Abstract

Goal-directed behavior lowers activity in brain areas that include the medial frontal cortex, the medial and lateral parietal cortex, and limbic and paralimbic brain regions, commonly referred to as the “default network.” These activity decreases are believed to reflect the interruption of processes that are ongoing when the mind is in a restful state. Previously, the nature of these processes was probed by varying cognitive task parameters, but the presence of emotional processes, while often assumed, was little investigated. With fMRI, we studied the effect of systematic variations of both cognitive load and emotional stimulus connotation on task-related decreases in the default network by employing an auditory working memory (WM) task with musical sounds. The performance of the WM task, compared to passive listening, lowered the activity in medial and lateral, prefrontal, parietal, temporal, and limbic regions. In a subset of these regions, the magnitude of decrease depended on the memory load; the greater the cognitive load, the larger the magnitude of the observed decrease. Furthermore, in the right amygdala and the left precuneus, areas previously associated with processing of unpleasant dissonant musical sounds, there was an interaction between the experimental condition and the stimulus type. The current results are consistent with the previously reported effect of task difficulty on task-related brain activation decreases. The results also indicate that task-related decreases may be further modulated by the emotional stimulus connotation.

INTRODUCTION

The current study addresses the question of the relationship between stimulus-independent activity during situations characterized by rest, also commonly referred to as default activity or resting state activity, and stimulus-related goal-directed activity in the human brain. During selective attention and stimulus-related cognitive operations, the default activity of the brain decreases in several cortical, limbic, and paralimbic locations. These decreases, which are believed to reflect the interruption of processes that are characteristic of a task-independent state (Gusnard & Raichle, 2001; Raichle et al., 2001; Shulman et al., 1997), could be of two kinds: (i) task-dependent decreases in areas not engaged in the current task, such as regions processing input generated by other stimulus modalities (Laurienti et al., 2002; Lewis, Beauchamp, & DeYoe, 2000; Haxby et al., 1994), or (ii) task-independent decreases across different types of tasks and stimuli (first described by Shulman et al., 1997). Brain regions representative of the posited task-independent decreases include the posterior cingulate gyrus, precuneus, lateral parietal regions, medial and orbital prefrontal cortex, anterior cingulate gyrus, inferior frontal cortex, and lateral frontal regions (Tomasi, Ernst, Caparelli, & Chang, 2006; McKiernan, Kaufman, Kucera-Thompson, & Binder, 2003; Gusnard, Akbudak, Shulman, & Raichle, 2001; Mazoyer et al., 2001; Binder et al., 1999; Petersen, van Mier, Fiez, & Raichle, 1998; Raichle, 1998; Shulman et al., 1997; Haxby et al., 1994; Raichle et al., 1994).

Psychological evidence suggests that the mind, in the absence of external stimulation, is occupied with the stimulus-independent thoughts of “mind-wandering” (Teasdale et al., 1995; Antrobus, Singer, & Greenberg, 1966) that may involve cognitive operations similar to those employed during the execution of exogenously imposed tasks (Teasdale et al., 1995). Hence, stimulus-independent thoughts may be the psychological baseline of the resting mind (Smallwood & Schooler, 2006). To further probe the nature of these processes, task-related decreases were studied during different types of tasks with a main focus on the variation of cognitive parameters. Evidence from brain imaging studies include semantic processes (McKiernan et al., 2003; Binder et al., 1999) and episodic memory (Mazoyer et al., 2001) in this baseline state. Another view has as its main premise the observation that consistent task-related decreases do not reflect specific activations (defined as activity elevation compared to the average brain activity during rest) in the resting state. This insight led to the formulation of a theory of a “default mode of brain function” that is functionally engaged in the general mental task of monitoring the external and internal environment. The default mode according to this theory critically involves the posterior cingulate gyrus (BA 31), the precuneus (BA 7), the medial prefrontal cortex (PFC, BAs 10, 32), the lateral parietal cortex (BAs 39, 40) (Raichle et al., 2001), and, as a more recent addition, the hippocampal formation (Vincent et al., 2006, 2007; Greicius, Srivastava, Reiss, & Menon, 2004).

Task-related decreases of activity in medial prefrontal and limbic cortices suggest that processing of self-referential and emotional materials occurs generally when more pressing tasks are not imposed from the outside (Gusnard et al., 2001; Raichle et al., 2001). Simpson, Snyder, Gusnard, and Raichle (2001) tested the influence of the emotional state of subjects performing a cognitive task on the magnitude of decreases in the medial PFC, a brain region assigned a general role in emotion processing (Steele & Lawrie, 2004; Phan, Wager, Taylor, & Liberzon, 2002). Task-related decreases of activity in the medial PFC (BAs 9, 10 and 32) were less when performance anxiety accompanied a word-generation task (compared to passive vision). The attenuation suggested an interaction between increased emotional activity and decreased cognitive activity in this region (Simpson et al., 2001).

The amygdala is critically involved in the experience of emotional valence (Habel, Klein, Kellermann, Shah, & Schneider, 2005; Kim et al., 2004; Anderson & Phelps, 2001; Blood & Zatorre, 2001; Isenberg et al., 1999; Morris et al., 1996, 1998; Bechara, Tranel, Damasio, & Damasio, 1996; Adolphs, Tranel, Damasio, & Damasio, 1994; LeDoux, Cicchetti, Xagoraris, & Romanski, 1990). During processing of emotionally charged visual stimuli, the amygdaloid responses were lowered during naming compared to superficial matching (Hariri, Mattay, Tessitore, Fera, & Weinberger, 2003; Hariri, Bookheimer, & Mazziotta, 2000), during rating compared to recognition (Liberzon et al., 2000), during explicit compared to implicit processing, and during the task of describing negative visual scenes in unemotional terms (Ochsner, Bunge, Gross, & Gabrieli, 2002). There is also evidence of increased responses in the amygdala during introspective emotion reappraisal compared to passive (task unrelated) perception (Ochsner, Ray, et al., 2004).

Temporary reallocation of resources saved by the task-related decrease of brain activity implies competition (McKiernan et al., 2003; Drevets & Raichle, 1998; Haxby et al., 1994) and predicts that the cognitive demand dictates the magnitude of the decrease. In a meta-analysis, Shulman et al. (1997) found no evidence that matched this prediction, but other studies found evidence of reduced activity of the default network in proportion to task difficulty during auditory discrimination and presentation rate detection (McKiernan et al., 2003), and visual tracking and visual working memory [WM] (Tomasi et al., 2006).

Here we test the hypothesis that both the cognitive load and the emotional stimulus connotation modulate the pattern of task-related activity decreases, especially that the emotional induction prevents the reassignment of neuronal resources that would otherwise be disengaged by a cognitive task. We designed an auditory experiment with systematic variation of the cognitive load and the emotional stimulus connotation. The stimuli were major, minor, and dissonant musical chords, characterized by happy, sad, and unpleasant emotional connotations, respectively (Pallesen, Brattico, & Carlson, 2003; Trainor & Heinmiller, 1998; Zentner & Kagan, 1998; Crowder, Reznick, & Rosenkrantz, 1991; Crowder, 1985). The emotional unpleasantness, or dissonance, of the sound stimuli had three levels, with major and dissonant chords representing the extremes. Processing of dissonance in musical stimuli was previously associated with the amygdala (Gosselin, Peretz, Johnsen, & Adolphs, 2007; Koelsch, Fritz, von Cramon, Muller, & Friederici, 2006), the parahippocampal gyrus (Gosselin et al., 2006; Koelsch et al., 2006), and the precuneus (Blood, Zatorre, Bermudez, & Evans, 1999). Hence, we predicted that cortical and limbic brain areas are deactivated during auditory cognition compared to a resting state of passive perception, and specifically, (i) that the decreases during the cognitive task depend on the cognitive load, and (ii) that the activity decrease in the amygdala, parahippocampal gyrus, and precuneus would be smaller in response to the negatively charged than to the positively charged auditory stimuli, resulting from the hypothesized countermodulation of task-related decreases by the emotional stimulus properties. Three experimental conditions varied the cognitive load: (1) passive listening (PL), (2) 1-back pitch WM (1B), and (3) 2-back pitch WM (2B). Because the subjects were required to perform a cognitive task on the chord stimuli, we thought that the emotional connotations implicitly would enter all experimental conditions. To allow wide generalization of the results, we presented the chords to musically untrained subjects while fMRI recorded the responses.

METHODS

Subjects

The 10 volunteers (age range = 22–31 years, mean = 25 years, 5 women) had minimal musical training, obtained exclusively as obligatory primary school music education. All subjects were right-handed and had no history of neurological disease or hearing deficit. The volunteers gave informed consent to the study as approved by the ethics committee of the Helsinki University Central Hospital.

Stimuli

The stimuli were nine sound combinations (chords) belonging to three different chord categories according to the Western tonal music theory (“major,” “minor,” and “dissonant”), and each spanning three frequency levels separated by an octave (frequency ratio 1:2, in musical notation the lowest pitches of the chords were A3, A4, and A5; see Figure 1). Each chord was produced with the grand-piano (Piano 1) timbre of the Roland Sound Canvas SC-50 synthesizer with built-in samples, and played using the ENCORE software. The chords were then edited by CoolEdit and SoundForge programs to be balanced in the loudness level and have the same duration (870 msec). The relatively long duration for single piano chords was chosen to maximize the emotional effects. The major chords consisted of A, C♯, E, A, C♯ and, as such, were characterized mostly by consonant intervals. The minor chords consisted of A, C, E, A, C, including the minor third interval, which is considered an imperfect consonance in music theory (cf. Schön, Regnault, Ystad, & Besson, 2005). The dissonant chords were made of A, B♭, G, A♭, C, including a minor second, the interval considered the most dissonant in the literature (Fishman et al., 2001; Kameoka & Kuriyagawa, 1969; Plomp & Levelt, 1965), and several other dissonant intervals.

Figure 1. 

The stimuli were nine piano chords belonging to three different chord categories: major, minor, and dissonant. Each chord spanned three frequency levels separated by an octave; the lowest pitches of the chords were A3, A4, and A5. The duration of the chords was 870 msec.

Figure 1. 

The stimuli were nine piano chords belonging to three different chord categories: major, minor, and dissonant. Each chord spanned three frequency levels separated by an octave; the lowest pitches of the chords were A3, A4, and A5. The duration of the chords was 870 msec.

Experimental Conditions

The three experimental conditions included two n-back task conditions in which subjects memorized the pitch of the chords, and a passive listening condition in which subjects listened to the stimuli with no cognitive evaluation. The subjects responded to each stimulus by pressing the left or right button of a response pad, with their right index and middle finger, respectively. In the 1B task, subjects pressed the left button when the pitch of the chord was the same as in the previous trial and in the 2B task when the pitch was the same as presented two trials back. The right button was pressed in all other trials. In the PL condition, subjects pressed the right button after each chord.

Magnetic Resonance Imaging

Magnetic resonance imaging measurements were performed in the Department of Radiology, Helsinki University Hospital, on a Siemens Sonata 1.5-T scanner. A 3-D T1-weighted anatomical image was acquired of each subject with the Magnetization Prepared RApid Gradient Echo sequence [MP-RAGE; TR = 1900 msec, TE = 3.86 msec, TI = 1100 msec, flip angle = 15°, FOV = 256 mm, 256 × 256 mm2 matrix, 176 sagittal slices, 1 mm3 isotropic voxel dimension]. For the functional imaging, a T2*-weighted, gradient-echo echo-planar sequence was used [EPI; TR = 3660 msec, TE = 40 msec, flip angle = 90°, FOV = 224 mm, 64 × 64 mm2 matrix, 4 mm slice thickness (no gaps), 36 slices]. A single functional volume was acquired in 2790 msec, introducing a period of scanner silence (870 msec; no gradient noise) during which the stimuli were presented. The slices were oriented axially and placed according to an initial scout image to cover the whole brain and the cerebellum.

Experimental Procedure

Subjects were informed of the experimental conditions with written and oral descriptions. They practiced the button presses in the scanner room until they felt confident in performing the WM tasks. With respect to the PL condition, subjects were told to rest their mind from the task while still pressing a button after each stimulus. The chords were presented binaurally with MR-compatible head phones (Commander XG, Resonance Technology). The chords were played at an intensity of approximately 80 dB, individually adjusted, so that subjects could clearly hear the sounds and did not feel any related discomfort. In order to assure optimal perception of the sounds, stimulus presentation was interleaved with image acquisition. The experiment was divided into two sessions, each session containing 18 blocks of 20 trials. Each trial consisted of a stimulus (870 msec) followed by scanning (2760 msec) and silence (30 msec), with a total trial duration of 3660 msec. The blocks lasted 73.2 sec and were separated by 4-sec intervals. In addition to the task condition (PL, 1B, or 2B), each block was characterized by a chord category (i.e., major, minor, or dissonant). Thus, there were nine types of blocks: (1) PL, major; (2) 1B, major; (3) 2B, major; (4) PL, minor; (5) 1B, minor; (6) 2B, minor; (7) PL, dissonant; (8) 1B, dissonant; (9) 2B, dissonant. Each block was presented four times in a counterbalanced design. An instruction figure (“Passive Listening,” “1-Back,” or “2-Back”) was presented on a screen for a period of 4 sec before each block. The subjects viewed the instructions through a mirror placed on the head coil. During task performance, subjects were asked to fixate on a central cross on the screen. Between the two sessions there was a 2-min break, during which subjects stayed quietly in the scanner but were allowed to move the eyes freely or close them. A short sound at the end of the break told subjects to be ready for the next session. Image acquisition followed immediately after each stimulus. The imaging part lasted 54 min, and the anatomical scan lasted approximately 7 min, thus the total time in the scanner was approximately one hour.

A questionnaire and a behavioral test followed the imaging part. The questionnaire involved evaluation of memory strategies by ticking one or several of the following options: auditory rehearsal, verbal rehearsal, visual imagery, somatosensory imagery, movement, no specific strategy. The task difficulty level was evaluated as: 1 = very easy, 2 = easy, 3 = intermediate, 4 = difficult, and 5 = very difficult. In the behavioral test, subjects were asked to rate the emotional connotation of each stimulus (the major, minor, and dissonant chords) along a happy–sad scale and along a pleasant–unpleasant scale. Each scale had 11 values, from −5 to +5, with −5 denoting the most negative rating and +5 the most positive rating, and zero denoting “neutral.” The chords were rated twice, with chords presented in a randomized sequence.

Analysis of Behavioral Data

All statistical analyses were performed using the R project for statistical computing (www.r-project.org/). Two-way repeated measures analysis of variance (RM-ANOVA) was applied to the log-transformed reaction time (RT) data, with task condition and chord category as within-subject factors, and subject treated as a random effect. The RM-ANOVA was implemented as a linear mixed effects model (R function lme), and t tests were used post hoc to assess significant differences within factors. Task performance was measured as the ratio (r) of correct responses over total (n) responses. The ratio was then transformed using the following formula: r′ = 2 × π × arcsin(p), where p = 1/(4 × n), if r = 0; p = r, if 0 < r < 1; p = (n − ¼)/n, if r = 1. The arcsine transformation is used to homogenize the variance of a binomial response variable (r). The transformed ratios (r′) and the emotional chord ratings were subjected to the same statistical test (RM-ANOVA) as the log-transformed RTs.

Analysis of fMRI Data

All analysis of functional and anatomical data was performed using the FMRIB Software Library (FSL, version 3.2b), Oxford Centre for Functional Magnetic Resonance Imaging of the Brain, UK (www.fmrib.ox.ac.uk/fsl/). Nonbrain tissue was removed from the T1-weighted anatomical images using the Brain Extraction Tool (BET) of FSL (Smith, 2002). The MNI/ICBM-152 average brain was used as the standard stereotaxic space template in group analyses (Collins et al., 1998; Collins, Neelin, Peters, & Evans, 1994; Evans & Collins, 1993). Each individual's brain volume was coregistered to the template using affine transformations (12 degrees of freedom) estimated by FMRIB's Linear Registration Tool, FLIRT (Jenkinson & Smith, 2001).

Functional data were processed using FMRIB's Expert Analysis Tool (FEAT). During the experiment, 896 volumes were collected, of which the first five were discarded allowing T1 effects in the signal to saturate. The functional volumes were realigned to the midpoint of the experiment using rigid-body transformations (MCFLIRT; Jenkinson, Bannister, Brady, & Smith, 2002). Spatial smoothing was performed using a low-pass Gaussian filter with a full width at half maximum of 8 mm. A piecewise linear temporal high-pass filter (longest period passed: 805 sec) was used to remove low-frequency components of the data.

The general linear model (GLM) implementation of FEAT was used for model fitting of preprocessed data. The design matrix consisted of nine columns representing each of the possible blocks: permutations of memory load (2B, 1B, or PL) and chord category (major, minor, or dissonant). All columns were convolved with the “canonical” double-gamma hemodynamic response function. Autocorrelations in the model fit residuals were estimated and removed using the FILM prewhitening step in FEAT (Woolrich, Ripley, Brady, & Smith, 2001). First-level contrast images between the PL and WM conditions were calculated separately within each chord category: PL (major) versus 2B (major), 1B (dissonant) versus 2B (dissonant), and so forth. In addition, the regressors for each of the nine stimulation blocks were contrasted to the implicit baseline of the study, the instruction periods between blocks, in order to evaluate effects of chord type on the neuronal responses.

Group results were obtained using full mixed effects (ME) modeling, thus allowing generalization to the subject population. Higher-level parameter estimates and the ME variance were estimated implicitly within FEAT using FMRIB's Local Analysis of Mixed Effects (FLAME) (Woolrich, Behrens, Beckmann, Jenkinson, & Smith, 2004; Beckmann, Jenkinson, & Smith, 2003). The inputs to group analyses consisted of first-level contrast images categorized by chord, for instance, PL (major) versus 2B (major), PL (minor) versus 2B (minor), and PL (dissonant) versus 2B (dissonant) for each subject. One-sample t tests were implemented across subjects and chord types to test for specific load effects (nonzero means) in the planned contrasts. In addition, we implemented an “omnibus” ANOVA model in FEAT at the group level with the factors “load” and “chord,” each at three levels. The inputs to the analysis were the nine regressors per subject representing the experimental blocks.

SPMs of the group contrasts of interest were created for “Gaussianized” t values, the so-called Z-scores. To control for the effects of multiple comparisons (type I error), Gaussian random field (GRF) theory was employed to assign p values to SPM peaks surviving the Z > 3.0 threshold (Friston, Worsley, Frackowiak, Mazziotta, & Evans, 1994; Worsley, Evans, Marrett, & Neelin, 1992). All results and conclusions are based on clusters that reached at least the p < .05, corrected significance level. Tables of local Z-score maxima produced by FSL for each contrast were translated to anatomical names using a structural parcellation of the MNI single-subject brain (Collins et al., 1998), and the extended naming procedure described in Tzourio-Mazoyer et al. (2002).

Chord category-specific effects were evaluated at standard-space coordinates selected as described in Results (cf. Table 1). A spherical ROI of 8 mm radius was created, centered on each coordinate, and resampled to each subject's functional data space using transformations estimated during intrasubject analysis. Mean percent BOLD signal changes were then extracted for the nine contrasts representing the experimental blocks (vs. instruction baseline). Two-way RM-ANOVA was used for statistical analysis, implemented as described above for the RT data.

Table 1. 

Group Activations for PL versus 2B Contrast; List of Local Maxima

Region of Activation
Hemisphere
Coordinates (MNI)
Peak Z Score
x
y
z
Medial superior frontal gyrus left −8 58 18 7.28 
right 62 8.58 
Orbital middle frontal gyrus right 56 −2 8.10 
Gyrus rectus left −4 28 −12 8.03 
Precuneus left −8 −52 16 7.21 
Middle cingulate gyrus left −10 −44 36 7.12 
Posterior cingulate gyrus left −44 30 6.39 
Angular gyrus left −46 −68 30 7.08 
Calcarine gyrus left −6 −64 22 6.89 
Rolandic operculum right 58 −4 10 6.57 
Superior frontal gyrus left −16 30 56 6.55 
Middle temporal gyrus right 56 −4 −22 6.51 
Precentral gyrus right 28 −26 74 6.26 
Lingual gyrus right −50 6.26 
Inferior temporal gyrus right 52 −6 −30 6.22 
Fusiform gyrus left −24 −38 −14 6.14 
right 28 −32 −20 6.22 
Temporal pole left −36 10 −38 6.20 
right 44 18 −36 6.21 
Hippocampus right 26 −10 −18 6.12 
Orbital inferior frontal gyrus left −32 34 −12 6.03 
Postcentral gyrus right 36 −30 70 6.01 
Middle frontal gyrus left −26 22 46 5.88 
Parahippocampal gyrus left −26 −20 −20 5.75 
Insula (long gyrus) right 38 −16 5.69 
Superior temporal/inferior parietal gyrus right 54 −30 22 5.49 
Amygdala right 28 −20 4.95 
Region of Activation
Hemisphere
Coordinates (MNI)
Peak Z Score
x
y
z
Medial superior frontal gyrus left −8 58 18 7.28 
right 62 8.58 
Orbital middle frontal gyrus right 56 −2 8.10 
Gyrus rectus left −4 28 −12 8.03 
Precuneus left −8 −52 16 7.21 
Middle cingulate gyrus left −10 −44 36 7.12 
Posterior cingulate gyrus left −44 30 6.39 
Angular gyrus left −46 −68 30 7.08 
Calcarine gyrus left −6 −64 22 6.89 
Rolandic operculum right 58 −4 10 6.57 
Superior frontal gyrus left −16 30 56 6.55 
Middle temporal gyrus right 56 −4 −22 6.51 
Precentral gyrus right 28 −26 74 6.26 
Lingual gyrus right −50 6.26 
Inferior temporal gyrus right 52 −6 −30 6.22 
Fusiform gyrus left −24 −38 −14 6.14 
right 28 −32 −20 6.22 
Temporal pole left −36 10 −38 6.20 
right 44 18 −36 6.21 
Hippocampus right 26 −10 −18 6.12 
Orbital inferior frontal gyrus left −32 34 −12 6.03 
Postcentral gyrus right 36 −30 70 6.01 
Middle frontal gyrus left −26 22 46 5.88 
Parahippocampal gyrus left −26 −20 −20 5.75 
Insula (long gyrus) right 38 −16 5.69 
Superior temporal/inferior parietal gyrus right 54 −30 22 5.49 
Amygdala right 28 −20 4.95 

RESULTS

Behavioral Results

Reaction Time and Task Performance

The task condition had a significant effect on RT [F(2, 6786) = 2107.62, p < .001], which increased as a function of cognitive load: The median 2B RT of 1001 msec exceeded the median 1B RT of 835 msec which, in turn, exceeded the median PL RT of 540 msec (PL vs. 1B: t = 45.58, p < .001; 1B vs. 2B: t = 17.24, p < .001). There was a significant effect of chord category [F(2, 6786) = 12.61, p < .001]: RTs were significantly shorter in response to dissonant chords (median 760 msec) than to both major (median 794 msec; t = 3.29; p < .01) and minor (median 798 msec; t = 3.40; p < .001) chords, whereas there was no significant difference of RTs to major and minor chords (t = 0.11, p > .05). Interactions between load and chord were not significant [F(4, 6782) = 0.91, p > .05], and were thus removed from the model. The effect of task condition on the RTs is illustrated in Figure 2, in which the effect of dissonance is present at all WM loads.

Figure 2. 

Detailed analysis of reaction times (RTs) as a function of WM load and chord type. The box-and-whisker plots illustrate the extent of variation of the RTs in the nine experimental conditions. Each box is drawn from the first (Q1) to the third (Q3) quartile around the median (solid horizontal line), and the whiskers extend to the robust extremes of the distribution, that is, 3/2 times the interquartile (Q1–Q3) range. The median values for each condition, and the across-chord averages at each load level, are shown above the plots. Significant differences are highlighted for the dissonant versus minor and dissonant versus major chord-type comparisons (*p < .05).

Figure 2. 

Detailed analysis of reaction times (RTs) as a function of WM load and chord type. The box-and-whisker plots illustrate the extent of variation of the RTs in the nine experimental conditions. Each box is drawn from the first (Q1) to the third (Q3) quartile around the median (solid horizontal line), and the whiskers extend to the robust extremes of the distribution, that is, 3/2 times the interquartile (Q1–Q3) range. The median values for each condition, and the across-chord averages at each load level, are shown above the plots. Significant differences are highlighted for the dissonant versus minor and dissonant versus major chord-type comparisons (*p < .05).

The mean error rates were 0.1%, 5.3%, and 10.4% for PL, 1B, and 2B, respectively. We used the arcsine-transformed ratio of correct-over-total responses (cf. Methods) to measure the effects of load and chord category on task performance. Neither the main effect of chord category [F(2, 72) = 0.18, p > .05] nor the Load-by-Chord interaction [F(4, 72) = 0.06, p > .05] significantly improved the ANOVA model fit, and both were removed from the regression. Providing evidence that the cognitive demands of the three WM conditions were indeed different, the main effect of WM load on performance was highly significant with an F ratio of F = 45.30, p < .001: Subjects performed better in PL than in 1B WM (t = 5.18, p < .001), and better in 1B than 2B (t = 4.32, p < .001).

Ratings of Difficulty Level and Employed Task Strategies

Nine of the 10 subjects rated the 1B task “very easy” or “easy” and 8 of 10 rated the 2B task as “difficult” or “intermediate.” Most subjects used several different task strategies (7/10 subjects). The “auditory rehearsal” strategy (explained to subjects as “keeping the tones in mind”) was used as a memory strategy by 8/10 subjects. “Verbal rehearsal” (explained to subjects as for example keeping in mind the tones as “low,” “middle,” or “high”) was used by 6/10 subjects. “Visual imagery” (explained to subjects as “for example imagining the high sound as being placed on a high location”) was used by 3/10 subjects.

Emotional Ratings of the Chords

During scanning, tasks were never related to the emotional connotations of the stimuli. After scanning, subjects' attention for the first time explicitly turned toward the emotional valence of the stimuli. As illustrated in Figure 3, chord type significantly influenced the perceived happiness of the stimuli [F(2, 75) = 34.70, p < .001]. Major chords were rated happier than both minor (t = 5.55, p < .001) and dissonant (t = 8.24, p < .001) chords, and minor chords were rated happier than dissonant chords (t = 2.72, p < .001). We also observed a significant effect of octave [F(2, 75) = 5.77, p < .005]: The fifth octave was judged happier than the third (the lowest in frequency; t = 3.38, p < .005).

Figure 3. 

Emotional ratings by chord type. Dissonant chords were judged more sad and unpleasant than both minor and major chords. Major chords were judged happier and more pleasant than minor chords. Bars represent the standard error (see Results section for statistical analysis).

Figure 3. 

Emotional ratings by chord type. Dissonant chords were judged more sad and unpleasant than both minor and major chords. Major chords were judged happier and more pleasant than minor chords. Bars represent the standard error (see Results section for statistical analysis).

The subjects' evaluations of the pleasantness of the stimuli were also affected by both chord type [F(2, 75) = 37.70, p < .001] and octave [F(2, 75) = 4.38, p < .05]. Major chords were rated more pleasant than minor (t = 2.37, p < .05) and dissonant (t = 8.35, p < .001) chords, and minor chords were rated more pleasant than dissonant chords (t = 6.03, p < .001). The fifth octave (the highest in frequency) was rated more pleasant than the third (t = 2.41, p < .05) and fourth (t = 2.71, p < .01) octave. The chord ratings confirm the distinct emotional connotations of the three chord types, which are not influenced by MR scanner noise (Pallesen et al., 2003).

Functional Imaging Results

Group-level parametric activation maps of the planned contrasts revealing WM load-dependent BOLD signal changes were calculated across chord categories. The group mean PL versus 2B contrast located the responses to the ventral and dorsal parts of the medial PFC (BAs 11, 32, 10, 9, 8), the middle/posterior cingulate gyrus (BAs 23/31), the precuneus (BA 7/31), the lateral parietal cortex (angular gyrus/BA 39), the temporal gyri (BAs 20, 21, 42), the temporal pole (BA 38), the Rolandic operculum (BA 43), the posterior insula, the hippocampal formation, and the right amygdala (Figure 4A, blue color scale and Table 1). The group mean PL versus 1B results (Table 2) mainly revealed the same areas found in the PL versus 2B contrast (Table 1), with the exception of the ventral medial PFC (gyrus rectus), which did not reach the significance criterion in the PL versus 1B contrast.

Figure 4. 

Significant BOLD signal modulations by WM. (A) WM increased (2B vs. PL; light-to-dark) and decreased (PL vs. 2B; light-to-dark) brain activity in distinct networks. (B) In a subset of the deactivated brain regions, the strength of down-regulation was WM load dependent (1B vs. 2B; green); (A and B) Activity maps were corrected for multiple comparisons at the cluster level (p < .05). The coordinate in MNI standard space is given below each image slice. The underlying structural image is the so-called MNI single-subject brain, which is displayed in radiological orientation.

Figure 4. 

Significant BOLD signal modulations by WM. (A) WM increased (2B vs. PL; light-to-dark) and decreased (PL vs. 2B; light-to-dark) brain activity in distinct networks. (B) In a subset of the deactivated brain regions, the strength of down-regulation was WM load dependent (1B vs. 2B; green); (A and B) Activity maps were corrected for multiple comparisons at the cluster level (p < .05). The coordinate in MNI standard space is given below each image slice. The underlying structural image is the so-called MNI single-subject brain, which is displayed in radiological orientation.

Table 2. 

Group Activations for PL versus 1B Contrast; List of Local Maxima

Region of Activation
Hemisphere
Coordinates (MNI)
Peak Z Score
x
y
z
Medial superior frontal gyrus left −2 62 14 7.35 
right 10 38 54 4.33 
Orbital middle frontal gyrus left −2 54 7.08 
right 44 −8 7.38 
Precuneus left −6 −52 22 7.14 
Middle cingulate gyrus left −2 −20 42 4.58 
Posterior cingulate gyrus left −46 32 7.04 
Angular gyrus left −40 −78 34 6.28 
Rolandic operculum right 52 −12 12 3.81 
Superior frontal gyrus left −14 58 28 5.84 
right 18 40 36 4.73 
Middle temporal gyrus left −48 −66 24 7.03 
Precentral gyrus right 18 −26 78 3.73 
Lingual gyrus right −46 5.17 
Inferior temporal gyrus left −48 −32 5.09 
Fusiform gyrus left −26 −38 −12 6.31 
right 28 −32 −18 5.17 
Temporal pole left −44 16 −34 5.00 
right 42 20 −38 4.96 
Hippocampus right 20 −8 −18 3.96 
Orbital inferior frontal gyrus left −40 28 −16 4.41 
Postcentral gyrus right 38 −30 70 4.25 
Middle frontal gyrus left −24 24 48 6.32 
Parahippocampal gyrus left −24 −20 −20 4.08 
right 18 −32 −12 4.38 
Insula (long gyrus) right 42 −8 −2 3.96 
Superior temporal/inferior parietal gyrus right 60 −10 10 4.04 
Amygdala right 34 −18 3.81 
Region of Activation
Hemisphere
Coordinates (MNI)
Peak Z Score
x
y
z
Medial superior frontal gyrus left −2 62 14 7.35 
right 10 38 54 4.33 
Orbital middle frontal gyrus left −2 54 7.08 
right 44 −8 7.38 
Precuneus left −6 −52 22 7.14 
Middle cingulate gyrus left −2 −20 42 4.58 
Posterior cingulate gyrus left −46 32 7.04 
Angular gyrus left −40 −78 34 6.28 
Rolandic operculum right 52 −12 12 3.81 
Superior frontal gyrus left −14 58 28 5.84 
right 18 40 36 4.73 
Middle temporal gyrus left −48 −66 24 7.03 
Precentral gyrus right 18 −26 78 3.73 
Lingual gyrus right −46 5.17 
Inferior temporal gyrus left −48 −32 5.09 
Fusiform gyrus left −26 −38 −12 6.31 
right 28 −32 −18 5.17 
Temporal pole left −44 16 −34 5.00 
right 42 20 −38 4.96 
Hippocampus right 20 −8 −18 3.96 
Orbital inferior frontal gyrus left −40 28 −16 4.41 
Postcentral gyrus right 38 −30 70 4.25 
Middle frontal gyrus left −24 24 48 6.32 
Parahippocampal gyrus left −24 −20 −20 4.08 
right 18 −32 −12 4.38 
Insula (long gyrus) right 42 −8 −2 3.96 
Superior temporal/inferior parietal gyrus right 60 −10 10 4.04 
Amygdala right 34 −18 3.81 

The opposite contrast, 2B versus PL, located responses to areas typical of WM tasks with a high level of executive demand such as the n-back WM task, as documented in other studies employing it (Figure 4A, red color scale and Table 4) (Rämä et al., 2001), including the dorsolateral PFC (BAs 6, 9), the superior and inferior lateral parietal cortex (BAs 7, 40), the striatum, the thalamus, the anterior insula, and the cerebellum.

To test the hypothesis of load-dependent attenuation of responses during WM, we determined the group-mean 1B versus 2B contrast across chord categories. Several brain regions were more active during 1B than 2B tasks, including the medial PFC (BAs 11, 25, 32, 10, 9), the cingulate gyrus/paracentral lobule (BA 23/31), the lateral parietal cortex (angular gyrus/BA 39), the temporal gyri (BAs 20, 21, 41/42), the temporal pole (BA 38), the Rolandic operculum (BA 43), and the posterior insula (Figure 4B and Table 3).

Table 3. 

Group Activations for 1B versus 2B Contrast; List of Local Maxima

Region of Activation
Hemisphere
Coordinates (MNI)
Peak Z Score
x
y
z
Gyrus rectus left −2 42 −16 5.05 
right 10 30 −14 3.89 
Medial superior frontal gyrus left −2 48 26 3.87 
right 52 20 4.64 
Precentral gyrus right 26 −28 74 4.51 
Rolandic operculum left −44 −24 18 4.03 
right 42 −14 20 4.38 
Middle cingulate gyrus left −2 −24 50 4.31 
right −20 48 3.89 
Postcentral gyrus right 40 −32 68 4.26 
Inferior temporal gyrus right 54 −6 −28 4.25 
left −40 −38 3.73 
Insula (long gyri) left −38 −18 4.12 
right 36 −18 4.23 
Superior temporal gyrus (A1) left −54 −24 12 4.04 
Superior temporal/inferior parietal gyrus right 54 −30 22 4.23 
Middle temporal gyrus right 56 −6 −24 4.22 
Middle temporal pole left −34 10 −40 4.09 
right 44 −34 3.95 
Orbital middle frontal gyrus left −8 48 −4 3.97 
right 12 42 −12 4.04 
Anterior cingulate gyrus left −10 48 3.91 
Medial superior frontal gyrus right 54 3.89 
Supramarginal gyrus right 60 −30 26 3.87 
Supplementary motor area left −2 −16 52 3.86 
Putamen right 32 10 3.78 
Paracentral lobule left −28 60 3.60 
Region of Activation
Hemisphere
Coordinates (MNI)
Peak Z Score
x
y
z
Gyrus rectus left −2 42 −16 5.05 
right 10 30 −14 3.89 
Medial superior frontal gyrus left −2 48 26 3.87 
right 52 20 4.64 
Precentral gyrus right 26 −28 74 4.51 
Rolandic operculum left −44 −24 18 4.03 
right 42 −14 20 4.38 
Middle cingulate gyrus left −2 −24 50 4.31 
right −20 48 3.89 
Postcentral gyrus right 40 −32 68 4.26 
Inferior temporal gyrus right 54 −6 −28 4.25 
left −40 −38 3.73 
Insula (long gyri) left −38 −18 4.12 
right 36 −18 4.23 
Superior temporal gyrus (A1) left −54 −24 12 4.04 
Superior temporal/inferior parietal gyrus right 54 −30 22 4.23 
Middle temporal gyrus right 56 −6 −24 4.22 
Middle temporal pole left −34 10 −40 4.09 
right 44 −34 3.95 
Orbital middle frontal gyrus left −8 48 −4 3.97 
right 12 42 −12 4.04 
Anterior cingulate gyrus left −10 48 3.91 
Medial superior frontal gyrus right 54 3.89 
Supramarginal gyrus right 60 −30 26 3.87 
Supplementary motor area left −2 −16 52 3.86 
Putamen right 32 10 3.78 
Paracentral lobule left −28 60 3.60 

ROI Analysis

A chord category effect on emotional evaluations of the major, minor, and dissonant sounds was observed in the behavioral data (Figure 3). We wished to test whether neural correlates of this effect could be detected in the BOLD signal, and whether the effect depended on the WM load level. An initial “omnibus” ANOVA (cf. Methods) showed evidence only of a main effect of load—No significant activation clusters were detected for the main effect of chord or the Load-by-Chord interaction. We proceeded to select the regions (Table 1 and Figure 4A) with significant decreases during WM (i.e., the PL vs. 2B group mean contrast) for further ROI-based analysis (as described in Methods). The two-way RM-ANOVA performed on the right amygdala revealed a significant Load-by-Chord interaction [F(4, 70) = 3.54, p < .05]. One-way RM-ANOVAs with chord category as factor within each load level separately gave a significant chord category effect only in the passive listening condition [F(2, 18) = 4.01, p < .05]. During PL, a significantly larger BOLD response was observed to dissonant chords than to major chords (t = 2.79, p < .05), whereas the differences between dissonant and minor chords and major and minor chords were not significant. We observed a significant main effect of chord category in the left precuneus [F(2, 72) = 3.22, p < .05], where the BOLD signal exhibited significantly greater change during blocks of dissonant and minor stimuli than during major stimuli in both the PL [F(2, 18) = 4.19, p <.05; dissonant vs. major: t = 2.24, p < .05; minor vs. major: t = 2.71, p < .05] and the 2B [F(2, 18) = 5.09, p < .05; dissonant vs. major: t = 2.11, p < .05; minor vs. major: t = 3.13, p < .01] conditions. The BOLD signal changes in the right amygdala and left precuneus are summarized in Figure 5 with the location and size of each ROI.

Figure 5. 

Mean percent BOLD signal changes within precuneus and amygdala ROIs (illustrated in the center) as a function of WM load and chord category. In the amygdala (right), the BOLD signal was significantly modulated by chord category in the PL condition, but not in either of the WM conditions. In the precuneus (left), the BOLD signal was significantly modulated by chord category in the PL condition and in the 2B condition. Note the different scales in the left and right plots. Whiskers represent the standard error; *p < .05.

Figure 5. 

Mean percent BOLD signal changes within precuneus and amygdala ROIs (illustrated in the center) as a function of WM load and chord category. In the amygdala (right), the BOLD signal was significantly modulated by chord category in the PL condition, but not in either of the WM conditions. In the precuneus (left), the BOLD signal was significantly modulated by chord category in the PL condition and in the 2B condition. Note the different scales in the left and right plots. Whiskers represent the standard error; *p < .05.

DISCUSSION

By employing an experimental design that allowed controlled manipulations of cognitive load and emotional stimulus connotation, we tested the hypothesis that task-related decreases during a cognitive task depend on both the cognitive load and the emotional stimulus connotation. The results were consistent with the hypothesis by showing a cognitive load-dependent effect on the magnitude of the decreases. As expected, the task of memorizing the pitch of musical chords deactivated medial and paleocortical regions of the brain: the posterior cingulate gyrus, precuneus, medial PFC, lateral parietal cortex, and hippocampal formation (Table 1 and Table 2), that is, areas that are core regions of the default processing mode (Raichle & Snyder, 2007). The WM load-dependent decreases (1B vs. 2B contrast) mainly occurred in areas outside key components of the default system. The results also indicated an interaction between the experimental condition and the emotional stimulus connotation in the amygdala and the precuneus, both regions linked to processing of dissonance in addition to their role in default operations.

Working Memory Down-regulates Activity in the Default Network

The finding of deactivation of the medial/posterior cingulate gyrus (BAs 23/31), the precuneus (BA 7), the angular gyrus (BA 39), and regions in the medial and orbital PFC (BAs 11, 32, 10, 9, and 8) agrees with previous observations of deactivation in these areas during goal-directed behavior (McKiernan et al., 2003; Mazoyer et al., 2001; Raichle et al., 2001; Shulman et al., 1997). By comparing task-related decreases during three different auditory tasks performed on tone stimuli (a discrimination task, a stimulus presentation task, and a short-term memory task), McKiernan et al. (2003) noted an influence of the task on the deactivation pattern, which was replicated by Tomasi et al. (2006), who found deactivation in partially overlapping nets including the cingulate gyrus (BA 24, 31), the insula (BA 13), the postcentral gyrus (BA 30), and the precuneus (BA 7) in a visual attention tracking task compared to passive viewing, and a visual WM n-back task compared to a passive reaction task (Tomasi et al., 2006). Although the current study describes a unique pattern of decreases, it shares the majority of deactivated regions with the aforementioned studies. These results include decreases in areas related to visual processing, as found in previous auditory studies (McKiernan et al., 2003; Laurienti et al., 2002; Lewis et al., 2000). The pattern of decrease in visual areas during auditory processing clearly depends on specific task characteristics. In our study, the task instructions were mediated visually, and the behavioral results indicated that visual imagery was part of the task strategies used by approximately one third of the subjects. Also observed in our study, responses in the superior temporal/inferior parietal and middle temporal gyri were enhanced during low cognitive load. It may be noted that the 2B versus PL contrast also included responses in these regions (Table 4), however, at more posterior locations. A response in the anterior part of the inferior temporal gyrus was also enhanced during low cognitive load.

Table 4. 

Group Activations for 2B versus PL Contrast; List of Local Maxima

Region of Activation
Hemisphere
Coordinates (MNI)
Peak Z Score
x
y
z
Inferior parietal gyrus left −48 −44 54 8.49 
right 44 −50 48 7.54 
Superior parietal gyrus left −30 −64 50 8.10 
right 34 −64 56 6.20 
Precentral gyrus left −50 42 7.90 
right 48 34 6.93 
Middle frontal gyrus left −26 54 7.34 
right 28 −2 54 8.30 
Middle frontal gyrus left −46 26 34 7.19 
right 42 36 26 7.89 
Middle cingulate gyrus right 14 46 7.39 
Inferior frontal operculum left −54 10 16 6.71 
right 46 24 7.33 
Superior frontal gyrus left −26 −6 62 7.31 
Supplementary motor area left −4 60 6.31 
Cerebellum lobule 8 right 30 −66 −58 7.29 
Insula (short gyri) left −34 24 −2 7.09 
right 34 24 −4 6.83 
Precuneus left −6 −64 54 6.24 
right 12 −68 54 6.86 
Cerebellum crus 1 left −32 −62 −30 6.69 
right 30 −58 −34 6.95 
Cerebellum lobule 6 right 28 −62 −30 6.79 
Caudate left −16 −10 22 6.47 
Thalamus left −10 −18 14 6.35 
Cerebellum crus 2 right −76 −28 6.22 
Putamen left −18 12 5.79 
Superior temporal gyrus left −54 −44 18 4.08 
Middle temporal gyrus left −54 −50 16 3.93 
Region of Activation
Hemisphere
Coordinates (MNI)
Peak Z Score
x
y
z
Inferior parietal gyrus left −48 −44 54 8.49 
right 44 −50 48 7.54 
Superior parietal gyrus left −30 −64 50 8.10 
right 34 −64 56 6.20 
Precentral gyrus left −50 42 7.90 
right 48 34 6.93 
Middle frontal gyrus left −26 54 7.34 
right 28 −2 54 8.30 
Middle frontal gyrus left −46 26 34 7.19 
right 42 36 26 7.89 
Middle cingulate gyrus right 14 46 7.39 
Inferior frontal operculum left −54 10 16 6.71 
right 46 24 7.33 
Superior frontal gyrus left −26 −6 62 7.31 
Supplementary motor area left −4 60 6.31 
Cerebellum lobule 8 right 30 −66 −58 7.29 
Insula (short gyri) left −34 24 −2 7.09 
right 34 24 −4 6.83 
Precuneus left −6 −64 54 6.24 
right 12 −68 54 6.86 
Cerebellum crus 1 left −32 −62 −30 6.69 
right 30 −58 −34 6.95 
Cerebellum lobule 6 right 28 −62 −30 6.79 
Caudate left −16 −10 22 6.47 
Thalamus left −10 −18 14 6.35 
Cerebellum crus 2 right −76 −28 6.22 
Putamen left −18 12 5.79 
Superior temporal gyrus left −54 −44 18 4.08 
Middle temporal gyrus left −54 −50 16 3.93 

The results are evidence of a stronger representation of limbic regions than seen in previous studies of task-related decreases. In addition to the frequently reported medial/posterior cingulate gyrus, medial and orbital PFC, our findings include the amygdala, the hippocampus and parahippocampal gyrus, the anterior cingulate gyrus, the insula, and temporal poles. The task-related decreases in multiple limbic structures may reflect the emotionally charged stimuli used in the present study. The same limbic areas previously were linked to processing of music (Gosselin et al., 2005, 2006, 2007; Mitterschiffthaler, Fu, Dalton, Andrew, & Williams, 2007; Koelsch et al., 2006; Khalfa, Schon, Anton, & Liegeois-Chauvel, 2005; Brown, Martinez, & Parsons, 2004; Blood & Zatorre, 2001; Blood et al., 1999). However, because no significant effect of chord type was found in most of these regions, this claim must be tested further with a greater number of subjects. Another past finding is that of decreases in limbic areas during processing of nonemotional stimuli, such as the consistent deactivation of the right amygdala in visual tasks in the meta-analysis by Shulman et al. (1997) and the parahippocampal gyrus during a visual n-back task compared to a passive reaction task (Tomasi et al., 2006), and during a tone task compared to rest (Binder et al., 1999), suggesting that activity in these limbic regions automatically is suspended in favor of other processing. Decreased activity in the parahippocampal gyrus and uncus during a tone WM task compared to passive listening can result from a higher level of automatic episodic memory encoding during passive listening than during WM (Zatorre, Evans, & Meyer, 1994). In other words, an automatic associative memorization of the stimuli as a whole would be contrasted with the goal-directed encoding. This interpretation is consistent with the suggestion that certain cognitive processes are prevalent at rest (Mazoyer et al., 2001; Binder et al., 1999). The present task-related decreases in the hippocampus and parahippocampal gyrus are relevant in relation to the functional connections of the hippocampal formation to the parietal cortex, both as members of the default network, associated with episodic memory processes (Vincent et al., 2006).

Effect of Working Memory Load on the Task-related Decreases

The present finding of significant task difficulty-dependent decreases adds to previous findings (Tomasi et al., 2006; McKiernan et al., 2003). As expected, the brain areas in which the metabolic signal varies inversely with the WM load formed a subset of the areas deactivated in the PL versus 2B contrast. McKiernan et al. (2003) reported an interaction between the difficulty level of their short-term memory tone task and the magnitude of deactivation (based on ROI analysis) in the right precuneus/superior parietal lobule, the right anterior cingulate gyrus, the left anterior cingulate/superior frontal gyrus, the left middle frontal gyrus, and the right posterior cingulate gyrus. Tomasi et al. (2006) reported that enhanced brain deactivation in the cingulate gyrus and insula followed parametric increases in visual WM load (based on ROI analysis). Our present results of the 1B versus 2B contrast, examining the entire brain, provides confirming evidence to the above findings and extends the putative network exhibiting WM load-dependent task-related decreases by several brain regions previously undetected in the ROI analyses.

Of note, the WM load-dependent decreases found in our study, by contrasting directly the easy 1B with the demanding 2B task, were not significant in the posterior cingulate gyrus (although in neighboring middle cingulate gyrus), precuneus, lateral parietal cortex, and the hippocampal formation. Thus, the change in WM load did not significantly affect the task-related decreases in these areas, suggested to be key components of the default mode of brain function (Raichle et al., 2001).

Effect of Stimulus Type on the Task-related Decreases

A strong motivation for the current study was the claim that the emotional stimulus connotation present during cognition may counteract task-related decreases. The behavioral data showed distinct effects of chord dissonance reflected both in the ratings of the emotional connotations of the chords and in the RTs, which were significantly shorter to dissonant chords than to major and minor chords in all the experimental conditions (Figure 2). The shorter RTs to dissonant chords may reflect previously observed distressed arousal related to increased motor activity (Zentner & Kagan, 1998). The presence of this differential behavioral effect of chord types on RT not only during PL but also during WM indicates that the neural coding of emotional stimulus aspects may be upheld in structures outside the network that is generally down-regulated by cognition. In support of this suggestion, we previously found differential responses to the chord types during PL in brain areas including, in addition to the amygdala, the retrosplenial cortex, the brainstem, and the cerebellum (Pallesen et al., 2005). We therefore expected the perceived unpleasantness of the dissonant chords to evoke the strongest emotional responses, also during cognition.

A significant Load-by-Chord BOLD signal interaction was present in the right amygdala and the left precuneus, two of the three regions hypothesized to mediate interaction effects. In the amygdala, the differential response as a function of chord type was significant during the PL condition, which required mere button presses after each stimulus. These chord-dependent responses were, however, absent in both the 1B and 2B tasks. Consistent with this observation, the pattern of task-related decreases in the amygdala showed significant changes in both the PL versus 1B and PL versus 2B contrasts (Tables 1 and 2). This suggests that differential amygdala responses to the chord types were diminished already during a low WM load. A more complex response pattern was detected in the left precuneus, in which dissonant and minor chords elicited stronger BOLD responses than did major chords. The effect was present in both the PL and 2B conditions. This indicates that the precuneus, in contrast to the amygdala, may be a part of a brain network that retains a larger part of its processing capability and function during goal-directed behavior. The conclusion from these results remains circumstantial, as the hypothesis of a countermodulation of task-related decreases by emotional processes was only partly confirmed.

The processes that mediate task-related decreases are still debated. The current results suggest that task-related decreases happen gradually in some areas while in other areas the decreases may depend on their role in information processing in relation to the particular demands imposed by current cognitive operations. Attention resources may be the critical factor that determines the need for reprioritization of ongoing processes. Although this may indeed be the case, the specific course of the progression of task-related decreases also depends on the nature of the cognitive task in question. The n-back task has uneven jumps in attentional and, specifically, cognitive requirements, hence, a linear progression in the decreases from PL to 1B to 2B would seem unlikely. From the perspective of the amygdala and the precuneus, the suspension of emotion-related processes would also be expected to be nonlinear, as was indeed the case.

Previous findings of a countermodulating effect of subjects' emotional state on cognitive down-regulation in the ventral region of the medial PFC (Simpson et al., 2001), which is frequently activated during emotional processing (Phan et al., 2002), inspired the current investigation similar effects related to emotionally charged stimuli. Although an up-regulating effect of the negative emotional state on task-related decreases was found by Simpson et al. (2001), our study did not show a similar effect in relation to the dissonant chords, despite subjects characterizing them as unpleasant (Figure 3). A potential explanation suggested by Simpson et al. implies that the presence of the same emotional influences in control and active conditions minimizes the effect in the ventromedial PFC. We do not rule out the possibility that a negative emotional state during the WM task potentially canceled responses to negative emotions in the PL condition, because we did not specifically interview subjects about task-induced stress or anxiety during the active task. However, this possibility is remote, given the observation that processing of dissonance was affirmed by the increased activity of the amygdala during the PL versus 2B WM. Rather, self-referential processing may be the critical function that activates parts of the medial PFC during emotional tasks (Ochsner, Knierim, et al., 2004; Gusnard et al., 2001; Lane, Fink, Chau, & Dolan, 1997). A recent review assigned “the anterior region of the rostral medial PFC” (including approximately BAs 32, 10, 24, and the lower part of BA 9) to the mental operation of assessing self-knowledge (Amodio & Frith, 2006). A certain presentation duration of emotional stimuli, leading to the occurrence of implicit cognitive processes, rather than emotional processing per se, was also recently found to be a critical predictor of ventral medial PFC activity (Geday, Kupers, & Gjedde, 2007). In the present study, subjects did not explicitly reflect upon the passive perception of an emotional state, which led to fewer emotion-related responses that may rely on enhanced awareness of self.

Task-related Decreases and the Default Network

The formulation of the theory of a default system in the brain addressed brain regions characterized by consistent activity decreases during exogenously imposed tasks, or specific goal-directed behavior. The identity of the default system as a group of areas that are functionally connected is evident in studies of temporally coherent BOLD signal fluctuations at rest (Fox et al., 2005; Greicius, Krasnow, Reiss, & Menon, 2003) as well as in independent component analyses (Damoiseaux et al., 2006; Greicius & Menon, 2004). The membership of the default network has evolved somewhat and now includes the hippocampal formation as a candidate on the basis of the strength of functional connections to the posterior parietal cortex (Vincent et al., 2006, 2007; Greicius et al., 2003, 2004).

The functions of the default network, on the other hand, are controversial. The hippocampal–parietal link involved in episodic memory supports the claim that exogenously generated tasks and stimulus-independent thoughts have some cognitive operations in common, as in the case of the positive correlation between the degree of mind-wandering and BOLD responses in the default network (Mason et al., 2007). Other evidence, such as the finding of an analogue of the default network in the monkey brain (Vincent et al., 2007), invokes the more fundamental function of maintenance of information needed to interpret and respond to environmental demands (Raichle & Snyder, 2007).

The claim that regions that process emotions are tonically active at rest and more active at rest than during goal-directed behavior is consistent with a fundamental role of emotions in the direction of behavior (Damasio, 2000). The evaluation of the behavioral relevance of intrinsic operations of the brain is a challenge, but better understanding of processes active at rest makes it easier to interpret activity increases and decreases during exogenously generated tasks used in functional brain imaging (De Luca, Beckmann, De Stefano, Matthews, & Smith, 2006; McKiernan et al., 2003). Also, individual variation influences the observed brain responses and frequently is reflected in task-related activation as well as in intrinsic brain activity (Greicius & Menon, 2004).

In summary, the current findings support the hypothesis that the neural coding of emotional stimulus properties depends on the cognitive context, and is generally consistent with the theory of a dynamic balance between emotional and cognitive neural processes. We suggest that the specific neural signature of brain responses during mental operations depends intricately on the specific context of variables in ways that remain to be further investigated. Overall, the current findings are a strong basis for further testing of the hypothesis of countermodulation of task-related decreases by emotional stimulus connotation.

Acknowledgments

We thank Professor Synnöve Carlson for help with the design of the study and for reading and commenting on several drafts of the manuscript, Dr. Kim Mouridsen for assistance with statistical analyses, and Dr. Peter Vuust for assistance with the stimuli. We are also grateful to Professor Marcus Raichle for guidance related to the implications of the theory of a default mode of brain function.

Reprint requests should be sent to Karen Johanne Pallesen, Center of Functionally Integrative Neuroscience, Building 30 Aarhus University Hospitals, Norrebrogade 44, 8000 Aarhus C, Denmark or via e-mail: karenjohanne@pet.auh.dk; karenjohanne.pallesen@gmail.com.

REFERENCES

Adolphs
,
R.
,
Tranel
,
D.
,
Damasio
,
H.
, &
Damasio
,
A.
(
1994
).
Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala.
Nature
,
372
,
669
672
.
Amodio
,
D. M.
, &
Frith
,
C. D.
(
2006
).
Meeting of minds: The medial frontal cortex and social cognition.
Nature Reviews Neuroscience
,
7
,
268
277
.
Anderson
,
A. K.
, &
Phelps
,
E. A.
(
2001
).
Lesions of the human amygdala impair enhanced perception of emotionally salient events.
Nature
,
411
,
305
309
.
Antrobus
,
J. S.
,
Singer
,
J. L.
, &
Greenberg
,
S.
(
1966
).
Studies in the stream of consciousness: Experimental enhancement and suppression of spontaneous cognitive processes.
Perceptual and Motor Skills
,
23
,
399
417
.
Bechara
,
A.
,
Tranel
,
D.
,
Damasio
,
H.
, &
Damasio
,
A. R.
(
1996
).
Failure to respond autonomically to anticipated future outcomes following damage to prefrontal cortex.
Cerebral Cortex
,
6
,
215
225
.
Beckmann
,
C. F.
,
Jenkinson
,
M.
, &
Smith
,
S. M.
(
2003
).
General multilevel linear modeling for group analysis in FMRI.
Neuroimage
,
20
,
1052
1063
.
Binder
,
J. R.
,
Frost
,
J. A.
,
Hammeke
,
T. A.
,
Bellgowan
,
P. S.
,
Rao
,
S. M.
, &
Cox
,
R. W.
(
1999
).
Conceptual processing during the conscious resting state. A functional MRI study.
Journal of Cognitive Neuroscience
,
11
,
80
95
.
Blood
,
A. J.
, &
Zatorre
,
R. J.
(
2001
).
Intensely pleasurable responses to music correlate with activity in brain regions implicated in reward and emotion.
Proceedings of the National Academy of Sciences, U.S.A.
,
98
,
11818
11823
.
Blood
,
A. J.
,
Zatorre
,
R. J.
,
Bermudez
,
P.
, &
Evans
,
A. C.
(
1999
).
Emotional responses to pleasant and unpleasant music correlate with activity in paralimbic brain regions.
Nature Neuroscience
,
2
,
382
387
.
Brown
,
S.
,
Martinez
,
M. J.
, &
Parsons
,
L. M.
(
2004
).
Passive music listening spontaneously engages limbic and paralimbic systems.
NeuroReport
,
15
,
2033
2037
.
Collins
,
D. L.
,
Neelin
,
P.
,
Peters
,
T. M.
, &
Evans
,
A. C.
(
1994
).
Automatic 3D intersubject registration of MR volumetric data in standardized Talairach space.
Journal of Computer Assisted Tomography
,
18
,
192
205
.
Collins
,
D. L.
,
Zijdenbos
,
A. P.
,
Kollokian
,
V.
,
Sled
,
J. G.
,
Kabani
,
N. J.
,
Holmes
,
C. J.
et al
(
1998
).
Design and construction of a realistic digital brain phantom.
IEEE Transactions on Medical Imaging
,
17
,
463
468
.
Crowder
,
R. G.
(
1985
).
Perception of the major/minor distinction: III. Hedonic, musical and affective discriminations.
Bulletin of the Psychonomic Society
,
23
,
314
316
.
Crowder
,
R. G.
,
Reznick
,
J. S.
, &
Rosenkrantz
,
S. L.
(
1991
).
Perception of the major/minor distinction: V. Preferences among infants.
Bulletin of the Psychonomic Society
,
29
,
187
188
.
Damasio
,
A. R.
(
2000
).
The feeling of what happens: Body, emotion and the making of consciousness.
London
:
Vintage
.
Damoiseaux
,
J. S.
,
Rombouts
,
S. A.
,
Barkhof
,
F.
,
Scheltens
,
P.
,
Stam
,
C. J.
,
Smith
,
S. M.
et al
(
2006
).
Consistent resting-state networks across healthy subjects.
Proceedings of the National Academy of Sciences, U.S.A.
,
103
,
13848
13853
.
De Luca
,
M.
,
Beckmann
,
C. F.
,
De Stefano
,
N.
,
Matthews
,
P. M.
, &
Smith
,
S. M.
(
2006
).
fMRI resting state networks define distinct modes of long-distance interactions in the human brain.
Neuroimage
,
29
,
1359
1367
.
Drevets
,
W. C.
, &
Raichle
,
C. E.
(
1998
).
Reciprocal suppression of regional cerebral blood flow during emotional versus higher cognitive processes: Implications for interactions between cognition and emotion.
Cognition and Emotion
,
12
,
353
385
.
Evans
,
C.
, &
Collins
,
D. L.
(
1993
).
A 305-member MRI-based stereotactic atlas for CBF activation studies.
Paper presented at the 40th Annual Meeting of the Society for Nuclear Medicine, 8–11 June, Toronto, Canada.
Fishman
,
Y. I.
,
Volkov
,
I. O.
,
Noh
,
M. D.
,
Garell
,
P. C.
,
Bakken
,
H.
,
Arezzo
,
J. C.
et al
(
2001
).
Consonance and dissonance of musical chords: Neural correlates in auditory cortex of monkeys and humans.
Journal of Neurophysiology
,
86
,
2761
2788
.
Fox
,
M. D.
,
Snyder
,
A. Z.
,
Vincent
,
J. L.
,
Corbetta
,
M.
,
Van Essen
,
D. C.
, &
Raichle
,
M. E.
(
2005
).
The human brain is intrinsically organized into dynamic, anticorrelated functional networks.
Proceedings of the National Academy of Sciences, U.S.A.
,
102
,
9673
9678
.
Friston
,
K. J.
,
Worsley
,
K. J.
,
Frackowiak
,
R. S. J.
,
Mazziotta
,
J. C.
, &
Evans
,
A. C.
(
1994
).
Assessing the significance of focal activations using their spatial extent.
Human Brain Mapping
,
1
,
214
220
.
Geday
,
J.
,
Kupers
,
R.
, &
Gjedde
,
A.
(
2007
).
As time goes by: Temporal constraints on emotional activation of inferior medial prefrontal cortex.
Cerebral Cortex
,
17
,
2753
2759
.
Gosselin
,
N.
,
Peretz
,
I.
,
Johnsen
,
E.
, &
Adolphs
,
R.
(
2007
).
Amygdala damage impairs emotion recognition from music.
Neuropsychologia
,
45
,
236
244
.
Gosselin
,
N.
,
Peretz
,
I.
,
Noulhiane
,
M.
,
Hasboun
,
D.
,
Beckett
,
C.
,
Baulac
,
M.
et al
(
2005
).
Impaired recognition of scary music following unilateral temporal lobe excision.
Brain
,
128
,
628
640
.
Gosselin
,
N.
,
Samson
,
S.
,
Adolphs
,
R.
,
Noulhiane
,
M.
,
Roy
,
M.
,
Hasboun
,
D.
et al
(
2006
).
Emotional responses to unpleasant music correlates with damage to the parahippocampal cortex.
Brain
,
129
,
2585
2592
.
Greicius
,
M. D.
,
Krasnow
,
B.
,
Reiss
,
A. L.
, &
Menon
,
V.
(
2003
).
Functional connectivity in the resting brain: A network analysis of the default mode hypothesis.
Proceedings of the National Academy of Sciences, U.S.A.
,
100
,
253
258
.
Greicius
,
M. D.
, &
Menon
,
V.
(
2004
).
Default-mode activity during a passive sensory task: Uncoupled from deactivation but impacting activation.
Journal of Cognitive Neuroscience
,
16
,
1484
1492
.
Greicius
,
M. D.
,
Srivastava
,
G.
,
Reiss
,
A. L.
, &
Menon
,
V.
(
2004
).
Default-mode network activity distinguishes Alzheimer's disease from healthy aging: Evidence from functional MRI.
Proceedings of the National Academy of Sciences, U.S.A.
,
101
,
4637
4642
.
Gusnard
,
D. A.
,
Akbudak
,
E.
,
Shulman
,
G. L.
, &
Raichle
,
M. E.
(
2001
).
Medial prefrontal cortex and self-referential mental activity: Relation to a default mode of brain function.
Proceedings of the National Academy of Sciences, U.S.A.
,
98
,
4259
4264
.
Gusnard
,
D. A.
, &
Raichle
,
M. E.
(
2001
).
Searching for a baseline: Functional imaging and the resting human brain.
Nature Reviews Neuroscience
,
2
,
685
694
.
Habel
,
U.
,
Klein
,
M.
,
Kellermann
,
T.
,
Shah
,
N. J.
, &
Schneider
,
F.
(
2005
).
Same or different? Neural correlates of happy and sad mood in healthy males.
Neuroimage
,
26
,
206
214
.
Hariri
,
A. R.
,
Bookheimer
,
S. Y.
, &
Mazziotta
,
J. C.
(
2000
).
Modulating emotional responses: Effects of a neocortical network on the limbic system.
NeuroReport
,
11
,
43
48
.
Hariri
,
A. R.
,
Mattay
,
V. S.
,
Tessitore
,
A.
,
Fera
,
F.
, &
Weinberger
,
D. R.
(
2003
).
Neocortical modulation of the amygdala response to fearful stimuli.
Biological Psychiatry
,
53
,
494
501
.
Haxby
,
J. V.
,
Horwitz
,
B.
,
Ungerleider
,
L. G.
,
Maisog
,
J. M.
,
Pietrini
,
P.
, &
Grady
,
C. L.
(
1994
).
The functional organization of human extrastriate cortex: A PET–rCBF study of selective attention to faces and locations.
Journal of Neuroscience
,
14
,
6336
6353
.
Isenberg
,
N.
,
Silbersweig
,
D.
,
Engelien
,
A.
,
Emmerich
,
S.
,
Malavade
,
K.
,
Beattie
,
B.
et al
(
1999
).
Linguistic threat activates the human amygdala.
Proceedings of the National Academy of Sciences, U.S.A.
,
96
,
10456
10459
.
Jenkinson
,
M.
,
Bannister
,
P.
,
Brady
,
M.
, &
Smith
,
S.
(
2002
).
Improved optimization for the robust and accurate linear registration and motion correction of brain images.
Neuroimage
,
17
,
825
841
.
Jenkinson
,
M.
, &
Smith
,
S.
(
2001
).
A global optimisation method for robust affine registration of brain images.
Medical Image Analysis
,
5
,
143
156
.
Kameoka
,
A.
, &
Kuriyagawa
,
M.
(
1969
).
Consonance theory: Part I. Consonance of dyads.
Journal of the Acoustical Society of America
,
45
,
1451
1459
.
Khalfa
,
S.
,
Schon
,
D.
,
Anton
,
J. L.
, &
Liegeois-Chauvel
,
C.
(
2005
).
Brain regions involved in the recognition of happiness and sadness in music.
NeuroReport
,
16
,
1981
1984
.
Kim
,
H.
,
Somerville
,
L. H.
,
Johnstone
,
T.
,
Polis
,
S.
,
Alexander
,
A. L.
,
Shin
,
L. M.
et al
(
2004
).
Contextual modulation of amygdala responsivity to surprised faces.
Journal of Cognitive Neuroscience
,
16
,
1730
1745
.
Koelsch
,
S.
,
Fritz
,
T.
,
von Cramon
,
D. Y.
,
Muller
,
K.
, &
Friederici
,
A. D.
(
2006
).
Investigating emotion with music: An fMRI study.
Human Brain Mapping
,
27
,
239
250
.
Lane
,
R. D.
,
Fink
,
G. R.
,
Chau
,
P. M.
, &
Dolan
,
R. J.
(
1997
).
Neural activation during selective attention to subjective emotional responses.
NeuroReport
,
8
,
3969
3972
.
Laurienti
,
P. J.
,
Burdette
,
J. H.
,
Wallace
,
M. T.
,
Yen
,
Y. F.
,
Field
,
A. S.
, &
Stein
,
B. E.
(
2002
).
Deactivation of sensory-specific cortex by cross-modal stimuli.
Journal of Cognitive Neuroscience
,
14
,
420
429
.
LeDoux
,
J. E.
,
Cicchetti
,
P.
,
Xagoraris
,
A.
, &
Romanski
,
L. M.
(
1990
).
The lateral amygdaloid nucleus: Sensory interface of the amygdala in fear conditioning.
Journal of Neuroscience
,
10
,
1062
1069
.
Lewis
,
J. W.
,
Beauchamp
,
M. S.
, &
DeYoe
,
E. A.
(
2000
).
A comparison of visual and auditory motion processing in human cerebral cortex.
Cerebral Cortex
,
10
,
873
888
.
Liberzon
,
I.
,
Taylor
,
S. F.
,
Fig
,
L. M.
,
Decker
,
L. R.
,
Koeppe
,
R. A.
, &
Minoshima
,
S.
(
2000
).
Limbic activation and psychophysiologic responses to aversive visual stimuli. Interaction with cognitive task.
Neuropsychopharmacology
,
23
,
508
516
.
Mason
,
M. F.
,
Norton
,
M. I.
,
Van Horn
,
J. D.
,
Wegner
,
D. M.
,
Grafton
,
S. T.
, &
Macrae
,
C. N.
(
2007
).
Wandering minds: The default network and stimulus-independent thought.
Science
,
315
,
393
395
.
Mazoyer
,
B.
,
Zago
,
L.
,
Mellet
,
E.
,
Bricogne
,
S.
,
Etard
,
O.
,
Houde
,
O.
et al
(
2001
).
Cortical networks for working memory and executive functions sustain the conscious resting state in man.
Brain Research Bulletin
,
54
,
287
298
.
McKiernan
,
K. A.
,
Kaufman
,
J. N.
,
Kucera-Thompson
,
J.
, &
Binder
,
J. R.
(
2003
).
A parametric manipulation of factors affecting task-induced deactivation in functional neuroimaging.
Journal of Cognitive Neuroscience
,
15
,
394
408
.
Mitterschiffthaler
,
M. T.
,
Fu
,
C. H.
,
Dalton
,
J. A.
,
Andrew
,
C. M.
, &
Williams
,
S. C.
(
2007
).
A functional MRI study of happy and sad affective states induced by classical music.
Human Brain Mapping
,
28
,
1150
1162
.
Morris
,
J. S.
,
Friston
,
K. J.
,
Buchel
,
C.
,
Frith
,
C. D.
,
Young
,
A W.
,
Calder
,
A. J.
et al
(
1998
).
A neuromodulatory role for the human amygdala in processing emotional facial expressions.
Brain
,
121
,
47
57
.
Morris
,
J. S.
,
Frith
,
C. D.
,
Perrett
,
D. I.
,
Rowland
,
D.
,
Young
,
A. W.
,
Calder
,
A. J.
et al
(
1996
).
A differential neural response in the human amygdala to fearful and happy facial expressions.
Nature
,
383
,
812
815
.
Ochsner
,
K. N.
,
Bunge
,
S. A.
,
Gross
,
J. J.
, &
Gabrieli
,
J. D.
(
2002
).
Rethinking feelings: An FMRI study of the cognitive regulation of emotion.
Journal of Cognitive Neuroscience
,
14
,
1215
1229
.
Ochsner
,
K. N.
,
Knierim
,
K.
,
Ludlow
,
D. H.
,
Hanelin
,
J.
,
Ramachandran
,
T.
,
Glover
,
G.
et al
(
2004
).
Reflecting upon feelings: An fMRI study of neural systems supporting the attribution of emotion to self and other.
Journal of Cognitive Neuroscience
,
16
,
1746
1772
.
Ochsner
,
K. N.
,
Ray
,
R. D.
,
Cooper
,
J. C.
,
Robertson
,
E. R.
,
Chopra
,
S.
,
Gabrieli
,
J. D.
et al
(
2004
).
For better or for worse: Neural systems supporting the cognitive down- and up-regulation of negative emotion.
Neuroimage
,
23
,
483
499
.
Pallesen
,
K. J.
,
Brattico
,
E.
,
Bailey
,
C.
,
Korvenoja
,
A.
,
Koivisto
,
J.
,
Gjedde
,
A.
et al
(
2005
).
Emotion processing of major, minor, and dissonant chords: A functional magnetic resonance imaging study.
Annals of the New York Academy of Sciences
,
1060
,
450
453
.
Pallesen
,
K. J.
,
Brattico
,
E.
, &
Carlson
,
S.
(
2003
).
Emotional connotations of major and minor musical chords in musically untrained listeners.
Brain and Cognition
,
51
,
188
190
.
Petersen
,
S. E.
,
van Mier
,
H.
,
Fiez
,
J. A.
, &
Raichle
,
M. E.
(
1998
).
The effects of practice on the functional anatomy of task performance.
Proceedings of the National Academy of Sciences, U.S.A.
,
95
,
853
860
.
Phan
,
K. L.
,
Wager
,
T.
,
Taylor
,
S. F.
, &
Liberzon
,
I.
(
2002
).
Functional neuroanatomy of emotion: A meta-analysis of emotion activation studies in PET and fMRI.
Neuroimage
,
16
,
331
348
.
Plomp
,
R.
, &
Levelt
,
W. J.
(
1965
).
Tonal consonance and critical bandwidth.
Journal of the Acoustical Society of America
,
38
,
548
560
.
Raichle
,
M. E.
(
1998
).
Behind the scenes of functional brain imaging: A historical and physiological perspective.
Proceedings of the National Academy of Sciences, U.S.A.
,
95
,
765
772
.
Raichle
,
M. E.
,
Fiez
,
J. A.
,
Videen
,
T. O.
,
MacLeod
,
A. M.
,
Pardo
,
J. V.
,
Fox
,
P. T.
et al
(
1994
).
Practice-related changes in human brain functional anatomy during nonmotor learning.
Cerebral Cortex
,
4
,
8
26
.
Raichle
,
M. E.
,
MacLeod
,
A. M.
,
Snyder
,
A. Z.
,
Powers
,
W. J.
,
Gusnard
,
D. A.
, &
Shulman
,
G. L.
(
2001
).
A default mode of brain function.
Proceedings of the National Academy of Sciences, U.S.A.
,
98
,
676
682
.
Raichle
,
M. E.
, &
Snyder
,
A. Z.
(
2007
).
A default mode of brain function: A brief history of an evolving idea.
Neuroimage
,
37
,
1083
1090
; discussion 1097–1099.
Rämä
,
P.
,
Martinkauppi
,
S.
,
Linnankoski
,
I.
,
Koivisto
,
J.
,
Aronen
,
H. J.
, &
Carlson
,
S.
(
2001
).
Working memory of identification of emotional vocal expressions: An fMRI study.
Neuroimage
,
13
,
1090
1101
.
Schön
,
D.
,
Regnault
,
P.
,
Ystad
,
S.
, &
Besson
,
M.
(
2005
).
Sensory consonance: An ERP study.
Music Perception
,
23
,
105
117
.
Shulman
,
G. L.
,
Fiez
,
J. A.
,
Corbetta
,
M.
,
Buckner
,
R. L.
,
Miezin
,
F. M.
,
Raichle
,
M. E.
et al
(
1997
).
Common blood flow changes across visual tasks: II. Decreases in cerebral cortex.
Journal of Cognitive Neuroscience
,
9
,
648
663
.
Simpson
,
J. R.
, Jr.,
Snyder
,
A. Z.
,
Gusnard
,
D. A.
, &
Raichle
,
M. E.
(
2001
).
Emotion-induced changes in human medial prefrontal cortex: I. During cognitive task performance.
Proceedings of the National Academy of Sciences, U.S.A.
,
98
,
683
687
.
Smallwood
,
J.
, &
Schooler
,
J. W.
(
2006
).
The restless mind.
Psychological Bulletin
,
132
,
946
958
.
Smith
,
S. M.
(
2002
).
Fast robust automated brain extraction.
Human Brain Mapping
,
17
,
143
155
.
Steele
,
J. D.
, &
Lawrie
,
S. M.
(
2004
).
Segregation of cognitive and emotional function in the prefrontal cortex: A stereotactic meta-analysis.
Neuroimage
,
21
,
868
875
.
Teasdale
,
J. D.
,
Dritschel
,
B. H.
,
Taylor
,
M. J.
,
Proctor
,
L.
,
Lloyd
,
C. A.
,
Nimmo-Smith
,
I.
et al
(
1995
).
Stimulus-independent thought depends on central executive resources.
Memory & Cognition
,
23
,
551
559
.
Tomasi
,
D.
,
Ernst
,
T.
,
Caparelli
,
E. C.
, &
Chang
,
L.
(
2006
).
Common deactivation patterns during working memory and visual attention tasks: An intra-subject fMRI study at 4 Tesla.
Human Brain Mapping
,
27
,
694
705
.
Trainor
,
L. J.
, &
Heinmiller
,
B. M.
(
1998
).
Infants prefer to listen to consonance over dissonance.
Infant Behaviour and Development
,
21
,
77
88
.
Tzourio-Mazoyer
,
N.
,
Landeau
,
B.
,
Papathanassiou
,
D.
,
Crivello
,
F.
,
Etard
,
O.
,
Delcroix
,
N.
et al
(
2002
).
Automated anatomical labeling of activations in SPM using a macroscopic anatomical parcellation of the MNI MRI single-subject brain.
Neuroimage
,
15
,
273
289
.
Vincent
,
J. L.
,
Patel
,
G. H.
,
Fox
,
M. D.
,
Snyder
,
A. Z.
,
Baker
,
J. T.
,
Van Essen
,
D. C.
et al
(
2007
).
Intrinsic functional architecture in the anaesthetized monkey brain.
Nature
,
447
,
83
86
.
Vincent
,
J. L.
,
Snyder
,
A. Z.
,
Fox
,
M. D.
,
Shannon
,
B. J.
,
Andrews
,
J. R.
,
Raichle
,
M. E.
et al
(
2006
).
Coherent spontaneous activity identifies a hippocampal–parietal memory network.
Journal of Neurophysiology
,
96
,
3517
3531
.
Woolrich
,
M. W.
,
Behrens
,
T. E.
,
Beckmann
,
C. F.
,
Jenkinson
,
M.
, &
Smith
,
S. M.
(
2004
).
Multilevel linear modelling for FMRI group analysis using Bayesian inference.
Neuroimage
,
21
,
1732
1747
.
Woolrich
,
M. W.
,
Ripley
,
B. D.
,
Brady
,
M.
, &
Smith
,
S. M.
(
2001
).
Temporal autocorrelation in univariate linear modeling of FMRI data.
Neuroimage
,
14
,
1370
1386
.
Worsley
,
K. J.
,
Evans
,
A. C.
,
Marrett
,
S.
, &
Neelin
,
P.
(
1992
).
A three-dimensional statistical analysis for CBF activation studies in human brain.
Journal of Cerebral Blood Flow and Metabolism
,
12
,
900
918
.
Zatorre
,
R. J.
,
Evans
,
A. C.
, &
Meyer
,
E.
(
1994
).
Neural mechanisms underlying melodic perception and memory for pitch.
Journal of Neuroscience
,
14
,
1908
1919
.
Zentner
,
M. R.
, &
Kagan
,
J.
(
1998
).
Infants' perception of consonance and dissonance in music.
Infant Behaviour and Development
,
21
,
483
492
.