing.
Brain and Language
,
108
,
89
96
.
Iidaka
,
T.
,
Sadato
,
N.
,
Yamada
,
H.
, &
Yonekura
,
Y.
(
2000
).
Functional asymmetry of human prefrontal cortex in verbal and non-verbal episodic memory as revealed by fMRI.
Brain Research, Cognitive Brain Research
,
9
,
73
83
.
Kapur
,
S.
,
Craik
,
F. I.
,
Tulving
,
E.
,
Wilson
,
A. A.
,
Houle
,
S.
, &
Brown
,
G. M.
(
1994
).
Neuroanatomical correlates of encoding in episodic memory: Levels of processing effect.
Proceedings of the National Academy of Sciences, U.S.A.
,
91
,
2008
2011
.
Kapur
,
S.
,
Tulving
,
E.
,
Cabeza
,
R.
,
McIntosh
,
A. R.
,
Houle
,
S.
, &
Craik
,
F. I.
(
1996
).
The neural correlates of intentional learning of verbal materials: A PET study in humans.
Brain Research, Cognitive Brain Research
,
4
,
243
249
.
Kim
,
H.
(
2011
).
Neural activity that predicts subsequent memory and forgetting: A meta-analysis of 74 fMRI studies.
Neuroimage
,
54
,
2446
2461
.
Kirchhoff
,
B. A.
, &
Buckner
,
R. L.
(
2006
).
Functional-anatomic correlates of individual difference in memory.
Neuron
,
51
,
263
274
.
Kirwan
,
C. B.
,
Wixted
,
J. T.
, &
Squire
,
L. R.
(
2008
).
Activity in the medial temporal lobe predicts memory strength, whereas activity in the prefrontal cortex predicts recollection.
Journal of Neuroscience
,
28
,
10541
10548
.
Kohler
,
S.
,
Moscovitch
,
M.
,
Winocur
,
G.
, &
McIntosh
,
A. R.
(
2000
).
Episodic encoding and recognition of pictures and words: Role of the human medial temporal lobes.
Acta Psychologica
, (Amsterdam),
105
,
159
179
.
Kriegeskorte
,
N.
,
Simmons
,
W. K.
,
Bellgowan
,
P. S.
, &
Baker
,
C. I.
(
2009
).
Circular analysis in systems neuroscience: The dangers of double dipping.
Nature Neuroscience
,
12
,
535
540
.
Lacey
,
S.
,
Flueckiger
,
P.
,
Stilla
,
R.
,
Lava
,
M.
, &
Sathian
,
K.
(
2010
).
Object familiarity modulates the relationship between visual object imagery and haptic shape perception.
Neuroimage
,
49
,
1977
1990
.
Lee
,
A. C.
,
Robbins
,
T. W.
,
Pickard
,
J. D.
, &
Owen
,
A. M.
(
2000
).
Asymmetric frontal activation during episodic memory: The effects of stimulus type on encoding and retrieval.
Neuropsychologia
,
38
,
677
692
.
Listerud
,
J.
,
Powers
,
C.
,
Moore
,
P.
,
Libon
,
D. J.
, &
Grossman
,
M.
(
2009
).
Neuropsychological patterns in magnetic resonance imaging-defined subgroups of patients with degenerative dementia.
Journal of the International Neuropsychological Society
,
15
,
459
470
.
Miller
,
M. I.
,
Beg
,
M. F.
,
Ceritoglu
,
C.
, &
Stark
,
C.
(
2005
).
Increasing the power of functional maps of the medial temporal lobe by using large deformation diffeomorphic metric mapping.
Proceedings of the National Academy of Sciences, U.S.A.
,
102
,
9685
9690
.
Mummery
,
C. J.
,
Patterson
,
K.
,
Price
,
C. J.
,
Ashburner
,
J.
,
Frackowiak
,
R. S.
, &
Hodges
,
J. R.
(
2000
).
A voxel-based morphometry study of semantic dementia: Relationship between temporal lobe atrophy and semantic memory.
Annals of Neurology
,
47
,
36
45
.
Otten
,
L. J.
, &
Rugg
,
M. D.
(
2001
).
Task-dependency of the neural correlates of episodic encoding as measured by fMRI.
Cerebral Cortex
,
11
,
1150
1160
.
Park
,
H.
, &
Rugg
,
M. D.
(
2008
).
Neural correlates of successful encoding of semantically and phonologically mediated inter-item associations.
Neuroimage
,
43
,
165
172
.
Prabhakaran
,
V.
,
Narayanan
,
K.
,
Zhao
,
Z.
, &
Gabrieli
,
J. D.
(
2000
).
Integration of diverse information in working memory within the frontal lobe.
Nature Neuroscience
,
3
,
85
90
.
Rama
,
P.
,
Sala
,
J. B.
,
Gillen
,
J. S.
,
Pekar
,
J. J.
, &
Courtney
,
S. M.
(
2001
).
Dissociation of the neural systems for working memory maintenance of verbal and nonspatial visual information.
Cognitive, Affective & Behavioral Neuroscience
,
1
,
161
171
.
Rossion
,
B.
, &
Pourtois
,
G.
(
2004
).
Revisiting Snodgrass and Vanderwart's object pictorial set: The role of surface detail in basic-level object recognition.
Perception
,
33
,
217
236
.
Rugg
,
M. D.
,
Otten
,
L. J.
, &
Henson
,
R. N.
(
2002
).
The neural basis of episodic memory: Evidence from functional neuroimaging.
Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences
,
357
,
1097
1110
.
Smith
,
S. M.
,
Jenkinson
,
M.
,
Woolrich
,
M. W.
,
Beckmann
,
C. F.
,
Behrens
,
T. E.
,
Johansen-Berg
,
H.
,
et al
(
2004
).
Advances in functional and structural MR image analysis and implementation as FSL.
Neuroimage
,
23(Suppl. 1)
,
S208
S219
.
Talairach
,
J.
, &
Tournoux
,
P.
(
1998
).
A co-planar stereotaxic atlas of the human brain.
Stuttgart
:
Thieme
.
Uncapher
,
M. R.
, &
Wagner
,
A. D.
(
2009
).
Posterior parietal cortex and episodic encoding: Insights from fMRI subsequent memory effects and dual-attention theory.
Neurobiology of Learning and Memory
,
91
,
139
154
.
Wagner
,
A. D.
,
Poldrack
,
R. A.
,
Eldridge
,
L. L.
,
Desmond
,
J. E.
,
Glover
,
G. H.
, &
Gabrieli
,
J. D.
(
1998
).
Material-specific lateralization of prefrontal activation during episodic encoding and retrieval.
NeuroReport
,
9
,
3711
3717
.
Ward
,
N. S.
(
2004
).
Functional reorganization of the cerebral motor system after stroke.
Current Opinion in Neurology
,
17
,
725
730
.