## Abstract

Visualizing emotionally loaded pictures intensifies peripheral reflexes toward sudden auditory stimuli, suggesting that the emotional context may potentiate responses elicited by novel events in the acoustic environment. However, psychophysiological results have reported that attentional resources available to sounds become depleted, as attention allocation to emotional pictures increases. These findings have raised the challenging question of whether an emotional context actually enhances or attenuates auditory novelty processing at a central level in the brain. To solve this issue, we used functional magnetic resonance imaging to first identify brain activations induced by novel sounds (NOV) when participants made a color decision on visual stimuli containing both negative (NEG) and neutral (NEU) facial expressions. We then measured modulation of these auditory responses by the emotional load of the task. Contrary to what was assumed, activation induced by NOV in superior temporal gyrus (STG) was enhanced when subjects responded to faces with a NEG emotional expression compared with NEU ones. Accordingly, NOV yielded stronger behavioral disruption on subjects’ performance in the NEG context. These results demonstrate that the emotional context modulates the excitability of auditory and possibly multimodal novelty cerebral regions, enhancing acoustic novelty processing in a potentially harming environment.

## Introduction

The human nervous system has evolved to efficiently detect salient information in a multisensory environment, which is a necessary skill for adaptive behavior. Particularly, a large body of evidence has shown that emotional stimuli have priority status in the neural processing systems, eliciting stronger and faster attention capture than nonemotional stimuli (e.g., Eastwood et al. 2001; Öhman et al. 2001; Carretié et al. 2004; Richards and Blanchette 2004). Such a mechanism of attentional bias arises from the limited processing capacity of sensory systems, while attending to adaptive and evolutionary advantages. In this sense, emotion interferes with the processing of concomitant stimuli, not only within (Öhman et al. 2001; Fox 2002) but also across sensory modalities. A number of psychophysiological studies have yielded potentiation of the effects elicited by auditory stimuli in emotional context. Peripheral responses, such as eyeblinks after the sudden burst of an acoustic stimulus (known as startle reflex), have been shown to intensify when visualizing emotionally loaded pictures (Stanley and Knight 2004; Bradley et al. 2006). However, psychophysiological measures have suggested that, at the same time, less attentional resources might be available for sound processing as a consequence of allocating more attention to these motivationally pertinent stimuli (Schupp et al. 1997; Cuthbert et al. 1998; Keil et al. 2007). This debate led us to investigate the neural mechanisms of sound processing and detection of acoustic changes in emotional environments and whether an affective context can either enhance or attenuate novelty processing in the auditory modality. Unraveling this effect would provide a better understanding of how emotion mediates involuntary attention and awareness.

In the present study, functional magnetic resonance imaging (fMRI) was used to assess the effects of manipulating the implicit negative (NEG) emotional load in a visual task, by means of fearful and angry faces, on the processing of task-irrelevant odd auditory stimuli. The occurrence of unexpected novel sounds (NOV) has been typically shown to elicit a specific pattern of neuronal activations localized in supratemporal (Alho et al. 1998; Downar et al. 2000; Kiehl et al. 2001), prefrontal (Downar et al. 2001; Bledowski et al. 2004), and parietal cortices (Clark et al. 2000; Downar et al. 2001), subserving the neural mechanisms of auditory novelty processing(Ranganath and Rainer 2003). This pattern of neuronal activations is usually accompanied by behavioral disruption of the ongoing task, a phenomenon called distraction (Escera et al. 1998, 2000, 2001; Escera and Corral 2007). If the emotional context facilitates the processing of task-irrelevant NOV, these NOV should elicit greater distraction when subjects respond to emotionally loaded stimuli as compared with neutral (NEU) ones. Moreover, areas related to auditory novelty processing should become more activated. If, on the contrary, emotional processing depletes most of the attentional resources, distraction should decrease and areas related to auditory novelty processing should become rather inhibited. To test this hypothesis, we presented our participants with task-irrelevant NOV while they judged whether the color of a monochromatic face, with task-irrelevant NEU or NEG facial expression, matched the color of the surrounding frame (Fig. 1a).

Figure 1.

Trial structure and behavioral data. (a) Sample visual and auditory stimuli. (b) Mean response times for STD and NOV trials, both when sounds were preceded and followed by NEU faces (NEU context) and when sounds were preceded and followed by NEG faces (NEG context). NOV sounds caused a delay on subjects’ responses, being this effect significantly enlarged when preceded and followed by fearful or angry faces. Bars indicate the standard error of the mean (±SEM).

Figure 1.

Trial structure and behavioral data. (a) Sample visual and auditory stimuli. (b) Mean response times for STD and NOV trials, both when sounds were preceded and followed by NEU faces (NEU context) and when sounds were preceded and followed by NEG faces (NEG context). NOV sounds caused a delay on subjects’ responses, being this effect significantly enlarged when preceded and followed by fearful or angry faces. Bars indicate the standard error of the mean (±SEM).

## Materials and Methods

### Subjects

Seventeen healthy right-handed female volunteers, aged between 19 and 30 years (mean 22 years, ±3.35), without past neurological or psychiatric history, who reported normal hearing and normal or corrected-to-normal vision, participated in the present study, which was approved by the board of directors of the Center of Advanced Imaging and the Ethical Committee of the University of Barcelona. All subjects gave informed consent according to procedures set by the local ethics committee and completed the color blindness Ishihara test (Kanehara Shuppen Company, Tokyo, Japan, 1974) to ensure normal color vision.

### Stimuli

#### Emotionally Valent Stimuli

The emotionally valent stimuli consisted of 120 monochromatic pictures including neutrally (60) and negatively valenced facial expressions (30 fearful, 30 angry) and 6 positively valenced facial expressions to provide an additional ending block. They were compiled at the Department of Neuropsychology and Behavioral Neurobiology, Bremen University, Bremen, Germany (Weitzel V, unpublished thesis) and evaluated by a sample of 73 subjects (age 19–40 years, mean 25.16 ± 5.1). Faces with the highest correctly categorized hit rate were selected (displaying NEU—mean 79.6% ± 8.9%; angry—mean 97.2% ± 2.2%; or fearful expressions—mean 93.3% ± 3.4%). Of the chosen pictures, 50% showed male faces and 50% showed female faces (resolution 500 × 500 pixels, duration on screen 400 ms). The pictures were back projected onto a mirror mounted on the MRI head coil. Visual angles varied with the distance from the screen (35–40 cm depending on the subjects’ head size) and were of 17°–15° horizontal angle and 18°–16° vertical angle.

All faces were surrounded by frames. The colors of both the faces and the frames were presented in green, blue, red, or orange in equal proportions, all equally and proportionally balanced. The faces and frames could be of the same color (50%) or of different colors (50%) across the trials. Every NEU face appeared a total of 6 times. From these NEU faces, the 7 most rated (8 for the no-match condition) appeared 8 times. All NEG faces (angry and fearful) appeared a total of 6 times. From these NEG faces, the 7 most rated (or 8 for match condition) appeared 8 times.

#### Auditory Stimuli

The auditory stimuli were presented at 20 dB sensation level and consisted of a 700 Hz standard tone (STD) and 100 unique environmental complex, NOV (200 ms duration), equalized for root mean square (RMS, see Supplementary Material) energy to keep the energy contour of all auditory stimuli constant over time, generated as described in previous studies (Escera et al. 1998). They were all similar in spectrotemporal features (see Supplementary Material) and were rated by a sample of 30 subjects, on a 1–5 Likert scale of semantic familiarity (which is defined as the quality of a novel sound having a particular meaning for the participants and being therefore defined as identifiable; see Escera et al. (2003) on the effects of familiarity on novelty processing). Sounds were chosen among those rated most familiar (2.54 mean rate ± 0.5). Throughout the session, all novel sounds appeared once except for the 50 most familiar, which appeared twice, all of them in random order.

Participants performed a modified version of a well-characterized auditory–visual distraction paradigm (Escera et al. 1998, 2000, 2001, 2002, 2003). The auditory stimuli, which preceded the images by 300 ms (onset-to-onset), were either a STD condition (P = 0.8) or a NOV condition (P = 0.2). Faces could be NEU (NEU context condition) or NEG (NEG context condition). The pictures appeared in the middle of the screen and the subjects were instructed to press a response button (left or right hand, counterbalanced across subjects) to respond as to whether the color of the frame matched the color of the face (Fig. 1a), while ignoring the sounds. The trial length ranged from 2600 to 3200 ms (mean 2900 ± 300 ms). A unique sequence was designed, which divided 750 trials into 50 blocks of 15, 10, or 20 trials. NEU and NEG faces were never mixed in the same block. In this manner, scanning time per condition was of a minimum of 26 s (10 trials, 26 000 ms). All blocks were pseudorandomized in a probabilistic nonstationary way so that, in the beginning, a higher proportion of NEU blocks was presented, decreasing progressively and turning into a higher proportion of NEG pictures at the end. These sequences were counterbalanced across subjects.

### Procedure

Before each measurement, the subjects performed an only-standard sound practice block of the task outside the scanner. Sound calibration was implemented inside the scanner to adjust hearing thresholds and correct balance between left- and right-ear inputs for each individual subject with the help of a sound amplifier. During the course of this procedure, a shim echo-planar sequence was applied to ensure that the subjects would hear the sounds also in the scanning session.

### FMRI Scanning

MRI data were acquired on a 3T Siemens Allegra scanner (Erlangen, Germany), using a whole-brain local gradient coil. Structural images were acquired with a T1-weighted MPRAGE sequence (160 slices, TR 2.3 s, TE 4.38 ms, flip angle 8°, 256 × 256 matrix, FOV 296 × 296, inversion time 900 ms, 1 mm3 voxels). Functional images were obtained using a gradient echo-planar (EPI) T2* sequence optimized for blood oxygenation level dependency (BOLD) contrast. The EPI sequence comprised 44 contiguous slices covering the whole brain, taken every 3 mm with no interslice gap and incorporating the following parameters: TE 30 ms, TR 2.5 s, 64 × 64 matrix, FOV 192 × 192, flip angle 90°, resolution 3 × 3 mm2, and interleaved ascending. Functional images were slice time corrected, realigned, normalized spatially to the Montreal Neurological Institute (MNI) template and smoothed with an 8-mm FWHM gaussian kernel using SPM2 (Wellcome Department of Cognitive Neurology, London, UK, 2003).

### Data Analysis

Behavioral performance was analyzed by calculating accuracy (hit rate) and response times for every condition, and post hoc comparisons were performed using the Bonferroni adjustment for multiple comparisons. Only those trials in which sounds were surrounded (both preceded and followed) by both NEU or NEG faces and containing correct responses were analyzed, both for behavioral and fMRI data. At the first level, 15 event types were modeled: 6 event types referred to trials containing sounds surrounded by equally valenced pictures and followed by a correct response, 6 more referred to trials comprising misses and errors, and a further 3 referred to all trials at the beginning of every valence block or after breaks. Data were high-pass filtered (1/128 Hz), corrected for intrinsic autocorrelations, and convolved with a standard hemodynamic response function (HRF) and its temporal derivative. Second-level analysis was performed on single-subject statistical parametric maps, which served as random effects. A pooled analysis was used to identify activations additively significant for ([STD − fear] + [STD − anger]) and ([NOV − fear] + [NOV − anger]), which resulted in the new NEG conditions. Three contrasts of interest were defined, by means of voxel-referred t–tests, to compare BOLD signal for the following conditions: STD − NEU < NOV − NEU (novelty processing in NEU context); STD − NEG < NOV − NEG (novelty processing in NEG context); and STD − NEU < STD − NEG (emotional face processing). A fourth contrast of interest (emotional effects on novelty processing) was examined with a 1-factor analysis of variance (ANOVA): ([STD − NEU < NOV − NEU] < {([STD − fear] + [STD − anger]) < ([NOV − fear] + [NOV − anger])}). An additional 1-factor ANOVA was performed (the reverse contrast): ([STD − NEU < NOV − NEU] > {([STD − fear] + [STD − anger]) < ([NOV − fear] + [NOV − anger])}).

Activation was considered significant when at least 20 contiguous voxels survived a threshold of P < 0.001 (except for STD − NEU < STD − NEG; P < 0.005), uncorrected for multiple comparisons. Stereotactic MNI coordinates were translated into standard Talairach space (Talairach and Tournoux 1988) following nonlinear transformations.

### Localization of BOLD Response in the Primary Auditory Cortex

The primary auditory cortex (PAC) is known to be particularly involved in the initial detection of changes in the acoustic environment (Ulanovsky et al. 2003; Schönwieser et al. 2007). Finding emotional modulation in this region would help us to constrain the timing of this effect in the novelty processing stream. One important issue that concerned us was the difficulty in localizing specific activations of PAC in functional imaging data. Cytoarchitectonic studies demonstrate considerable interindividual and interhemispheric differences in size and location of PAC, as well as duplications of Heschl's Gyrus (Penhune et al. 1996). The MNI template is known to be roughly based on the Talairach space, and it does not totally match the Talairach brain size or shape (Brett et al. 2002). Therefore, Talairach labels for the cytoarchitectonic allocation of functional activations is problematic because it does not provide information about the interindividual variability of the cytoarchitectonic areas (Eickhoff et al. 2005). Thus, there was a risk of Type II errors (false negatives) in respect to PAC for the first- and second-level analyses. To overcome this limitation, we used a 3-dimensional probabilistic cytoarchitectonic map based on an observer-independent quantification of cell volume densities and areal borders of PAC on 10 human postmortem brains (Morosan et al. 2001; Rademacher et al. 2001). PAC, microstructurally defined and normalized in space (Eickhoff et al. 2005), was then set as an anatomical region of interest (ROI) for every single subject. Then, the relative extent of activation for those clusters showing the highest probability to be part of the 3 cytoarchitectonic areas of bilateral PAC (TE1.0, TE1.1, TE1.2) was computed for the contrasts of interest emotional effects on novelty processing and the reverse contrast.

## Results

### Behavioral Data

The subjects had an overall hit rate of approximately 95% (STD in NEU: $x-=95.9%$, σ = 2.84%; NOV in NEU: $x-=96.5%$, σ = 2.46%; STD in NEG: $x-=97.1%$, σ = 1.98%; NOV in NEG: $x-=96.0%$, σ = 4.19%). There were no statistical differences in accuracy between sound effects for NEU and NEG conditions, although a significant decrease in accuracy was found for NOV sounds in fear condition (context × sound: F(1,16) = 7.19; P = 0.016). A 2-factor repeated-measurement ANOVA revealed significantly longer response times in trials containing NOV sounds compared with those containing STD sounds (STD in NEU: $x-=536$ms, σ = 78.6 ms; NOV in NEU: $x-=570$ms, σ = 75.6 ms; STD in NEG: $x-=537$ms, σ = 76.0 ms; NOV in NEG: $x-=581$ms, σ = 79.6 ms; sound: F(1,16) = 35.93; P = 0.001), indicating by their delayed responses, so that the subjects were distracted from task performance by the task-irrelevant occurrence of novel sounds, in agreement with many previous studies (Alho et al. 1997; Escera et al. 1998, 2001, 2003) (Fig. 1b). Distraction effects (NOV > STD trials) were of 34 ms for the NEU context and 44 ms for the NEG context. Thus, these response time differences between STD and NOV trials were larger for the NEG context compared with the NEU context (context × sound: F(1,16) = 5.3; P = 0.035). Two-tailed dependent t-tests showed a difference between NOV trials in NEG and NEU context (NOV in NEU vs. NOV in NEG: T16 = 2.6; P = 0.019), whereas the response time in STD trials was similar for both NEG and NEU context (T16 = 0.06; P = 0.95).

### Imaging Data

#### Emotional Face Processing

Hemodynamic responses for the STD sounds showed a differential activation pattern during the processing of angry and fearful expressions compared with the processing of NEU ones in bilateral fusiform gyrus and right inferior temporal gyrus (Fig. 2, Table 1), areas known to be involved in face processing (both fusiform gyrus—George et al. 1999—and inferior temporal gyrus—Rolls 1992). Fusiform cortex has been found to be modulated by emotional faces in a variety of studies (Sugase et al. 1999; Vuilleumier et al. 2001).

Table 1

Emotional face processing (STD in NEU context < STD in NEG context): regions activated

 Brain region Brodmann area z value Size (voxels) Coordinates x y z R fusiform gyrus 37 4.14 278 46 −41 −11 R fusiform gyrus/inferior temporal gyrus 37 4.14 46 −41 −11 R fusiform gyrus 37 3.55 46 −48 −18 R fusiform gyrus 37 3.76 336 42 −61 −10 L fusiform gyrus 37 4.44 1222 −42 −57 −16 L fusiform gyrus 19 4.36 −44 −74 −11 L fusiform gyrus 20 3.47 24 −44 −15 −24 R inferior temporal gyrus/uncus 20 4.12 237 34 −8 −35 R inferior temporal gyrus 20 4.08 38 −2 −39 L inferior temporal gyrus/uncus 20 3.26 40 −32 −8 −35 R superior frontal gyrus/medial frontal gyrus 6 3.90 62 12 7 55 L cuneus/middle occipital gyrus 18 3.74 337 −10 −100 14 L middle occipital gyrus 19 3.59 −34 −89 10 L middle occipital gyrus 18 3.58 −28 −87 4 R middle occipital gyrus 18/19 3.69 440 34 −85 10 R middle occipital gyrus 19 3.06 51 −74 −5 L lingual gyrus 17 3.46 32 −14 −94 −9 R STG 38 3.45 37 42 7 −17 R STG 38 2.80 48 15 −18 R medial frontal gyrus 11 3.33 65 6 42 −16 L rectal gyrus 11 2.97 −2 36 −22 R precentral gyrus 6 3.12 25 50 0 46 R precentral gyrus 4 2.70 53 −7 46 L insula 13 3.11 35 −36 20 12 R amygdala — 3.09 83 28 −1 −17 R parahippocampal gyrus–hippocampus — 2.95 28 −9 −20
 Brain region Brodmann area z value Size (voxels) Coordinates x y z R fusiform gyrus 37 4.14 278 46 −41 −11 R fusiform gyrus/inferior temporal gyrus 37 4.14 46 −41 −11 R fusiform gyrus 37 3.55 46 −48 −18 R fusiform gyrus 37 3.76 336 42 −61 −10 L fusiform gyrus 37 4.44 1222 −42 −57 −16 L fusiform gyrus 19 4.36 −44 −74 −11 L fusiform gyrus 20 3.47 24 −44 −15 −24 R inferior temporal gyrus/uncus 20 4.12 237 34 −8 −35 R inferior temporal gyrus 20 4.08 38 −2 −39 L inferior temporal gyrus/uncus 20 3.26 40 −32 −8 −35 R superior frontal gyrus/medial frontal gyrus 6 3.90 62 12 7 55 L cuneus/middle occipital gyrus 18 3.74 337 −10 −100 14 L middle occipital gyrus 19 3.59 −34 −89 10 L middle occipital gyrus 18 3.58 −28 −87 4 R middle occipital gyrus 18/19 3.69 440 34 −85 10 R middle occipital gyrus 19 3.06 51 −74 −5 L lingual gyrus 17 3.46 32 −14 −94 −9 R STG 38 3.45 37 42 7 −17 R STG 38 2.80 48 15 −18 R medial frontal gyrus 11 3.33 65 6 42 −16 L rectal gyrus 11 2.97 −2 36 −22 R precentral gyrus 6 3.12 25 50 0 46 R precentral gyrus 4 2.70 53 −7 46 L insula 13 3.11 35 −36 20 12 R amygdala — 3.09 83 28 −1 −17 R parahippocampal gyrus–hippocampus — 2.95 28 −9 −20

All coordinates reported in Talairach space. Activations shown are based on a voxelwise P < 0.005, uncorrected, k = 20.

Figure 2.

Emotional face processing. (a) Bilateral fusiform gyrus (FUS, left corner) and right amygdala activations (AMG, right corner, white circle) for NEG versus NEU faces. The plane coordinates of each slice are indicated in the upper right-hand corner. Bright colors in coronal slides represent significance levels of contrasts, as indicated by the scale bar. (b) Average activity (± [standard error of the mean] SEM) across conditions for a left fusiform cluster (mean x, y, z, −42, −57, −16, 1222 voxels at P < 0.005). (c) Average activity (±SEM) across conditions for the right amygdala cluster (mean x, y, z, 28, −1, −17, 83 voxels at P < 0.005).

Figure 2.

Emotional face processing. (a) Bilateral fusiform gyrus (FUS, left corner) and right amygdala activations (AMG, right corner, white circle) for NEG versus NEU faces. The plane coordinates of each slice are indicated in the upper right-hand corner. Bright colors in coronal slides represent significance levels of contrasts, as indicated by the scale bar. (b) Average activity (± [standard error of the mean] SEM) across conditions for a left fusiform cluster (mean x, y, z, −42, −57, −16, 1222 voxels at P < 0.005). (c) Average activity (±SEM) across conditions for the right amygdala cluster (mean x, y, z, 28, −1, −17, 83 voxels at P < 0.005).

Additionally, early visual processing areas (bilateral middle occipital gyrus) as well as right precentral gyrus (premotor cortex) were widely activated. Visual cortices, such as V2, are known to exhibit an increased response to faces with fearful expressions compared with NEU ones (Morris et al. 1998). Passive viewing of emotional faces has also previously elicited enhanced activations in the right ventral premotor area (Leslie et al. 2004), suggesting a right hemisphere mirroring system that may subserve the neural substrate for empathy. Interestingly, in our results, BOLD response of limbic lobe areas such as right parahippocampal gyrus and right amygdala was observed. This limbic right lateralization is consistent with evidence attributing affective comprehension to the right hemisphere, as seen in patients with right amygdala damage who were unable to interpret facial expressions (Anderson and Phelps 2000) or patients with left hemisphere damage who presented difficulties in comprehending words but could report the meaning of their emotional prosody (Barrett et al. 1999). A large body of evidence indicates that the amygdala is critically involved in processing aversive information (Adolphs and Tranel 2004) and is capable of organizing rapid reactions to danger, even without the participation of the cerebral cortex (LeDoux 1998).

Differential brain activations in fearful and angry faces compared with NEU expressions were necessary prerequisite for evaluating any further emotional effect on novelty processing and distraction.

#### Auditory Novelty Processing

Before examining emotional modulation of novelty processing, we tested the effects of novelty processing in every single context. Novelty processing engaged a cerebral network expressed in prefrontal (bilateral inferior frontal gyrus—IFG), parietal (bilateral precuneus), and temporal cortices (bilateral superior temporal gyrus [STG] and middle temporal gyrus) both for the NEU and the NEG context (Fig. 3, Table 2).

Table 2

Novelty processing (STD < NOV): regions activated

 Brain region Brodmann area z value Size (voxels) Coordinates x y z STD < NOV (NEU context) R STG 22 5.96 4513 50 −14 −8 L STG 41 5.67 3962 −50 −29 9 L STG/transverse temporal gyrus 41 5.56 −57 −23 9 R middle temporal gyrus 21 5.59 4513 55 −20 −4 L middle temporal gyrus 21 3.43 21 −59 −54 3 R inferior frontal gyrus 9/46 4.39 146 46 17 21 R inferior frontal gyrus 47 3.87 88 51 29 −1 L inferior frontal gyrus 44/45 3.64 72 −44 18 12 L inferior frontal gyrus 13 3.61 −38 22 8 R middle frontal gyrus 46 4.39 146 46 17 21 R medial frontal gyrus 9 3.41 23 4 40 27 R lingual gyrus 18 3.66 22 −56 6 R paracentral lobule 5 4.14 264 6 −42 50 L paracentral lobule 5/7 3.18 −10 −46 58 L posterior cingulate 30/31 3.67 85 −18 −63 14 R cuneus 17 3.66 24 6 −77 9 R precuneus 7 4.14 264 6 −42 50 R precuneus 31 3.59 63 18 −67 20 L precuneus 7 3.82 264 −2 −46 45 L precuneus 7 3.18 −10 −46 58 L precuneus 31 4.35 85 −10 −67 20 STD < NOV (NEG context) R STG/transverse temporal gyrus 42 5.45 3339 67 −17 10 L STG 22 5.02 3021 −51 −13 8 R inferior frontal gyrus 45 4.73 80 36 27 4 L inferior frontal gyrus 9 3.79 87 −48 19 21 L middle frontal gyrus 46 3.79 −48 19 21 L insula 13 4.17 85 −38 19 1 R precuneus 7 3.86 36 10 −46 45 L precuneus 7 4.49 88 −8 −48 45
 Brain region Brodmann area z value Size (voxels) Coordinates x y z STD < NOV (NEU context) R STG 22 5.96 4513 50 −14 −8 L STG 41 5.67 3962 −50 −29 9 L STG/transverse temporal gyrus 41 5.56 −57 −23 9 R middle temporal gyrus 21 5.59 4513 55 −20 −4 L middle temporal gyrus 21 3.43 21 −59 −54 3 R inferior frontal gyrus 9/46 4.39 146 46 17 21 R inferior frontal gyrus 47 3.87 88 51 29 −1 L inferior frontal gyrus 44/45 3.64 72 −44 18 12 L inferior frontal gyrus 13 3.61 −38 22 8 R middle frontal gyrus 46 4.39 146 46 17 21 R medial frontal gyrus 9 3.41 23 4 40 27 R lingual gyrus 18 3.66 22 −56 6 R paracentral lobule 5 4.14 264 6 −42 50 L paracentral lobule 5/7 3.18 −10 −46 58 L posterior cingulate 30/31 3.67 85 −18 −63 14 R cuneus 17 3.66 24 6 −77 9 R precuneus 7 4.14 264 6 −42 50 R precuneus 31 3.59 63 18 −67 20 L precuneus 7 3.82 264 −2 −46 45 L precuneus 7 3.18 −10 −46 58 L precuneus 31 4.35 85 −10 −67 20 STD < NOV (NEG context) R STG/transverse temporal gyrus 42 5.45 3339 67 −17 10 L STG 22 5.02 3021 −51 −13 8 R inferior frontal gyrus 45 4.73 80 36 27 4 L inferior frontal gyrus 9 3.79 87 −48 19 21 L middle frontal gyrus 46 3.79 −48 19 21 L insula 13 4.17 85 −38 19 1 R precuneus 7 3.86 36 10 −46 45 L precuneus 7 4.49 88 −8 −48 45

All coordinates reported in Talairach space. Activations shown are based on a voxelwise P < 0.001, uncorrected, k = 20.

Figure 3.

Novelty processing (NOV > STD). BOLD activations for NOV versus STD trials (novelty processing) in NEU context (left) and NEG context (right). In both contexts, main activations were located in bilateral STG, bilateral inferior frontal gyrus (IFG), and bilateral precuneus.

Figure 3.

Novelty processing (NOV > STD). BOLD activations for NOV versus STD trials (novelty processing) in NEU context (left) and NEG context (right). In both contexts, main activations were located in bilateral STG, bilateral inferior frontal gyrus (IFG), and bilateral precuneus.

### Modulation of Novelty Processing in NEG Emotional Context

As can be observed in Figure 4 and Table 3, bilateral STG, including bilateral secondary auditory cortex (cytoarchitectonic area Te2—BA42 and BA21), and bilateral planum temporale (cytoarchitectonic area Te3, BA22), showed a stronger response when the subjects processed auditory NOV events in a NEG emotional context as compared with the NEU one. Subsequent analysis of the relative extents of activation, based on probabilistic maps, showed that 2.9% of the activated voxels were allocated to right Te1.0 and, from these, 3.4% showed significantly increased activations. A smaller proportion of activated voxels was allocated to Te1.2 (0.2%), from which 0.5% was more activated. On the left hemisphere, 13.4% of the activated voxels were allocated to Te1.0, showing significant activations in 15% of them, and 6.1% were included in Te1.1, with 7.9% significantly activated. It can therefore be concluded that, based on probabilistic maps (Eickhoff et al. 2007), the part of the voxels that showed more activation when processing auditory novelty in the NEG context, compared with the NEU, were probably allocated to PAC.

Table 3

Emotional effects on novelty processing (novelty processing in NEU < novelty processing in NEG)

 Brain region Brodmann area z value Size (voxels) Coordinates x y z Novelty processing (NEU < NEG) R STG 21/22 3.87 295 59 −10 −1 L STG 22/42 3.72 215 −65 −32 13 L STG 22 3.62 −51 −13 8
 Brain region Brodmann area z value Size (voxels) Coordinates x y z Novelty processing (NEU < NEG) R STG 21/22 3.87 295 59 −10 −1 L STG 22/42 3.72 215 −65 −32 13 L STG 22 3.62 −51 −13 8

All coordinates reported in Talairach space. Activations shown are based on a voxelwise P < 0.001, uncorrected, k = 20.

Figure 4.

Emotional effects on novelty processing (NOV > STD in NEU context) < (NOV > STD in NEG context). (a) Bilateral STG activations for novelty processing in NEU context < novelty processing in NEG context. The plane coordinate of the slice is indicated in the upper right-hand corner. Bright colors in axial and coronal slides represent significance levels of contrasts, as indicated by the scale bar. (b) Average activity (±standard error of the mean) across conditions for the right STG cluster (mean x, y, z, 59, −10, −1, 295 voxels at P < 0.001, k = 20).

Figure 4.

Emotional effects on novelty processing (NOV > STD in NEU context) < (NOV > STD in NEG context). (a) Bilateral STG activations for novelty processing in NEU context < novelty processing in NEG context. The plane coordinate of the slice is indicated in the upper right-hand corner. Bright colors in axial and coronal slides represent significance levels of contrasts, as indicated by the scale bar. (b) Average activity (±standard error of the mean) across conditions for the right STG cluster (mean x, y, z, 59, −10, −1, 295 voxels at P < 0.001, k = 20).

Activations obtained with the reverse contrast were expressed in bilateral fusiform gyrus, bilateral inferior temporal gyrus, bilateral middle occipital gyri, left middle frontal gyrus, and right superior parietal lobule (Table 4, Fig. 5). Analysis of the relative extent of activation, based on probabilistic maps, showed no activations in any of the 3 PAC subregions.

Table 4

Emotional effect on novelty processing: reverse contrast (novelty processing in NEU > novelty processing in NEG)

 Brain region Brodmann area z value Size (voxels) Coordinates x y z Novelty processing (NEU > NEG) R fusiform gyrus 19 3.51 43 50 −67 −13 R fusiform gyrus 37 3.83 53 −53 −14 L fusiform gyrus 19/37 3.73 122 −53 −65 −12 L fusiform gyrus 19 3.71 90 −22 −69 −12 L inferior temporal gyrus 20 3.90 46 −57 −26 −19 R middle temporal gyrus 39 3.57 72 50 −73 13 R inferior temporal gyrus/middle occipital gyrus 19 3.51 57 −66 −2 R inferior temporal gyrus 20 3.83 43 53 −53 −14 R middle occipital gyrus 19 3.95 72 53 −68 7 L middle occipital gyrus 18/19 3.61 29 −48 −78 −8 L superior frontal gyrus 22 3.46 22 −10 −4 68 L middle frontal gyrus 10 3.66 58 −38 58 −5 L middle frontal gyrus 11 3.65 −38 52 −13 L postcentral gyrus 2 3.72 21 −57 −22 32 L precentral gyrus 6 4.74 77 −55 1 20 L parahippocampal gyrus 19/37 4.05 25 −28 −45 −6 L lingual gyrus 18 3.71 90 −22 −69 −12 L caudate — 3.63 23 −14 20 8 R superior parietal lobule 7 3.47 32 28 −55 60
 Brain region Brodmann area z value Size (voxels) Coordinates x y z Novelty processing (NEU > NEG) R fusiform gyrus 19 3.51 43 50 −67 −13 R fusiform gyrus 37 3.83 53 −53 −14 L fusiform gyrus 19/37 3.73 122 −53 −65 −12 L fusiform gyrus 19 3.71 90 −22 −69 −12 L inferior temporal gyrus 20 3.90 46 −57 −26 −19 R middle temporal gyrus 39 3.57 72 50 −73 13 R inferior temporal gyrus/middle occipital gyrus 19 3.51 57 −66 −2 R inferior temporal gyrus 20 3.83 43 53 −53 −14 R middle occipital gyrus 19 3.95 72 53 −68 7 L middle occipital gyrus 18/19 3.61 29 −48 −78 −8 L superior frontal gyrus 22 3.46 22 −10 −4 68 L middle frontal gyrus 10 3.66 58 −38 58 −5 L middle frontal gyrus 11 3.65 −38 52 −13 L postcentral gyrus 2 3.72 21 −57 −22 32 L precentral gyrus 6 4.74 77 −55 1 20 L parahippocampal gyrus 19/37 4.05 25 −28 −45 −6 L lingual gyrus 18 3.71 90 −22 −69 −12 L caudate — 3.63 23 −14 20 8 R superior parietal lobule 7 3.47 32 28 −55 60

All coordinates reported in Talairach space. Activations shown are based on a voxelwise P < 0.001, uncorrected, k = 20.

Figure 5.

Emotional effects on novelty processing, reverse contrast (NOV > STD in NEU context) > (NOV > STD in NEG context). Bilateral (right shown) FUS (top, left corner, white circle), left middle frontal gyrus (middle, left corner, white circle), and right superior parietal lobule (bottom, left corner, white circle) activations for novelty processing in NEU context > novelty processing in NEG context. The plane coordinate of the slice is indicated in the upper right-hand corner. Bright colors in sagittal and axial slides represent significance levels of contrasts, as indicated by the scale bar. Average activity (±standard error of the mean) across conditions for the right FUS cluster (top, right corner, mean x, y, z, 53, −53, −14, 43 voxels at P < 0.001, k = 20), the left middle frontal gyrus cluster (middle, right corner, mean x, y, z, −38, 58, −5, 58 voxels at P < 0.001, k = 20), and the right superior parietal lobule cluster (bottom, right corner, mean x, y, z, 28, −55, 60, 32 voxels at P < 0.001, k = 20).

Figure 5.

Emotional effects on novelty processing, reverse contrast (NOV > STD in NEU context) > (NOV > STD in NEG context). Bilateral (right shown) FUS (top, left corner, white circle), left middle frontal gyrus (middle, left corner, white circle), and right superior parietal lobule (bottom, left corner, white circle) activations for novelty processing in NEU context > novelty processing in NEG context. The plane coordinate of the slice is indicated in the upper right-hand corner. Bright colors in sagittal and axial slides represent significance levels of contrasts, as indicated by the scale bar. Average activity (±standard error of the mean) across conditions for the right FUS cluster (top, right corner, mean x, y, z, 53, −53, −14, 43 voxels at P < 0.001, k = 20), the left middle frontal gyrus cluster (middle, right corner, mean x, y, z, −38, 58, −5, 58 voxels at P < 0.001, k = 20), and the right superior parietal lobule cluster (bottom, right corner, mean x, y, z, 28, −55, 60, 32 voxels at P < 0.001, k = 20).

## Discussion

### Modulation of Auditory Novelty Processing

The processing of stimulus salience depends not only on novelty or frequency of occurrence but also on the behavioral context (Katayama and Polich 1998) and, specifically, on the emotional relevance of the context. In the present study, we have shown that subjects were unable to fully ignore emotional information even when it was task irrelevant (Vuilleumier et al. 2001). Emotional faces were processed differently compared with NEU ones by activating the amygdala (LeDoux 2000; Vuilleumier et al. 2001; Adolphs and Tranel 2004) and, importantly, by modulating face-selective regions such as fusiform gyrus (George et al. 1999; Sugase et al. 1999; Vuilleumier et al. 2001) or early visual areas in the occipital lobe (Morris et al. 1998).

Novel auditory events activated a network of frontal (inferior frontal gyrus), temporal (STG), and parietal (precuneus) components, all areas known to be crucially involved in novelty processing (Downar et al. 2000, 2001; Bledowski et al. 2004). Despite the lack of consensus on the neural substrates required for novelty processing, some common results can be observed across oddball studies. Investigation using novelty oddball paradigms, both with auditory and visual novel events, has yielded novelty responses in inferior parietal (Clark et al. 2000), precuneus (Downar et al. 2001), superior temporal sulcus and gyrus (Alho et al. 1998; Clark et al. 2000; Downar et al. 2000; Kiehl et al. 2001), and prefrontal areas (IFG -Bledowski et al. 2004; Downar et al. 2001—dorsolateral—Bledowski et al. 2004). In general, the network observed in our results may subserve those mechanisms responsible for identification and evaluation of salient sensory stimuli in a complex environment, underlying a novelty evaluation system that refers to a mechanism concerned with the evaluation of novel stimuli that has already captured attention (Näätänen 1992; Friedman et al. 2001; Ranganath and Rainer 2003).

The combination of hemodynamic and behavioral results in our study led us to confirm that auditory novel sounds were processed differently with regard to the repetitive standard sounds (novelty processing), recruiting attentional resources from the ongoing task in subjects and causing a delay in the average performance (distraction) regardless of the emotional load within the task.

Furthermore, novel sounds, which already caused a delay on the subjects’ responses regardless of the context, elicited a stronger distraction effect when preceding and subsequent faces displayed a fearful or angry expression in contrast to NEU faces. In other words, when novel sounds were surrounded by NEG faces and thus processed in a NEG context. In this vein, bilateral superior temporal gyri, comprising bilateral planum temporale, bilateral secondary auditory cortex, and a portion of PAC, showed a stronger response when subjects processed auditory novel events in a NEG emotional context as compared with a NEU emotional context.

STG has been consistently linked to novelty processing (Alho et al. 1998; Opitz et al. 1999a, 1999b) and has been argued to subserve the novelty-P3 event-related brain potential (ERP; Opitz et al. 1999b). This neuroelectric pattern of activation responds to a positive-going deflection that peaks around 300 ms and reflects orienting response toward novel auditory events (Escera et al. 1998, 2000). Temporoparietal lesions, centered in the superior temporal cortex, have been shown to attenuate P3 to novel sounds (Knight et al. 1989). Specifically, PAC and planum temporale have been related to the initial detection and detailed analysis of changes in the acoustic environment, respectively (Ulanovsky et al. 2003; Schönwieser et al. 2007).

Our results did not reveal any modulatory effect on novelty processing caused by emotion within or in the vicinity of other areas also known to be involved in novelty processing (activations that we indeed observed in every single context separately), such as IFG (Downar et al. 2000). According to the novelty encoding hypothesis (Tulving et al. 1996), temporal regions would provide the input for frontal encoding networks and, thus, STG would be involved in novelty detection, whereas accessing and retrieving semantic concepts related to novel sounds would additionally engage prefrontal cortex (Opitz et al. 1999b). Neuropsychological data (Knight 1984) suggest that the prefrontal cortex might not be a primary P3 generator. All these arguments and the partial activation of PAC led us to suggest that an emotional context modulates novelty processing in the very primary stage of detection.

However, it is still unclear whether these modulated areas correspond to a unimodal network related to auditory novelty processing or to a multimodal circuitry common in multisensory novelty processes. Superior temporal cortices have been claimed to be specific for the auditory sensory modality in novelty and target detection (Downar et al. 2000; Kiehl et al. 2001). Indeed, the coordinates of STG related to unimodal auditory processing reported before (Downar et al. 2000; Talairach [x, y, z] 53/−21/5; −51/−37/10; −53/−15/3) are close to areas identified in the present study and, similarly to our activations, were bilaterally distributed.

On the other hand, some of these regions located in STG could also correspond to the areas described previously as temporoparietal junction (Downar et al. 2000, 2001, 2002). It has been proposed that this brain region plays a general role in identifying salient stimuli in the sensory environment across multiple modalities (Downar et al. 2000, 2002). A lesion study concerning spatial neglect (Karnath et al. 2001), a disorder typically associated with lesions in the posterior parietal lobe, suggested STG to be the neural substrate for spatial awareness in humans. The coordinates corresponding to BA22 and BA42 reported in this study are located close to ours. However, all these results pointed to a right hemisphere lateralization, postulating a phylogenetic transition of this function from the bilateral distribution seen in the monkey brain (Karnath et al. 2001). Given this evidence, the data in our study support the idea that both unimodal and multimodal novelty processing areas may be modulated by the emotional context.

Moreover, when performing the 1-way ANOVA for emotional effects on novelty processing in the reverse order to test which areas were more activated during novelty processing in the NEU context than in the NEG one (the reverse contrast), we observed activations located in fusiform, inferior temporal, and middle occipital gyri (Rolls 1992; George et al. 1999). Previous studies using similar auditory–visual paradigms have suggested that the involuntary switching of attention to changes in the acoustic environment can interfere with early processing of the successive visual stimuli (Alho et al. 1997). Therefore, this finding suggests that a context of emotional faces not only modulated novelty processing areas but also made novel sounds especially attenuate the processing of subsequent visual target stimuli. This effect was consistent with the behavioral data.

In conclusion, emotion processing not only modulates perception when presented within a certain sensory pathway, as has been previously proposed (Öhman et al. 2001; Fox 2002), but also might exert a strong influence on the processing of other sensory signals presented concomitantly through an effect of facilitation. This modulation would be independent from voluntary mechanisms of attentional control but would still respond to common top–down regulation (Vuilleumier 2005). The amygdala should indirectly play a critical role in this attentional modulation. It is known to receive sensory inputs from all modalities and to send projections toward different cortical and subcortical areas (Holland and Gallagher 1999). Animal studies focused on fear conditioning to a simple auditory conditioned stimulus postulate that both auditory thalamus and auditory cortex send inputs to the lateral nucleus of the amygdala (Romanski and LeDoux 1993). Furthermore, they point to the importance of the direct thalamo-amygdala pathway in this learning process (Morris et al. 1999). The direct or indirect connections between auditory pathways and the amygdala might be directly involved in the STG enhancement we observed.

The results of the present experiment are in agreement with recent findings in ERP studies, where similar experimental tasks were used, revealing an enhancement of auditory novelty-P3 amplitude in the NEG context in relation to the NEU context (Domínguez-Borràs et al. 2008a, 2008b), but differ from previous electrophysiological studies using P3 and startle reflex measurements (Schupp et al. 1997; Cuthbert et al. 1998; Bradley et al. 2006; Keil et al. 2007). In the latter studies, the amplitude of P3 to the startle probes appeared smaller when blinks were potentiated by the processing of unpleasant pictures (Schupp et al. 1997; Cuthbert et al. 1998; Bradley et al. 2006; Keil et al. 2007) even to unexpected simple tones (Cuthbert et al. 1998), suggesting a greater allocation of attentional resources to the affective visual stimuli in a limited-capacity system (Keil et al. 2007). These effects seem contrary to the modulatory effects we observed in STG. The reasons for this disagreement are still unclear but, however, any comparison between these studies and the experiment here reported should be taken with extreme caution. In startle experiments, probes were identical across presentations, whereas we used 100 different novel sounds. Although both startle probes and novel sounds appeared unexpectedly in all studies and, thus, can be treated as novel auditory events, the structure of the audio–visual oddball paradigm used here differs notably from that used by startle reflex measurements. For instance, in those experiments, sounds appeared while the picture was on screen and the stimulus duration was of 1.5 or 6 s. In our study, sounds appeared when no image was being displayed. With this and the short duration of the stimuli, we manipulated the emotional load of the task while the images and sounds never overlapped, and attentional resources were never forced to compete. This trial structure was optimal to favor novelty processing and behavioral distraction (Escera et al. 1998, 2000; Escera and Corral 2007.

### Potential Limitations and Shortcomings

It is important to note that it may be difficult to dissociate between activations related to novelty specifically from those related to spectrotemporal processing, as the novel sounds differed widely from the standard ones in this regard. Moreover, it has been demonstrated that lateral belt areas of rhesus monkey auditory cortex are more responsive to complex sounds than to pure tones (Rauschecker et al. 1995). However, the novel sounds also differed in their probability of occurrence and may therefore be considered as contextually novel (see Ranganath and Rainer 2003). According to magnetoencephalography (MEG) data, auditory cortex activation related to these types of novel sounds (actually to a similar set of sounds as the one used here) was located halfway between the MEG source of N1 and that of the mismatch field, reflecting novelty detection (Alho et al. 1998). Moreover, the fact that the novel sounds presented in the NEU context were statistically similar to those presented in the NEG context, both in loudness (RMS level) and spectrotemporal features (Cepstral Coefficients; see Supplementary Material), would still demonstrate that a NEG emotional context modulates the processing of novel sounds in auditory cortical areas.

Furthermore, the possibility that differential brain activations during emotional face processing could be caused by differences in basic visual features is still present, although any effects related to emotional face processing disappeared, in other studies, when faces were inverted (and, therefore, the holistic face perception was reduced—Fox et al. 2000).

In addition, it is worth emphasizing that the emotional effects on auditory processing observed in the present study were elicited by faces displaying an expression of fear and anger. It would therefore be rash to generalize and suggest that the same effects could also be elicited by a task loaded with any other emotional valence.

Finally, to fully assume the role of the amygdala both in the enhancement of the STG and a decrease of activations in different cortical areas, further research should address this issue focussing on functional connectivity and correlations between the amygdala activation and behavioral effects.

## Conclusions

Taken together, the results obtained in the present study led us to propose that processing NEG emotionally salient information would alter cortical excitability in STG and thus would define this temporal region as an emotional context-dependent area in the auditory novelty system. When mediated by emotion, specific unimodal and multimodal regions of the STG would enhance processing of novel auditory events in novelty-related areas. These events, irrelevant in NEU environmental conditions, could potentially convey crucial information in a context of affective relevance and, through the mechanisms described in this study, would warranty greater chances of survival.

## Supplementary Material

Supplementary material can be found at: http://www.cercor.oxfordjournals.org/.

## Funding

Spanish Ministry of Education and Science (SEJ2006-00496/PSIC; Consolider-Ingenio 2010 CSD 2007-00012; HA2005-0024); Generalitat de Catalunya (SGR2005-00953).

Conflict of Interest: None declared.

## References

R
Tranel
D
Impaired judgments of sadness but not happiness following bilateral amygdala damage
J Cogn Neurosci
,
2004
, vol.
16

3
(pg.
453
-
462
)
Alho
K
Escera
C
Díaz
R
Yago
E
Serra
JM
Effects of involuntary auditory attention on visual task performance and brain activity
NeuroReport
,
1997
, vol.
8
(pg.
3233
-
3237
)
Alho
K
Winkler
I
Escera
C
Huotilainen
M
Virtanen
J
Jääskelainen
IP
Pekkonen
E
Ilmoniemi
RJ
Processing of novel sounds and frequency changes in the human auditory cortex: magnetoencephalographic recordings
Psychophysiology
,
1998
, vol.
35
(pg.
211
-
224
)
Anderson
AK
Phelps
EA
Expression without recognition: contributions of the human amygdala to emotional communication
Psychol Sci
,
2000
, vol.
11
(pg.
106
-
111
)
Barrett
AM
Crucian
GP
Raymer
AM
Heilman
KM
Spared comprehension of emotional prosody in a patient with global aphasia
Neuropsychiatry Neuropsychol Behav Neurol
,
1999
, vol.
12

2
(pg.
117
-
120
)
Bledowski
C
Prvulovic
D
Hoechstetter
K
Scherg
M
Wibral
M
Goebel
R
Linden
DEJ
Localizing P300 generators in visual target and distractor processing: a combined event-related potential and functional magnetic resonance imaging study
J Neurosci
,
2004
, vol.
24

42
(pg.
9353
-
9360
)
MM
Codispoti
M
Lang
PJ
A multi-process account of startle modulation during affective perception
Psychophysiology
,
2006
, vol.
43
(pg.
486
-
497
)
Brett
M
Johnsrude
IS
Owen
AM
The problem of functional localization in the human brain
Nat Rev Neurosci
,
2002
, vol.
3
(pg.
243
-
249
)
Carretié
L
Hinojosa
JA
Martín-Loeches
M
F
Tapia
M
Automatic attention to emotional stimuli: neural correlates
Hum Brain Mapp
,
2004
, vol.
22
(pg.
290
-
299
)
Clark
VP
Fannon
S
Lai
S
Benson
R
Bauer
L
Responses to rare visual target and distractor stimuli using event-related fMRI
J Neurophysiol
,
2000
, vol.
83
(pg.
3133
-
3139
)
Cuthbert
BN
Schupp
HT
MM
McManis
M
Lang
PJ
Probing affective pictures: attended startle and tone probes
Psychophysiology
,
1998
, vol.
35
(pg.
344
-
347
)
Domínguez-Borràs
J
Garcia-Garcia
M
Escera
C
Emotional context enhances auditory novelty processing: behavioural and electrophysiological evidence
Eur J Neurosci
,
2008
, vol.
28

6
(pg.
1199
-
206
)
Domínguez-Borràs
J
Garcia-Garcia
M
Escera
C
Negative emotional context enhances auditory novelty processing
Neuroreport
,
2008
, vol.
19
(pg.
503
-
507
)
Downar
J
Crawley
AP
Mikulis
DJ
Davis
KD
A multimodal cortical network for the detection of changes in the sensory environment
Nat Neurosci
,
2000
, vol.
3

3
(pg.
277
-
283
)
Downar
J
Crawley
AP
Mikulis
DJ
Davis
KD
The effect of task relevance on the cortical response to changes in visual and auditory stimuli: an event-related fMRI study
NeuroImage
,
2001
, vol.
14
(pg.
1256
-
1267
)
Downar
J
Crawley
AP
Mikulis
DJ
Davis
KD
A cortical network sensitive to stimulus salience in a neutral behavioral context across multiple sensory modalities
J Neurophysiol
,
2002
, vol.
87
(pg.
615
-
620
)
Eastwood
JD
Smilek
D
Merikle
PM
Differential attentional guidance by unattended faces expressing positive and negative emotion
Percept Psychophys
,
2001
, vol.
63
(pg.
1004
-
1013
)
Eickhoff
SB
Paus
T
Caspers
S
Grosbras
MH
Evans
A
Zilles
K
Amunts
K
Assignment of functional activations to probabilistic cytoarchitectonic areas revisited
NeuroImage
,
2007
, vol.
36
(pg.
511
-
521
)
Eickhoff
SB
Stephan
KE
Mohlberg
H
Grefkes
C
Fink
G
Amunts
K
Zilles
K
A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data
NeuroImage
,
2005
, vol.
25
(pg.
1325
-
1335
)
Escera
C
Alho
K
Schröger
E
Winkler
I
Involuntary attention and distractibility as evaluated with event-related brain potentials
Audiol Neurootol
,
2000
, vol.
5
(pg.
151
-
166
)
Escera
C
Alho
K
Winkler
I
Näätäänen
R
Neural mechanisms of involuntary attention to acoustic novelty and change
J Cogn Neurosci
,
1998
, vol.
10

5
(pg.
590
-
604
)
Escera
C
Corral
MJ
Role of mismatch negativity and novelty-P3 in involuntary auditory attention
Int J Psychophysiol.
,
2007
, vol.
21
(pg.
251
-
264
)
Escera
C
Corral
MJ
Yago
E
An electrophysiological and behavioral investigation of involuntary attention towards auditory frequency, duration and intensity changes
Cogn Brain Res
,
2002
, vol.
14
(pg.
325
-
332
)
Escera
C
Yago
E
Alho
K
Electrical responses reveal the temporal dynamics of brain events during involuntary attention switching
Eur J Neurosci
,
2001
, vol.
14
(pg.
877
-
883
)
Escera
C
Yago
E
Corral
MJ
Corbera
S
Núñez
MI
Attention capture by auditory significant stimuli: semantic analysis follows attention switching
Eur J Neurosci
,
2003
, vol.
18
(pg.
2408
-
2412
)
Fox
E
Processing of emotional facial expressions: the role of anxiety and awareness
Cogn Affect Behav Neurosci
,
2002
, vol.
2
(pg.
52
-
63
)
Fox
E
Lester
V
Russo
R
Bowles
RJ
Pichler
A
Dutton
K
Facial expressions of emotion: are angry faces detected more efficiently?
Cogn Emot
,
2000
, vol.
14
(pg.
61
-
92
)
Friedman
D
Cycowicz
YM
Gaeta
H
The novelty P3: an event-related brain potential (ERP) sign of the brain's evaluation of novelty
Neurosci Biobehav Rev
,
2001
, vol.
25
(pg.
355
-
373
)
George
N
Dolan
RJ
Fink
GR
Baylis
GC
Russell
C
Driver
J
Contrast polarity and face recognition in the human fusiform gyrus
Nat Neurosci
,
1999
, vol.
2
(pg.
574
-
580
)
Holland
PC
Gallagher
M
Amygdala circuitry in attentional and representational processes
Trends Cogn Sci
,
1999
, vol.
3
(pg.
65
-
73
)
Karnath
HO
Ferber
S
Himmelbach
M
Spatial awareness is a function of the temporal not the posterior parietal lobe
Nature
,
2001
, vol.
411
(pg.
950
-
953
)
Katayama
JI
Polich
J
Stimulus context determines P3a and P3b
Psychophysiology
,
1998
, vol.
35
(pg.
23
-
33
)
Keil
A
MM
Junghöfer
M
Russmann
T
Lowenthal
W
Lang
PJ
Cross-modal attention capture by affective stimuli: evidence from event-related potentials
Cogn Affect Behav Neurosci
,
2007
, vol.
7

1
(pg.
18
-
24
)
Kiehl
KA
Laurens
KR
Duty
TL
Forster
BB
Liddle
PF
Neural sources involved in auditory target detection and novelty processing: an event-related fMRI study
Psychophysiology
,
2001
, vol.
38
(pg.
133
-
142
)
Knight
RT
Decreased response to novel stimuli after prefrontal lesions in man
Electroencephalogr Clin Neurophysiol
,
1984
, vol.
59
(pg.
9
-
20
)
Knight
RT
Scabini
D
Woods
DL
Clayworth
C
Contributions of the temporoparietal junction to the human auditory P3
Brain Res
,
1989
, vol.
502
(pg.
109
-
116
)
LeDoux
J
Fear and the brain: where have we been, and where are we going?
Biol Psychiatry
,
1998
, vol.
44
(pg.
1229
-
1238
)
LeDoux
JE
Lane
RD
IL
Cognitive-emotional interactions: listen to the brain
Cognitive neuroscience of emotion
,
2000
Oxford
Oxford University Press
(pg.
129
-
155
)
Leslie
KR
Johnson-Frey
SH
Grafton
ST
Functional imaging of face and hand imitation: towards a motor theory of empathy
NeuroImage
,
2004
, vol.
21

2
(pg.
601
-
607
)
Morosan
P
J
Schleicher
A
Amunts
K
Schormann
T
Zilles
K
Human primary auditory cortex: cytoarchitectonic subdivisions and mapping into a spatial reference system
NeuroImage
,
2001
, vol.
13
(pg.
684
-
701
)
Morris
JS
Friston
KJ
Buchel
C
Frith
CD
Young
AW
Calder
AJ
Dolan
RJ
A neuromodulatory role for the human amygdala in processing emotional facial expressions
Brain
,
1998
, vol.
121
(pg.
47
-
57
)
Morris
JS
Ohman
A
Dolan
RJ
A subcortical pathway to the right amygdala mediating ‘‘unseen’’ fear
,
1999
, vol.
96
(pg.
1680
-
1685
)
Näätänen
R
Attention and brain function
,
1992
Hillsdale, NJ
Erlbaum
pg.
494

Öhman
A
Lundqvist
D
Esteves
F
The face in the crowd revisited: a threat advantage with schematic stimuli
J Pers Soc Psychol
,
2001
, vol.
80
(pg.
381
-
396
)
Opitz
B
Mecklinger
A
Friederici
von Cramon
DY
The functional neuroanatomy of novelty processing: integrating ERP and fMRI results
Cereb Cortex
,
1999
, vol.
9
(pg.
379
-
391
)
Opitz
B
Mecklinger
A
von Cramon
DY
Kruggel
F
Combining electrophysiological and hemodynamic measures of the auditory oddball
Psychophysiology
,
1999
, vol.
36
(pg.
142
-
147
)
Penhune
VB
Zatorre
RJ
MacDonald
JD
Evans
AC
Interhemispheric anatomical differences in human primary auditory cortex: probabilistic mapping and volume measurement from magnetic resonance scans
Cereb Cortex
,
1996
, vol.
6
(pg.
661
-
672
)
J
Morosan
P
Schleicher
A
Freund
HJ
Zilles
K
Human primary auditory cortex in women and men
NeuroReport
,
2001
, vol.
12

8
(pg.
1561
-
1565
)
Ranganath
C
Rainer
G
Neural mechanisms for detecting and remembering novel events
Nat Rev Neurosci
,
2003
, vol.
4

3
(pg.
193
-
202
)
Rauschecker
JP
Tian
B
Hauser
M
Processing of complex sounds in the macaque nonprimary auditory cortex
Science
,
1995
, vol.
268

5207
(pg.
111
-
114
)
Richards
A
Blanchette
I
Independent manipulation of emotion in an emotional stroop task using classical conditioning
Emotion
,
2004
, vol.
4

3
(pg.
275
-
281
)
Rolls
ET
Neurophysiological mechanisms underlying face processing within and beyond the temporal cortical visual areas
Philos Trans R Soc Lond B Biol Sci
,
1992
, vol.
335
(pg.
11
-
20
)
Romanski
LM
LeDoux
JE
Information cascade from primary auditory cortex to the amygdala: corticocortical and corticoamygdaloid projections of temporal cortex in the rat
Cereb Cortex
,
1993
, vol.
3

6
(pg.
515
-
532
)
Schönwieser
M
Novitski
N
Pakarinen
S
Carlson
S
Tervaniemi
M
Näätänen
R
Heschl's gyrus, posterior superior temporal gyrus, and mid-ventrolateral prefrontal cortex have different roles in the detection of acoustic changes
J Neurophysiol
,
2007
, vol.
97
(pg.
2075
-
2082
)
Schupp
HT
Cuthbert
BN
MM
Birbaumer
N
Lang
PJ
Probe P3 and blinks: two measures of affective startle modulation
Psychophysiology
,
1997
, vol.
34
(pg.
1
-
6
)
Stanley
J
Knight
RG
Emotional specificity of startle potentiation during the early stages of picture viewing
Psychophysiology
,
2004
, vol.
41
(pg.
935
-
940
)
Sugase
Y
Yamane
S
Ueno
S
Kawano
K
Global and fine information coded by single neurons in the temporal visual cortex
Nature
,
1999
, vol.
400
(pg.
869
-
873
)
Talairach
J
Tournoux
PA
Co-planar stereotactic atlas of the human brain
,
1988

Stuttgart (Germany): Thieme Medical
Tulving
E
Markowitsch
HJ
Craik
FIM
Habib
R
Houle
S
Novelty and familiarity activations in PET studies of memory encoding and retrieval
Cereb Cortex
,
1996
, vol.
6
(pg.
71
-
79
)
Ulanovsky
N
Las
L
Nelken
I
Processing of low-probability sounds by cortical neurons
Nat Neurosci
,
2003
, vol.
6

4
(pg.
391
-
398
)
Vuilleumier
P
How brains beware: neural mechanisms of emotional attention
Trends Cogn Sci
,
2005
, vol.
9

12
(pg.
585
-
592
)
Vuilleumier
P
Armony
JL
Driver
J
Dolan
RJ
Effects of attention and emotion on face processing in the human brain: an event-related fMRI study
Neuron
,
2001
, vol.
30
(pg.
829
-
841
)