Abstract

Visual threat-related signals are not only processed via a cortical geniculo-striatal pathway to the amygdala but also via a subcortical colliculo-pulvinar-amygdala pathway, which presumably mediates implicit processing of fearful stimuli. Indeed, hemianopic patients with unilateral damage to the geniculo-striatal pathway have been shown to respond faster to seen happy faces in their intact visual field when unseen fearful faces were concurrently presented in their blind field [Bertini, C., Cecere, R., & Làdavas, E. I am blind, but I “see” fear. Cortex, 49, 985–993, 2013]. This behavioral facilitation in the presence of unseen fear might reflect enhanced processing of consciously perceived faces because of early activation of the subcortical pathway for implicit fear perception, which possibly leads to a modulation of cortical activity. To test this hypothesis, we examined ERPs elicited by fearful and happy faces presented to the intact visual field of right and left hemianopic patients, whereas fearful, happy, or neutral faces were concurrently presented in their blind field. Results showed that the amplitude of the N170 elicited by seen happy faces was selectively increased when an unseen fearful face was concurrently presented in the blind field of right hemianopic patients. These results suggest that when the geniculo-striate visual pathway is lesioned, the rapid and implicit processing of threat signals can enhance facial encoding. Notably, the N170 modulation was only observed in left-lesioned patients, favoring the hypothesis that implicit subcortical processing of fearful signals can influence face encoding only when the right hemisphere is intact.

INTRODUCTION

As fearful stimuli signal potential threats for the organism, fast processing of fear-related information has a pivotal role for survival, learning, and adaptive behavior. Indeed, rapid detection of potential threats, which enables the immediate activation of defensive, aggressive, or aversive motor responses, is highly advantageous from an evolutionary perspective.

The amygdala is a core structure for perception and recognition of fear-related information (Öhman, 2005; Zald, 2003; Davis & Whalen, 2001; Whalen, 1998) in particular fearful faces (Hariri, Tessitore, Mattay, Fera, & Weinberger, 2002; Anderson & Phelps, 2000; Sprengelmeyer et al., 1999; Broks et al., 1998; Calder et al., 1996; Adolphs, Tranel, Damasio, & Damasio, 1995). Citing evidence from animal fear conditioning, lesion, and neuroimaging studies, LeDoux (1996) suggested that sensory information about fearful stimuli is conveyed to the amygdala by a dual-route system: a direct, subcortical and short-latency “low road” from the superior colliculus-pulvinar pathway to the amygdala and an indirect cortical “high road” from the LGN of the thalamus via the striate cortex to the amygdala. These two pathways are functionally different in terms of speed and accuracy, with the “low road” allowing faster but cruder processing of fearful information and the “high road” being slower but also more detailed.

Previous investigations of the functional properties of the “low-road” have focused on implicit visual processing of emotional faces in patients with visual field defects (Bertini, Cecere, & Làdavas, 2013; Pegna, Khateb, Lazeyras, & Seghier, 2005; de Gelder, Pourtois, & Weiskrantz, 2002; de Gelder, Pourtois, van Raamsdonk, Vroomen, & Weiskrantz, 2001; de Gelder, Vroomen, Pourtois, & Weiskrantz, 1999), in whom the “high road” to the amygdala is lesioned but the “low road” is spared and functional. Interestingly, different populations of hemianopic patients, namely blindsight patients (de Gelder et al., 2001) and hemianopes without blindsight (Bertini et al., 2013), although both reporting no subjective visual awareness for stimuli presented in the blind field, showed different performances in tasks with redundant emotional face stimuli, where implicit processing of unseen stimuli is inferred from their effect on the response to seen stimuli. In particular, the study by de Gelder et al. (2001) on blindsight patient G.Y. reported a congruency effect (i.e., faster RTs) when he responded to fearful or sad emotional faces in his intact visual field that were coupled with emotionally identical faces in his blind visual field. In contrast, the study by Bertini et al. (2013) failed to report any congruency effect, showing that RTs of hemianopes without blindsight to both emotional (i.e., happy) and neutral target faces in their intact visual field were selectively reduced when fearful faces, but not happy or scrambled faces, were concurrently presented in their blind visual field. Conversely, no effect of unseen fearful faces was found when they were coupled with seen fearful faces in the intact visual field. Moreover, in contrast to blindsight patients (e.g., de Gelder et al., 1999, 2001), these patients performed at chance level when explicitly asked to guess the content of unseen stimuli in their blind visual field (two-alternative forced-choice task [2AFC]). These different performances in tasks with redundant emotional faces and in 2AFC tasks suggest that different neural circuits might account for the visual residual abilities of the two groups of patients. In particular, the implicit visual processing of blindsight patients might rely both on the “low” and the “high road,” possibly because of specific postlesion plasticity phenomena promoting functional reorganization within the cortical visual pathway. By contrast, hemianopes without blindsight might be able to recruit only the spared subcortical “low road” bypassing the lesioned cortex. The fear-specific effects shown by hemianopes without blindsight (Bertini et al., 2013) suggest that when the striate cortex is damaged and only the subcortical visual pathway is available, only unseen fearful signals are rapidly and implicitly processed by this alternative “low-road” and influence explicit processing of seen faces mediated by cortical circuits. Yet, it is not clear at which stage of cortical face processing the “low road” activity exerts its modulatory effect. However, considering that the “low road” performs rapid and coarse analysis of visual threat, its activity could modulate face processing already at an early stage, that is, at the stage of structural face encoding. Specifically, rapid and nonconscious fear perception mediated by the “low road” might enhance face encoding, which in turn could drive the behavioral facilitation observed in response to seen faces when coupled with unseen fearful faces (Bertini et al., 2013).

To investigate the effect of implicit fear perception on conscious face processing, in this study we recorded EEG in right and left hemianopic patients during a task requiring them to categorize the emotional expression (fearful or happy) of faces presented in their intact visual field, while fearful, happy or neutral faces were concurrently presented in their blind field. To ensure that the patients were totally unaware of visual stimuli presented in their blind field, before the main experiment, we administered a series of tests of visual abilities, such as visual detection, shape discrimination, and discrimination of the gender and the emotional expression of faces. Chance-level performance in all of these tests was considered an inclusion criterion. In the main experiment, we recorded ERPs elicited by seen (fearful or happy) faces during the emotional categorization task to test whether they could be specifically modulated by the concurrent presentation of unseen fearful faces. On the basis of our hypothesis that ERP modulation because of the implicit processing of unseen fearful faces could be already evident at early stages of face processing, we focused on the N170 component of ERPs, which has been suggested to reflect the rapid structural encoding of faces (Itier & Taylor, 2004; Bentin & Deouell, 2000; Bentin, Allison, Puce, Perez, & McCarthy, 1996) and to be modulated by emotional expressions, both when they are explicitly (Stekelenburg & de Gelder, 2004; Batty & Taylor, 2003) and implicitly perceived (Pegna, Darque, Berrut, & Khateb, 2011; Pegna, Landis, & Khateb, 2008). In particular, studies on healthy participants using backward masked faces to prevent overt recognition of stimuli reported N170 enhancement in response to subliminally perceived fearful faces compared with nonfearful faces (Smith, 2012; Pegna et al., 2008, 2011).

In line with our previous study on hemianopic patients reporting faster responses to seen happy faces when paired with unseen fearful faces (Bertini et al., 2013) and according to the hypothesis that such faster responses might reflect enhanced face processing, we predicted an N170 enhancement in response to (seen) happy faces in the intact field only when presented together with (unseen) fearful faces in the blind field. In contrast, no N170 modulation was expected by happy or neutral faces in the blind field or by unseen fearful faces in the blind field that were paired with seen fearful faces in the intact field. Moreover, the expected N170 modulation should be only evident in patients with right hemianopia, that is, in patients with intact right hemisphere. This was hypothesized because right hemisphere dominance for the regulation of the emotional responses has been a well-established notion since early studies on cerebral asymmetries in emotional processing (Gainotti, Caltagirone, & Zoccolotti, 1993; Làdavas, Cimatti, Del Pesce, & Tuozzi, 1993; Gainotti, 1972).

METHODS

Participants

Seven patients (three women; mean age = 50.1 years; range = 28–72 years) with chronic right visual field defect and seven patients (one woman; mean age = 52 years; range = 40–59 years) with chronic left visual field defect gave their written informed consent to participate in the study, which was approved by the Departmental Ethics Committee and was designed in accordance with the ethical principles of the Declaration of Helsinki. Patients P2, P3, P4, P5, P6, P10, and P14 had already participated in our previous behavioral study (Bertini et al., 2013). All patients presented deafferentation or destruction of primary visual areas consequent to postgeniculate lesions, as documented by CT and MRI scans (see Table 1 for lesion details and Figure 1 for CT or MRI scans).

Table 1. 

Summary of Clinical, Demographic, and Lesional Data

CaseSexAgeYears of EducationTime since Onset (months)Visual Field DefectEtiologyLesion Site
P1 72 22 Right hemianopia Vascular L temporal-occipital 
P2 40 17 239 Right hemianopia Traumatic L temporal-occipital 
P3 34 11 Right hemianopia Vascular L parietal-occipital 
P4 60 13 Right hemianopia Vascular L occipital 
P5 44 Lower right quadrantopia Vascular L parietal-occipital 
P6 29 13 Upper right quadrantopia Vascular (AVM) L temporal-parietal-occipital 
P7 72 10 Upper right quadrantopia Vascular L temporal-parietal-occipital 
P8 56 13 12 Left hemianopia Vascular R temporal-parietal-occipital 
P9 48 18 34 Left hemianopia Vascular R temporal-occipital 
P10 56 16 55 Left hemianopia Vascular R temporal-occipital 
P11 47 13 Left hemianopia Vascular R temporal-parietal-occipital 
P12 40 13 Left hemianopia Vascular (AVM) R temporal-parietal 
P13 58 Left hemianopia Astrocytoma R parietal-occipital 
P14 59 13 Lower left quadrantopia Vascular R temporal-parietal-occipital 
CaseSexAgeYears of EducationTime since Onset (months)Visual Field DefectEtiologyLesion Site
P1 72 22 Right hemianopia Vascular L temporal-occipital 
P2 40 17 239 Right hemianopia Traumatic L temporal-occipital 
P3 34 11 Right hemianopia Vascular L parietal-occipital 
P4 60 13 Right hemianopia Vascular L occipital 
P5 44 Lower right quadrantopia Vascular L parietal-occipital 
P6 29 13 Upper right quadrantopia Vascular (AVM) L temporal-parietal-occipital 
P7 72 10 Upper right quadrantopia Vascular L temporal-parietal-occipital 
P8 56 13 12 Left hemianopia Vascular R temporal-parietal-occipital 
P9 48 18 34 Left hemianopia Vascular R temporal-occipital 
P10 56 16 55 Left hemianopia Vascular R temporal-occipital 
P11 47 13 Left hemianopia Vascular R temporal-parietal-occipital 
P12 40 13 Left hemianopia Vascular (AVM) R temporal-parietal 
P13 58 Left hemianopia Astrocytoma R parietal-occipital 
P14 59 13 Lower left quadrantopia Vascular R temporal-parietal-occipital 

M = male; F = female; AVM = arteriovenous malformation; L = left; R = right.

Figure 1. 

CT or MRI scans of patients. All the scans are presented in axial views except for those of P9, which are presented in coronal view. R = right; L = left.

Figure 1. 

CT or MRI scans of patients. All the scans are presented in axial views except for those of P9, which are presented in coronal view. R = right; L = left.

Mapping of brain lesions was performed using MRIcron (Rorden, Karnath, & Bonilha, 2007; Rorden & Brett, 2000). Lesions as documented by the most recent clinical CT or MRI were traced on the T1-weighted template MRI scan from the Montreal Neurological Institute provided with the MRIcron software (Rorden et al., 2007; Rorden & Brett, 2000). Lesion volumes were computed for each patient, and the extents of the lesions were compared between the two groups, revealing no significant differences between left (16,286 mm3) and right (31,707 mm3) damage patients, t(12) = −1.77; p = .1. Probability maps, revealing the localization of areas of maximal lesion overlap, were generated for patients with left and right lesions separately, averaging the individual lesion volumes. Patients with left lesions revealed a maxov area (overlap: 57%) in the occipital lobe–lingual gyrus (Talairach coordinates: x = −16 y = −75 z = −3; Figure 2A). The maxov areas for patients with right lesions (overlap: 42%) were located in the subgyral white matter of the temporal lobe (Talairach coordinates: x = 38 y = −45 z = 0; Figure 2B) and in the occipital lobe–Brodmann's area 30 (Talairach coordinates: x = 12 y = −68 z = 12; Figure 2B). To compare Talairach coordinates of maxov between the two groups, lesion volumes of patients with left lesions were flipped to the right side. The frequency of occurrence of damage at Talairach coordinates of maxov areas between the two groups were compared with Fisher exact test, revealing no significant difference between left and right damage patients (all ps > .19). All patients showed lesions to cortical areas excluding amygdala, posterior thalamus, and superior colliculus (Figure 2C, D, E). All patients suffered from complete right or left homonymous hemianopia, except for P5 who had a lower right quadrantopia, P6 and P7 who had an upper right quadrantopia, and P14 who had a lower left quadrantopia. The presence of visual field blindness was confirmed by automated perimetry, whereas visual acuity in the intact visual field was normal or corrected-to-normal in all patients. None of the participants presented any coexisting neurological or psychiatric disorder or cognitive deficit.

Figure 2. 

Location and overlap of brain lesions. (A, B) The image shows the lesions of the left lesioned patients (A) and right lesioned patients (B) projected on the same four axial slices of the standard MNI brain. In each slice, the left hemisphere is on the left side. The level of axial and coronal slices has been marked by white lines on the sagittal view of the brain. The color bar indicates the number of overlapping lesions. (C–E) Overlap of lesions of both left and right lesioned patients projected of the axial slices where the amygdala (C), the posterior thalamus (D), and the superior colliculus (E) are visible. The arrows indicate the amygdala (C), the posterior thalamus (D), and the superior colliculus (E).

Figure 2. 

Location and overlap of brain lesions. (A, B) The image shows the lesions of the left lesioned patients (A) and right lesioned patients (B) projected on the same four axial slices of the standard MNI brain. In each slice, the left hemisphere is on the left side. The level of axial and coronal slices has been marked by white lines on the sagittal view of the brain. The color bar indicates the number of overlapping lesions. (C–E) Overlap of lesions of both left and right lesioned patients projected of the axial slices where the amygdala (C), the posterior thalamus (D), and the superior colliculus (E) are visible. The arrows indicate the amygdala (C), the posterior thalamus (D), and the superior colliculus (E).

Apparatus

During the experimental sessions, patients were placed 57 cm distant from a 17-in. PC monitor (refresh rate 60 Hz) in a quiet and dimly lit room. Eye movements were constantly monitored by a Pan/Tilt optic eye-tracker (Eye-Track ASL-6000; sampling rate 60 Hz) during the test of residual visual abilities, whereas an EOG was used during EEG registration. Stimuli were presented on a PC running Presentation software (Version 0.60, www.neurobs.com). Patients were asked to hold their gaze on a central fixation crosshair (2°). For patients with quadrantopia, the fixation spot was displaced upwards or downwards (11° above or below the midline) to ensure the display of stimuli in the blind field.

Test of Residual Visual Abilities

Patients' residual visual abilities in the blind field were tested in four separate sessions of a 2AFC task using different stimuli (Figure 3). In the visual detection task (Figure 3A), the stimulus was a white dot (2° diameter). In the emotional task (Figure 3B), stimuli consisted of 12 grayscale photographs (six men, six women) showing happy or fearful facial expressions. In the gender task (Figure 3C), six grayscale photographs of male and female models with a neutral expression were used. The pictures displayed both in the emotional and in the gender task were the same size (7.5° × 11°) and were taken from the Pictures of Facial Affect set (Ekman & Friesen, 1976). In the shape task (Figure 3D), white-colored squares (5° × 5°) and circles (5° diameter) were used as stimuli. All the stimuli in all sessions were presented at 10° eccentricity (i.e., centered on this point) from the fixation spot. In each of the four experimental sessions, a single block of 180 trials (90 trials × 2 possible choices) was administered. Stimuli were displayed in the blind visual field in random order, whereas no stimuli were presented to the intact visual field. In the visual detection task, patients were asked to indicate the presence of a white dot in the blind visual field (50% catch trials). The remaining three tasks required them to guess the image presented to the blind visual field, choosing between two possibilities: fearful versus happy faces in the emotional task, male versus female faces in the gender task, and circle versus square in the shape task. Each trial (2250 msec total duration) began with a central fixation crosshair (2°) on a black background (500 msec), followed by the display of the target stimulus in the blind visual field (1500 msec) and then finally by a black screen with the fixation. At the end of the stimulus presentation, a sound prompted patients to verbally respond. During the task patients were instructed to maintain fixation on the central crosshair and a new trial was manually launched only when they were keeping their gaze on the fixation spot. Trials where eye movements occurred were discarded from the analysis (<2%). The percentage of correct responses given by each patient was compared with the chance level (50% correct responses) by means of a binomial test.

Figure 3. 

Graphical representation of the trial structure of the test of residual visual abilities. (A) visual detection task; (B) emotional task; (C) gender task; (D) shape task.

Figure 3. 

Graphical representation of the trial structure of the test of residual visual abilities. (A) visual detection task; (B) emotional task; (C) gender task; (D) shape task.

EEG Experiment: Behavioral Task

The EEG was recorded while the patients performed a double-choice discrimination task with redundant stimuli. The stimuli were 18 grayscale photographs of faces (7.5° × 11°; three men, three women) showing fearful, happy, or neutral expressions (Ekman & Friesen, 1976). All stimuli were displayed against a black background, 10° distant from the central fixation (i.e., centered on this point). Patients were required to discriminate between faces displaying fearful or happy expressions (target emotions), presented in their intact visual field by pressing one of two vertically arranged buttons on a keyboard.

The target seen stimuli were always coupled, in the blind field, with a face of the same identity showing a neutral, fearful, or happy expression (Figure 4). This resulted in six possible conditions: (1) target fearful face–neutral face in the blind field (Fn); (2) target fearful face–fearful face in the blind field (Ff); (3) target fearful face–happy face in the blind field (Fh); (4) target happy face–neutral face in the blind field (Hn); (5) target happy face–happy face in the blind field (Hh); (6) target happy face–fearful face in the blind field (Hf). Patients were instructed to respond as quickly and accurately as possible to target fearful and happy faces, respectively, by pressing the upper button with the middle finger or the lower one with the index finger of their right hand. A central fixation crosshair (2°) always remained on the screen. Each trial started with a fixation period (1500 msec) followed by the display of two faces, one to left and one to the right of the fixation spot (400 msec). After stimulus offset, responses were recorded during a maximal time interval of 1600 msec. As soon as a response occurred, a new trial started. Each experimental block consisted of 36 trials (six trials per condition). A total of 36 experimental blocks were run (1296 total trials, 216 trials per condition), split in two separate sessions of approximately 1 hr each.

Figure 4. 

Graphical representation of the trial structure in the behavioral task. The figure depicts examples of neutral (A), congruent (B), and incongruent (C) trials, when the target is a happy face.

Figure 4. 

Graphical representation of the trial structure in the behavioral task. The figure depicts examples of neutral (A), congruent (B), and incongruent (C) trials, when the target is a happy face.

Behavioral Data Analysis

RT was defined as the time interval between the onset of the stimulus and the subsequent button press. To control for outliers, trials were excluded on which the RT of the choice response was 3 standard deviations above or below the condition mean (<1.5%). RTs and error rates were subjected to three-way repeated-measures ANOVAs with the between-subject variable Group (left-sided lesion, right-sided lesion), and the within-subject variables Target Emotion (fearful, happy) and Condition (neutral, congruent, incongruent).

EEG Experiment: ERPs

The EEG was recorded with Ag/AgCl electrodes (Fast'n Easy-Electrodes, Easycap, Herrsching, Germany) from 27 electrode sites (Fp1, F3, F7, FC1, FC5, C3, T7, CP1, CP5, P3, P7, O1, Fz, Cz, Pz, Fp2, F4, F8, FC2, FC6, C4, T8, CP2, CP6, P4, P8, O2) as well as from the right mastoid. The left mastoid was used as reference electrode, and the ground electrode was placed on the right cheek. All electrodes were offline re-referenced to the average of both mastoids. Vertical and horizontal EOG was recorded from above and below the left eye and from the outer canthi of both eyes. EEG and EOG were recorded with a band-pass of 0.01–100 Hz and amplified by a BrainAmp DC amplifier (Brain Products, Gilching, Germany). The amplified signals were digitized at a sampling rate of 500 Hz and offline filtered with a 30-Hz elliptic infinite impulse response low-pass filter with a filter order of 11 and a transition bandwidth of 1 Hz. Further filtering was only applied for graphical display (16 Hz low-pass filter).

ERP Data Analysis

ERP data were analyzed using custom routines in MatLab 7.0.4 (The Mathworks, Natick, MA) as well as EEGLAB 5.03 (Delorme & Makeig, 2004), an open source toolbox for EEG data analysis (EEGLAB toolbox for single-trial EEG data analysis, Swartz Center for Computational Neurosciences, La Jolla, CA; www.sccn.ucsd.edu/eeglab). Only epochs with correct responses were considered in the ERP analyses (mean = 88.3% of all trials). Segments of 200 msec before and 500 msec after stimulus onset were extracted from the continuous EEG. The baseline window ran from −100 msec to 0 msec relative to stimulus onset. Epochs contaminated with large artifacts were identified using two methods from the EEGLAB toolbox (Delorme, Sejnowski, & Makeig, 2007): (1) An epoch was excluded whenever the voltage on an EOG channel exceeded an individually adjusted threshold (mean 126 μV) to remove epochs with large EOG peaks. (2) An epoch was excluded whenever the joint probability of a trial exceeded five standard deviations to remove epochs with improbable data. The mean percentage of epochs excluded in this way was 16.28%. Remaining horizontal and vertical EOG artifacts were corrected by an eye movement correction procedure (Automatic Artifact Removal Toolbox Version 1.3; www.germangh.com/eeglab_plugin_aar/index.html) based on a linear regression method (Gratton, Coles, & Donchin, 1983). Finally, epochs were discarded from the analysis when saccadic movements were registered in a time window of 400 msec following stimulus onset (11.6%) to make sure that faces displayed in the blind field could not be seen because of eye movements. The remaining epochs (mean: 156 epochs per condition) were averaged separately for each participant and each condition. For patients with lesions in the right hemisphere, ERP channels over one hemisphere were flipped to the other and vice versa. In this way, all patients' lesions resulted on the left side for all analyses and the corresponding Figure 6E and F. Instead, the scalp topography displayed in Figure 5D shows unflipped data for better understanding. The N170 amplitude was quantified using a peak-to-peak measure. We took the difference between the most positive peak in a time window of 50–150 msec following stimulus onset and the most negative peak in a time window of 150–250 msec following stimulus onset. Peak-to-peak amplitudes were subjected to a four-way, repeated-measures ANOVA with the between-subject variable Group (left-sided lesion, right-sided lesion) and the within-subject variables Electrode (P8, P4, O2), Target Emotion (fearful, happy), and Condition (neutral, congruent, incongruent). In case of a significant interaction between Group, Electrode, Target Emotion and Condition, three-way ANOVAs with Electrode (P4, P8, O2), Target Emotion (fearful, happy), and Condition (neutral, congruent, incongruent) as within-subject variables were computed separately for the two patient groups.

Figure 5. 

Grand-averaged ERPs elicited by happy and fearful target faces presented in the intact field as a function of Condition (neutral, congruent, incongruent). ERP waveforms at the representative electrode P8 (B, C, E, and F), and scalp topographies of the most negative peak in a time window of 150–250 msec following stimulus onset averaged over all conditions (A and D) are shown.

Figure 5. 

Grand-averaged ERPs elicited by happy and fearful target faces presented in the intact field as a function of Condition (neutral, congruent, incongruent). ERP waveforms at the representative electrode P8 (B, C, E, and F), and scalp topographies of the most negative peak in a time window of 150–250 msec following stimulus onset averaged over all conditions (A and D) are shown.

The peak-to-peak quantification of the N170 was chosen to control for possible effects of differences in the immediately preceding P1 component on the N170 (e.g., Goffaux, Gauthier, & Rossion, 2003; Rossion, Joyce, Cottrell, & Tarr, 2003), which is thought to reflect low-level properties of visual stimuli (e.g., Rossion & Caharel, 2011). Additionally, two control analyses were performed to ascertain that our N170 results were not influenced by the preceding P1 component. First, analyses of peak-to-peak N170 amplitudes were confirmed by analogous analyses of peak N170 amplitudes, quantified as the most negative peak in a time window of 150–250 msec following stimulus onset. Second, an analogous ANOVA was calculated also for the P1 amplitudes quantified as the most positive peak in a time window of 50–150 msec poststimulus onset. To compensate for violations of sphericity, Huynh–Feldt corrections were applied whenever appropriate (Huynh & Feldt, 1970), and corrected p values (but uncorrected degrees of freedom) are reported. Post hoc testing was done using Tukey's HSD test.

RESULTS

Test of Residual Visual Abilities

During screening, participants underwent four experiments (visual detection task, emotional task, gender task, shape task; see Methods) to determine if they could guess above chance regarding the presence and content of stimuli presented in their blind field. Results of the binomial test showed that the percentage of correct responses did not significantly differ from chance (50% correct responses) in any of the patients across all tasks (Table 2; visual detection task: all ps > .3; emotional task: all ps > .14; gender task: all ps > .14; shape task: all ps > .3).

Table 2. 

Percentage of Correct Answers in the Four Sessions of the Test of Residual Visual Abilities

CaseVisual Detection TaskEmotional TaskGender TaskShape Task
P1 52 44 47 47 
P2 52 56 47 50 
P3 48 55 53 48 
P4 53 54 53 53 
P5 49 47 48 51 
P6 51 50 47 47 
P7 50 45 50 52 
P8 53 50 56 48 
P9 48 46 46 51 
P10 53 47 50 47 
P11 52 51 46 51 
P12 51 48 47 49 
P13 47 51 53 52 
P14 54 55 50 53 
CaseVisual Detection TaskEmotional TaskGender TaskShape Task
P1 52 44 47 47 
P2 52 56 47 50 
P3 48 55 53 48 
P4 53 54 53 53 
P5 49 47 48 51 
P6 51 50 47 47 
P7 50 45 50 52 
P8 53 50 56 48 
P9 48 46 46 51 
P10 53 47 50 47 
P11 52 51 46 51 
P12 51 48 47 49 
P13 47 51 53 52 
P14 54 55 50 53 

Results are reported for each patient separately.

EEG Experiment

During EEG recording, patients were required to classify the emotional expression of target faces, which showed either a happy or a fearful expression. The target faces presented in their intact visual field could be coupled with a neutral, emotionally congruent, or emotionally incongruent face in the blind visual field (see Methods).

For the behavioral task, the three-way ANOVAs on RTs and error rates with the between-subject variable Group (left-sided lesion, right-sided lesion) and the within-subject variables Target Emotion (fearful, happy) and Condition (neutral, congruent, incongruent) showed no significant main effects (RT: all ps > .09, ηp2 < .22; error rates: all ps > .32, ηp2 < .09) or interactions (RT: all ps > .13, ηp2 < .16; error rates: all ps > .1, ηp2 < .17). The mean RT was 792 msec, and the mean error rate was 11.7%. Neither RT nor error rates were significantly different between the two patient groups (p > .34, ηp2 < .08, and p > .33, ηp2 < .08, respectively).

To determine an ROI for the N170 component, we first visually examined the grand-averaged scalp topographies for both patient groups, identifying the most negative peak in a time window of 140–200 msec following stimulus onset (see Figure 5A and D). It is evident from the topographies (Figure 5A and D) that the N170 was largest over the intact brain areas in both groups and the N170 was small over the lesioned hemisphere, probably because of the large lesions in occipito-parietal areas (see Figures 1 and 2). For this reason, only data from the intact hemisphere were analyzed. Note that except for the topographies shown in Figure 5A and D, hemispheres were flipped for patients with right-sided lesions for comparison with patients with left-sided lesions in all analyses (see Methods). Because the topography of the most negative peak in the time window of the N170 reached a maximum on channels P4, P8, and O2, these channels were chosen as an ROI for the N170 analyses. Grand-averaged waveforms for the representative electrode P8 are depicted in Figure 5B, C, E, and F, where a clear N170 component is visible in all experimental conditions. The ANOVA on peak-to-peak N170 amplitudes revealed a three-way interaction between Group, Target Emotion, and Condition, F(2, 24) = 6.02, p < .01, ηp2 = .33. No further main effects (all ps > .09, ηp2 < .21) or interactions (all ps > .26, ηp2 < .15) were significant. The ANOVA for the patients with left-sided lesions revealed a main effect of Condition, F(2, 12) = 4.83, p < .03, ηp2 = .45, qualified by a larger N170 amplitude for the incongruent pairs (Fh and Hf: 7.57 μV), followed by the neutral (Fn and Hn: 7.11 μV) and the congruent (Ff and Hh: 6.77 μV) pairs. Importantly, this main effect was qualified by a significant interaction between Target Emotion and Condition, F(2, 12) = 6.92, p < .01, ηp2 = .54. No further main effects (all ps > .32, ηp2 < .17) or interactions (all ps > .60, ηp2 < .12) reached significance. Post hoc testing showed that the N170 amplitude was larger when the target image in the intact field was a happy face coupled with a fearful face in the blind field (Hf: 7.95 μV), compared with the remaining conditions (Fn: 7.15 μV, p < .04; Ff: 6.97 μV, p = .01; Fh: 7.20 μV, p = .05; Hn: 7.07 μV, p < .02; Hh: 6.57 μV, p < .001; see Figure 6A for average data and Figure 6C for individual data), which were not statistically different from each other (all ps > .13).

Figure 6. . 

Top row: Mean peak-to-peak amplitude of the N170 elicited by happy and fearful target faces in the intact visual field as a function of Condition (neutral, congruent, incongruent), averaged over electrodes P4, P8, and O2. Data are displayed separately for patients with left (A) and right (B) hemisphere lesions. In patients with left hemisphere lesions (A), the N170 amplitude was significantly higher in the Hf condition (indicated by the asterisk), compared with all the remaining conditions (Fn, Ff, Fh, Hn and Hh). Central row: individual peak-to-peak amplitude of N170 in patients with left (C) and right (D) hemisphere lesions. Bottom row: scalp topographies of the difference in peak-to-peak N170 amplitude between congruent and incongruent trials in patients with left (E) and right (F) hemisphere lesions. n = neutral condition; C = congruent condition; IC = incongruent condition.

Figure 6. . 

Top row: Mean peak-to-peak amplitude of the N170 elicited by happy and fearful target faces in the intact visual field as a function of Condition (neutral, congruent, incongruent), averaged over electrodes P4, P8, and O2. Data are displayed separately for patients with left (A) and right (B) hemisphere lesions. In patients with left hemisphere lesions (A), the N170 amplitude was significantly higher in the Hf condition (indicated by the asterisk), compared with all the remaining conditions (Fn, Ff, Fh, Hn and Hh). Central row: individual peak-to-peak amplitude of N170 in patients with left (C) and right (D) hemisphere lesions. Bottom row: scalp topographies of the difference in peak-to-peak N170 amplitude between congruent and incongruent trials in patients with left (E) and right (F) hemisphere lesions. n = neutral condition; C = congruent condition; IC = incongruent condition.

In contrast, the ANOVA on peak-to-peak N170 amplitudes for the patients with right-sided lesions did not reveal any significant main effects (all ps > .14, ηp2 < .31) or interactions (all ps > .16, ηp2 < .25). Of particular interest, the interaction between Target Emotion and Condition was not significant, p > .25, ηp2 < .20 (see Figure 6B for average data and Figure 6D for individual data). Figure 6E and F shows grand-averaged scalp distributions of the difference in peak-to-peak N170 amplitude between the congruent and the incongruent conditions as a function of Target Emotion for patient with left and right hemisphere lesions, respectively. A clear negativity over right occipito-temporal brain areas is visible for happy targets in the group with left hemisphere lesions (Figure 6E), demonstrating that the scalp distribution of the observed effect in these patients is consistent with an N170 component generated in the right hemisphere.

The four-way ANOVA on peak N170 amplitudes revealed an interaction between Group, Target Emotion, and Condition, F(2, 24) = 4.61, p < .04, ηp2 < .28. The three-way ANOVAs performed separately for each group revealed an interaction of Target Emotion and Condition for patients with left-sided lesions, F(2, 12) = 4.64, p < .05, ηp2 < .44, but not for patients with right-sided lesions (p > .58, ηp2 < .18). Thus, analyses of peak N170 amplitudes replicated the analyses of peak-to-peak N170 amplitudes.

Finally, the ANOVA on peak P1 amplitudes revealed no significant effects (all ps > .24, ηp2 < .14). Crucially, the interaction between the variables Group, Target Emotion, and Condition was not significant (p > .81, ηp2 < .02), suggesting that effects on the peak-to-peak N170 amplitudes were not because of differences in the P1 peak preceding the N170.

In summary, patients with left-sided lesions showed larger N170 amplitudes only when a fearful face in the blind field was presented concurrently with a happy face in the intact field. By contrast, in patients with right-sided lesions, no effects of the stimuli presented in the blind field on the N170 were observed.

DISCUSSION

The aim of this study was to test whether the activity of a subcortical “low road” for fast visual processing of fearful stimuli can modulate cortical visual processing of consciously perceived stimuli. To this end, we investigated the effects of emotional faces presented in the blind field of hemianopic patients on the processing of faces presented concurrently in their intact field, as measured by the N170 component of ERPs elicited by these faces. The N170 is thought to represent the visual encoding of faces (Itier & Taylor, 2004; Bentin & Deouell, 2000; Bentin et al., 1996). Consequently, an enhancement of this component reflects an enhancement of facial encoding.

In hemianopic patients with left hemisphere lesions, we found that the N170 recorded from right occipito-temporal sites in response to a happy face in their intact visual field was selectively increased when a fearful face was concurrently presented in their blind visual field. No such enhancement was evident when seen (happy and fearful) faces were coupled with unseen happy or neutral faces. Likewise, unseen fearful faces did not elicit a higher N170 component when concurrently presented with seen fearful faces. These results are in agreement with previous studies on healthy participants reporting higher N170 amplitude in response to negative versus positive emotional faces (Meaux, Roux, & Batty, 2013; Morel, Ponz, Mercier, Vuilleumier, & George, 2009; Blau, Maurer, Tottenham, & McCandliss, 2007; Krombholz, Schaefer, & Boucsein, 2007; Leppanen, Kauppinen, Peltola, & Hietanen, 2007; Caharel, Courtay, Bernard, Lalonde, & Rebai, 2005; Batty & Taylor, 2003), even with subliminally presented stimuli (Smith, 2012; Pegna et al., 2008, 2011). However, the literature about ERP modulation by emotional faces is rather controversial, as other work suggested that the P1 component, but not the N170, is affected by the emotional valence of facial expressions (e.g., Rotshtein et al., 2010; Eimer & Holmes, 2002; Halgren, Raij, Marinkovic, Jousmaki, & Hari, 2000). Some authors argued that these discrepancies might arise from the heterogeneity of stimuli and paradigms used to investigate these aspects (Blau et al., 2007). Noteworthy, this study revealed N170 modulation by unseen fearful faces using a redundant target paradigm where one face was explicitly presented and the other one was not visible because displayed in the patients' blind visual field. In contrast, previous studies which found P1 modulations used only one explicitly presented face at a time (e.g., Rotshtein et al., 2010; Eimer & Holmes, 2002; Halgren et al., 2000). These differences in the paradigms might account for the discrepant results. In line with other recent studies using subliminal emotional faces as stimuli (Smith, 2012; Pegna et al., 2008, 2011), the results of this study support the idea that in the context of implicit emotion perception only the N170, but not the P1 component is sensitive to emotional (fearful) facial expressions.

The present EEG findings demonstrate that when the cortical “high road” for emotional processing is lesioned, the “low road” bypassing the striate cortex can extract threat-related visual information from the blind field and boost face perception in the intact field. Notably, the modulatory effect of unseen fear was evident at the stage of facial encoding, that is, at a fairly early stage of visual processing, which favors the hypothesis that the “low road” visual pathway provides a fast and coarse visual analysis in absence of awareness. Although the existence of a subcortical circuit mediating implicit visual processing of fearful stimuli has been debated (see Pessoa & Adolphs, 2010; Tamietto & de Gelder, 2010, for alternative accounts), animal studies have reported evidence of a network encompassing subcortical structures classically involved in emotional processing, such as the superior colliculus, the thalamus, and the amygdala (Day-Brown, Wei, Chomsung, Petry, & Bickford, 2010; Chomsung, Petry, & Bickford, 2008; Shi & Davis, 2001; Linke, De Lima, Schwegler, & Pape, 1999; Romanski, Giguere, Bates, & Goldman-Rakic, 1997; Jones & Burton, 1976). More recently, a diffusion tensor imaging study also demonstrated the existence of a subcortical “low road” to the amygdala in humans, reporting the presence of fiber connections between the SC and amygdala via the pulvinar, both in a blindsight patient (G.Y.) with unilateral destruction of the striate cortex and in healthy controls (Tamietto, Pullens, de Gelder, Weiskrantz, & Goebel, 2012).

Interestingly, in this study, a modulation of the N170 was evident only in the condition where a fearful face was presented in the blind field concurrently with a happy face in the intact field, whereas no effect of unseen fearful faces was observed when another fearful face was concurrently presented in the intact field. This lack of modulation suggests that conscious perception of a fearful face might have inhibited the activity of the subcortical “low road” to the amygdala engaged by the unseen fearful face to promote a slower and more detailed analysis of the seen fearful stimulus (Tamietto & de Gelder, 2010). Accordingly, previous studies using fMRI and connectivity analyses have shown that, whereas subliminal fear perception was characterized by positive connectivity within the subcortical pathways to the amygdala, conscious fear perception elicited negative functional connectivity within visual pathways to the amygdala (Williams et al., 2006). This latter finding supports our results, suggesting that the presence of seen fear in the Ff condition may have prevented the fear-related modulation of temporal-occipital areas by the subcortical pathways. Such a down-regulation of parallel visual inputs to the amygdala during conscious perception of threat might serve to deploy selective attentional resources toward the salient threatening stimuli, allowing deeper elaboration in higher-order cortical areas (Williams et al., 2006).

Alternatively, it could be that the ambiguous condition where a seen happy face is paired with an unseen fearful face (Hf condition) is the only condition capable of enhancing the N170. This would be in line with the suggested role of the amygdala in constantly monitoring the environment to boost vigilance as soon as something unpredicted or ambiguous occurs (Herry et al., 2007; Whalen, 1998, 2007), as well as with the observation of sustained amygdala activity and enhanced attention toward emotional faces in the presence of unexpected and unpredicted events (Herry et al., 2007). Notably, in our paradigm there were two emotionally ambiguous conditions, namely Hf and Fh. However, as only unconscious fearful stimuli are processed through the “low road” (Whalen et al., 1998, 2004), the Hf condition is the only one in which ambiguity can arise between information from the intact and blind visual field.

The electrophysiological data of this study are in line with previous behavioral findings on hemianopic patients showing that unseen fearful faces facilitated manual responses both to emotional and nonemotional stimuli presented in the intact field (Bertini et al., 2013). Notably, the same fear-specific behavioral facilitations were found in healthy participants when tested in a similar task after receiving inhibitory transcranial direct current stimulation over the left occipital cortex (Cecere, Bertini, & Làdavas, 2013). The present results extend these previous findings, suggesting that the behavioral facilitations observed in the presence of unseen fearful faces (Bertini et al., 2013; Cecere et al., 2013) are likely because of an enhancement of face encoding induced by unseen fear. However, it is worth noting that, in contrast to Bertini et al. (2013), we did not find a behavioral facilitatory effect of unseen fearful faces in this study, although this apparent discrepancy could be because of the different tasks used. In particular, whereas Bertini et al. (2013) employed a go/no-go task, this study used a double-choice task to avoid comparing go trials where a response was provided and no-go trials where no response was provided, while maintaining the experiment duration within reasonable limits. As suggested by previous psychophysical studies (Grice & Reed, 1992; Grice & Canham, 1990) response competition in double-choice tasks can mask the occurrence of facilitatory effects with redundant stimuli. Accordingly, studies with redundant emotional faces and employing a double-choice task (e.g., Schweinberger, Baird, Blumler, Kaufmann, & Mohr, 2003) failed to find any behavioral gain, in contrast to studies using similar redundant stimuli but in a go/no-go task (e.g., Tamietto, Latini Corazzini, de Gelder, & Geminiani, 2006).

The selective effect for unseen fear in hemianopes without blindsight, evident both at the behavioral (Bertini et al., 2013) and at the electrophysiological level (present study), is in contrast with previous evidence on blindsight patients (de Gelder et al., 2001, 2002). Indeed, de Gelder and colleagues (2001) reported a congruency effect (i.e., faster responses) when a blindsight patient was asked to classify the emotional expression of faces presented in his intact visual field, and the same emotion was presented in the blind visual field. Similarly, an electrophysiological study on blindsight patients showed that congruent and incongruent pairs of voices and unseen emotional faces differentially affected the auditory N1 component (de Gelder et al., 2002). Notably, hemianopes without blindsight and blindsight patients perform differently not only in tasks with redundant stimuli but also in classical 2AFC tasks. In these latter tasks, hemianopes without blindsight perform at chance level (Bertini et al., 2013), whereas blindsight patients demonstrate a wide variety of residual visual abilities, performing above chance during the discrimination of emotional faces (Tamietto et al., 2009; de Gelder et al., 1999, 2002) and other visual features (Morris, DeGelder, Weiskrantz, & Dolan, 2001; Brent, Kennard, & Ruddock, 1994). More importantly, the congruency effect shown by blindsight patients is strikingly similar to the effect reported in healthy participants (e.g., Tamietto & de Gelder, 2008), in whom, although conscious visual perception is prevented by backward masking, occipital areas are functional. This different pattern of results suggests that, although both hemianopes without blindsight and blindsight patients suffer from lesions to the primary visual cortex, their visual residual abilities might rely on different neural substrates. In particular, the performance of blindsight patients might stem not only from the activity of the “low road” but also from the contribution of some spared cortical extrastriate area, most likely because of postlesional functional reorganization of the visual system. Accordingly, neuroimaging evidence revealed a sustained hemodynamic response in the lateral occipital areas and within the posterior fusiform gyrus of blindsight patients during presentation of images of natural objects in their blind field (Goebel, Muckli, Zanella, Singer, & Stoerig, 2001). In contrast, the fear-specific effects demonstrated by hemianopes without blindsight both at the behavioral level (Bertini et al., 2013) and at the early stages of facial encoding (present results) are likely to exclusively reflect the activity of the subcortical pathway.

Another interesting finding of this study is that the modulation of the N170 for unseen fearful faces was evident only in patients with left hemisphere lesions, whereas no effect was observed in patients with right hemisphere lesions. Notably, in our patients both “low roads” (in the left and right hemispheres) were intact, as none of them had lesions of any of the critical subcortical structures encompassed by the “low road” (amygdala, pulvinar, superior colliculus; see Figure 2C, D, E). This suggests that unseen fearful facial expressions that are processed through the “low road” can only modulate the visual encoding of emotional faces mediated by the right hemisphere. This finding is in keeping with the well-documented right hemisphere advantage in processing emotions (Gainotti et al., 1993; Làdavas et al., 1993; Gainotti, 1972), as well as with the hypothesis of a dominance of the right hemisphere in withdrawal behavior, which is often associated to negative emotions, such as fear (Davidson, Jackson, & Kalin, 2000; Davidson, 1993). Furthermore, our results are consistent with other EEG studies showing that unseen fearful faces enhance the N170 component in the right hemisphere (Pegna et al., 2008). Although the techniques employed in this study do not allow to draw any conclusions about anatomical connections mediating the effects found, we speculate that this was likely achieved through reciprocal anatomical connections between the left and right hemispheres, possibly via the anterior commissure, as suggested by previous studies (Brown, Jeeves, Dietrich, & Burnison, 1999; Demeter, Rosene, & Van Hoesen, 1990; Klingler & Gloor, 1960).

In summary, we showed that unseen fearful faces presented in the blind field of hemianopic patients can enhance the structural encoding of seen happy faces presented in their intact field. Notably, the patients tested here could only process unseen fearful faces via the subcortical “low road” because of damage of the cortical “high road” (i.e., the geniculo-striate pathway). This suggests that the fear-specialized subcortical “low road” affords the implicit detection of fearful faces and that this system can operate independently of consciousness to ensure a fast processing of threatening stimuli. Such fast and automatic processing of stimuli signaling a potential threat is extremely valuable in adaptive terms, as it allows the individual to rapidly orient attentional resources toward danger and to produce an appropriate behavioral response.

Acknowledgments

This research was supported by grants from the Ministero Istruzione Università e Ricerca (PRIN) to E. L. We are grateful to Neil Dundon and Brianna Beck for their help in editing the manuscript and Manuela Sellitto for helping with the lesion reconstructions.

Reprint requests should be sent to Prof. Elisabetta Làdavas, Department of Psychology, University of Bologna, Viale Berti Pichat, 5 - 40127 Bologna, Italy, or via e-mail: elisabetta.ladavas@unibo.it.

REFERENCES

REFERENCES
Adolphs
,
R.
,
Tranel
,
D.
,
Damasio
,
H.
, &
Damasio
,
A. R.
(
1995
).
Fear and the human amygdala.
Journal of Neuroscience
,
15
,
5879
5891
.
Anderson
,
A. K.
, &
Phelps
,
E. A.
(
2000
).
Expression without recognition: Contributions of the human amygdala to emotional communication.
Psychological Science
,
11
,
106
111
.
Batty
,
M.
, &
Taylor
,
M. J.
(
2003
).
Early processing of the six basic facial emotional expressions.
Brain Research. Cognitive Brain Research
,
17
,
613
620
.
Bentin
,
S.
,
Allison
,
T.
,
Puce
,
A.
,
Perez
,
E.
, &
McCarthy
,
G.
(
1996
).
Electrophysiological studies of face perception in humans.
Journal of Cognitive Neuroscience
,
8
,
551
565
.
Bentin
,
S.
, &
Deouell
,
L. Y.
(
2000
).
Structural encoding and identification in face processing: ERP evidence for separate mechanisms.
Cognitive Neuropsychology
,
17
,
35
55
.
Bertini
,
C.
,
Cecere
,
R.
, &
Làdavas
,
E.
(
2013
).
I am blind, but I “see” fear.
Cortex
,
49
,
985
993
.
Blau
,
V. C.
,
Maurer
,
U.
,
Tottenham
,
N.
, &
McCandliss
,
B. D.
(
2007
).
The face-specific N170 component is modulated by emotional facial expression.
Behavioral and Brain Function
,
3
,
7
.
Brent
,
P. J.
,
Kennard
,
C.
, &
Ruddock
,
K. H.
(
1994
).
Residual colour vision in a human hemianope: Spectral responses and colour discrimination.
Proceedings: Biological Science
,
256
,
219
225
.
Broks
,
P.
,
Young
,
A. W.
,
Maratos
,
E. J.
,
Coffey
,
P. J.
,
Calder
,
A. J.
,
Isaac
,
C. L.
,
et al
(
1998
).
Face processing impairments after encephalitis: Amygdala damage and recognition of fear.
Neuropsychologia
,
36
,
59
70
.
Brown
,
W. S.
,
Jeeves
,
M. A.
,
Dietrich
,
R.
, &
Burnison
,
D. S.
(
1999
).
Bilateral field advantage and evoked potential interhemispheric transmission in commissurotomy and callosal agenesis.
Neuropsychologia
,
37
,
1165
1180
.
Caharel
,
S.
,
Courtay
,
N.
,
Bernard
,
C.
,
Lalonde
,
R.
, &
Rebai
,
M.
(
2005
).
Familiarity and emotional expression influence an early stage of face processing: An electrophysiological study.
Brain and Cognition
,
59
,
96
100
.
Calder
,
A. J.
,
Young
,
A. W.
,
Rowland
,
D.
,
Perrett
,
D. I.
,
Hodges
,
J. R.
, &
Etcoff
,
N. L.
(
1996
).
Facial emotion recognition after bilateral amygdala damage: Differentially severe impairment of fear.
Cognitive Neuropsychology
,
13
,
669
745
.
Cecere
,
R.
,
Bertini
,
C.
, &
Làdavas
,
E.
(
2013
).
Differential contribution of cortical and subcortical visual pathways to the implicit processing of emotional faces: A tDCS study.
Journal of Neuroscience
,
33
,
6469
6475
.
Chomsung
,
R. D.
,
Petry
,
H. M.
, &
Bickford
,
M. E.
(
2008
).
Ultrastructural examination of diffuse and specific tectopulvinar projections in the tree shrew.
Journal of Comparative Neurology
,
510
,
24
46
.
Davidson
,
R. J.
(
1993
).
Cerebral asymmetry and emotion: Conceptual and methodological conundrums.
Cognition and Emotion
,
7
,
115
138
.
Davidson
,
R. J.
,
Jackson
,
D. C.
, &
Kalin
,
N. H.
(
2000
).
Emotion, plasticity, context, and regulation: Perspectives from affective neuroscience.
Psychological Bulletin
,
126
,
890
909
.
Davis
,
M.
, &
Whalen
,
P. J.
(
2001
).
The amygdala: Vigilance and emotion.
Molecular Psychiatry
,
6
,
13
34
.
Day-Brown
,
J. D.
,
Wei
,
H.
,
Chomsung
,
R. D.
,
Petry
,
H. M.
, &
Bickford
,
M. E.
(
2010
).
Pulvinar projections to the striatum and amygdala in the tree shrew.
Frontiers in Neuroanatomy
,
4
,
143
.
de Gelder
,
B.
,
Pourtois
,
G.
,
van Raamsdonk
,
M.
,
Vroomen
,
J.
, &
Weiskrantz
,
L.
(
2001
).
Unseen stimuli modulate conscious visual experience: Evidence from inter-hemispheric summation.
NeuroReport
,
12
,
385
391
.
de Gelder
,
B.
,
Pourtois
,
G.
, &
Weiskrantz
,
L.
(
2002
).
Fear recognition in the voice is modulated by unconsciously recognized facial expressions but not by unconsciously recognized affective pictures.
Proceedings of the National Academy of Sciences, U.S.A.
,
99
,
4121
4126
.
de Gelder
,
B.
,
Vroomen
,
J.
,
Pourtois
,
G.
, &
Weiskrantz
,
L.
(
1999
).
Non-conscious recognition of affect in the absence of striate cortex.
NeuroReport
,
10
,
3759
3763
.
Delorme
,
A.
, &
Makeig
,
S.
(
2004
).
EEGLAB: An open source toolbox for analysis of single-trial EEG dynamics including independent component analysis.
Journal of Neuroscience Methods
,
134
,
9
21
.
Delorme
,
A.
,
Sejnowski
,
T.
, &
Makeig
,
S.
(
2007
).
Enhanced detection of artifacts in EEG data using higher-order statistics and independent component analysis.
Neuroimage
,
34
,
1443
1449
.
Demeter
,
S.
,
Rosene
,
D. L.
, &
Van Hoesen
,
G. W.
(
1990
).
Fields of origin and pathways of the interhemispheric commissures in the temporal lobe of macaques.
Journal of Comparative Neurology
,
302
,
29
53
.
Eimer
,
M.
, &
Holmes
,
A.
(
2002
).
An ERP study on the time course of emotional face processing.
NeuroReport
,
13
,
427
431
.
Ekman
,
P.
, &
Friesen
,
W.
(
1976
).
Pictures of facial affect.
Palo Alto, CA
:
Consulting Psychologists Press
.
Gainotti
,
G.
(
1972
).
Emotional behavior and hemispheric side of the lesion.
Cortex
,
8
,
41
55
.
Gainotti
,
G.
,
Caltagirone
,
C.
, &
Zoccolotti
,
P.
(
1993
).
Left/right and cortical/subcortical dichotomies in the neuropsychological study of human emotions.
Cognition and Emotion
,
7
,
71
93
.
Goebel
,
R.
,
Muckli
,
L.
,
Zanella
,
F. E.
,
Singer
,
W.
, &
Stoerig
,
P.
(
2001
).
Sustained extrastriate cortical activation without visual awareness revealed by fMRI studies of hemianopic patients.
Vision Research
,
41
,
1459
1474
.
Goffaux
,
V.
,
Gauthier
,
I.
, &
Rossion
,
B.
(
2003
).
Spatial scale contribution to early visual differences between face and object processing.
Brain Research. Cognitive Brain Research
,
16
,
416
424
.
Gratton
,
G.
,
Coles
,
M. G.
, &
Donchin
,
E.
(
1983
).
A new method for off-line removal of ocular artifact.
Electroencephalography and Clinical Neurophysiology
,
55
,
468
484
.
Grice
,
G. R.
, &
Canham
,
L.
(
1990
).
Redundancy phenomena are affected by response requirements.
Perception and Psychophysics
,
48
,
209
213
.
Grice
,
G. R.
, &
Reed
,
J. M.
(
1992
).
What makes targets redundant?
Perception and Psychophysics
,
51
,
437
442
.
Halgren
,
E.
,
Raij
,
T.
,
Marinkovic
,
K.
,
Jousmaki
,
V.
, &
Hari
,
R.
(
2000
).
Cognitive response profile of the human fusiform face area as determined by MEG.
Cerebral Cortex
,
10
,
69
81
.
Hariri
,
A. R.
,
Tessitore
,
A.
,
Mattay
,
V. S.
,
Fera
,
F.
, &
Weinberger
,
D. R.
(
2002
).
The amygdala response to emotional stimuli: A comparison of faces and scenes.
Neuroimage
,
17
,
317
323
.
Herry
,
C.
,
Bach
,
D. R.
,
Esposito
,
F.
,
Di Salle
,
F.
,
Perrig
,
W. J.
,
Scheffler
,
K.
,
et al
(
2007
).
Processing of temporal unpredictability in human and animal amygdala.
Journal of Neuroscience
,
27
,
5958
5966
.
Huynh
,
H.
, &
Feldt
,
L. S.
(
1970
).
Conditions under which mean square ratios in repeated measurement designs have exact F-distributions.
Journal of the American Statistical Association
,
65
,
1582
1589
.
Itier
,
R. J.
, &
Taylor
,
M. J.
(
2004
).
N170 or N1? Spatiotemporal differences between object and face processing using ERPs.
Cerebral Cortex
,
14
,
132
142
.
Jones
,
E. G.
, &
Burton
,
H.
(
1976
).
A projection from the medial pulvinar to the amygdala in primates.
Brain Research
,
104
,
142
147
.
Klingler
,
J.
, &
Gloor
,
P.
(
1960
).
The connections of the amygdala and of the anterior temporal cortex in the human brain.
Journal of Comparative Neurology
,
115
,
333
369
.
Krombholz
,
A.
,
Schaefer
,
F.
, &
Boucsein
,
W.
(
2007
).
Modification of N170 by different emotional expression of schematic faces.
Biological Psychology
,
76
,
156
162
.
Làdavas
,
E.
,
Cimatti
,
D.
,
Del Pesce
,
M.
, &
Tuozzi
,
G.
(
1993
).
Emotional evaluation with and without conscious stimulus identification: Evidence from a split-brain patients.
Cognition and Emotion
,
7
,
95
114
.
LeDoux
,
J. E.
(
1996
).
The emotional brain: The mysterious underpinnings of emotional life.
New York
:
Simon & Schuster
.
Leppanen
,
J. M.
,
Kauppinen
,
P.
,
Peltola
,
M. J.
, &
Hietanen
,
J. K.
(
2007
).
Differential electrocortical responses to increasing intensities of fearful and happy emotional expressions.
Brain Research
,
1166
,
103
109
.
Linke
,
R.
,
De Lima
,
A. D.
,
Schwegler
,
H.
, &
Pape
,
H. C.
(
1999
).
Direct synaptic connections of axons from superior colliculus with identified thalamo-amygdaloid projection neurons in the rat: Possible substrates of a subcortical visual pathway to the amygdala.
Journal of Comparative Neurology
,
403
,
158
170
.
Meaux
,
E.
,
Roux
,
S.
, &
Batty
,
M.
(
2013
).
Early visual ERPs are influenced by individual emotional skills.
Social Cognitive and Affective Neuroscience
.
doi:10.1093/scan/nst084
.
Morel
,
S.
,
Ponz
,
A.
,
Mercier
,
M.
,
Vuilleumier
,
P.
, &
George
,
N.
(
2009
).
EEG-MEG evidence for early differential repetition effects for fearful, happy and neutral faces.
Brain Research
,
1254
,
84
98
.
Morris
,
J. S.
,
DeGelder
,
B.
,
Weiskrantz
,
L.
, &
Dolan
,
R. J.
(
2001
).
Differential extrageniculostriate and amygdala responses to presentation of emotional faces in a cortically blind field.
Brain
,
124
,
1241
1252
.
Öhman
,
A.
(
2005
).
The role of the amygdala in human fear: Automatic detection of threat.
Psychoneuroendocrinology
,
30
,
953
958
.
Pegna
,
A. J.
,
Darque
,
A.
,
Berrut
,
C.
, &
Khateb
,
A.
(
2011
).
Early ERP modulation for task-irrelevant subliminal faces.
Frontiers in Psychology
,
2
,
88
.
Pegna
,
A. J.
,
Khateb
,
A.
,
Lazeyras
,
F.
, &
Seghier
,
M. L.
(
2005
).
Discriminating emotional faces without primary visual cortices involves the right amygdala.
Nature Neuroscience
,
8
,
24
25
.
Pegna
,
A. J.
,
Landis
,
T.
, &
Khateb
,
A.
(
2008
).
Electrophysiological evidence for early non-conscious processing of fearful facial expressions.
International Journal of Psychophysiology
,
70
,
127
136
.
Pessoa
,
L.
, &
Adolphs
,
R.
(
2010
).
Emotion processing and the amygdala: From a “low road” to “many roads” of evaluating biological significance.
Nature Review Neuroscience
,
11
,
773
783
.
Romanski
,
L. M.
,
Giguere
,
M.
,
Bates
,
J. F.
, &
Goldman-Rakic
,
P. S.
(
1997
).
Topographic organization of medial pulvinar connections with the prefrontal cortex in the rhesus monkey.
Journal of Comparative Neurology
,
379
,
313
332
.
Rorden
,
C.
, &
Brett
,
M.
(
2000
).
Stereotaxic display of brain lesions.
Behavioural Neurology
,
12
,
191
200
.
Rorden
,
C.
,
Karnath
,
H. O.
, &
Bonilha
,
L.
(
2007
).
Improving lesion-symptom mapping.
Journal of Cognitive Neuroscience
,
19
,
1081
1088
.
Rossion
,
B.
, &
Caharel
,
S.
(
2011
).
ERP evidence for the speed of face categorization in the human brain: Disentangling the contribution of low-level visual cues from face perception.
Vision Research
,
51
,
1297
1311
.
Rossion
,
B.
,
Joyce
,
C. A.
,
Cottrell
,
G. W.
, &
Tarr
,
M. J.
(
2003
).
Early lateralization and orientation tuning for face, word, and object processing in the visual cortex.
Neuroimage
,
20
,
1609
1624
.
Rotshtein
,
P.
,
Richardson
,
M. P.
,
Winston
,
J. S.
,
Kiebel
,
S. J.
,
Vuilleumier
,
P.
,
Eimer
,
M.
,
et al
(
2010
).
Amygdala damage affects event-related potentials for fearful faces at specific time windows.
Human Brain Mapping
,
31
,
1089
1105
.
Schweinberger
,
S. R.
,
Baird
,
L. M.
,
Blumler
,
M.
,
Kaufmann
,
J. M.
, &
Mohr
,
B.
(
2003
).
Interhemispheric cooperation for face recognition but not for affective facial expressions.
Neuropsychologia
,
41
,
407
414
.
Shi
,
C.
, &
Davis
,
M.
(
2001
).
Visual pathways involved in fear conditioning measured with fear-potentiated startle: Behavioral and anatomic studies.
Journal of Neuroscience
,
21
,
9844
9855
.
Smith
,
M. L.
(
2012
).
Rapid processing of emotional expressions without conscious awareness.
Cerebral Cortex
,
22
,
1748
1760
.
Sprengelmeyer
,
R.
,
Young
,
A. W.
,
Schroeder
,
U.
,
Grossenbacher
,
P. G.
,
Federlein
,
J.
,
Buttner
,
T.
,
et al
(
1999
).
Knowing no fear.
Proceedings of the Royal Society B
,
266
,
2451
2456
.
Stekelenburg
,
J. J.
, &
de Gelder
,
B.
(
2004
).
The neural correlates of perceiving human bodies: An ERP study on the body-inversion effect.
NeuroReport
,
15
,
777
780
.
Tamietto
,
M.
,
Castelli
,
L.
,
Vighetti
,
S.
,
Perozzo
,
P.
,
Geminiani
,
G.
,
Weiskrantz
,
L.
,
et al
(
2009
).
Unseen facial and bodily expressions trigger fast emotional reactions.
Proceedings of the National Academy of Sciences, U.S.A.
,
106
,
17661
17666
.
Tamietto
,
M.
, &
de Gelder
,
B.
(
2008
).
Affective blindsight in the intact brain: Neural interhemispheric summation for unseen fearful expressions.
Neuropsychologia
,
46
,
820
828
.
Tamietto
,
M.
, &
de Gelder
,
B.
(
2010
).
Neural bases of the non-conscious perception of emotional signals.
Nature Review Neuroscience
,
11
,
697
709
.
Tamietto
,
M.
,
Latini Corazzini
,
L.
,
de Gelder
,
B.
, &
Geminiani
,
G.
(
2006
).
Functional asymmetry and interhemispheric cooperation in the perception of emotions from facial expressions.
Experimental Brain Research
,
171
,
389
404
.
Tamietto
,
M.
,
Pullens
,
P.
,
de Gelder
,
B.
,
Weiskrantz
,
L.
, &
Goebel
,
R.
(
2012
).
Subcortical connections to human amygdala and changes following destruction of the visual cortex.
Current Biology
,
22
,
1449
1455
.
Whalen
,
P. J.
(
1998
).
Fear, vigilance, and ambiguity: Initial neuroimaging studies of the human amygdala.
Current Directions in Psychological Science
,
7
,
177
188
.
Whalen
,
P. J.
(
2007
).
The uncertainty of it all.
Trends in Cognitive Science
,
11
,
499
500
.
Whalen
,
P. J.
,
Kagan
,
J.
,
Cook
,
R. G.
,
Davis
,
F. C.
,
Kim
,
H.
,
Polis
,
S.
,
et al
(
2004
).
Human amygdala responsivity to masked fearful eye whites.
Science
,
306
,
2061
.
Whalen
,
P. J.
,
Rauch
,
S. L.
,
Etcoff
,
N. L.
,
McInerney
,
S. C.
,
Lee
,
M. B.
, &
Jenike
,
M. A.
(
1998
).
Masked presentations of emotional facial expressions modulate amygdala activity without explicit knowledge.
Journal of Neuroscience
,
18
,
411
418
.
Williams
,
L. M.
,
Das
,
P.
,
Liddell
,
B. J.
,
Kemp
,
A. H.
,
Rennie
,
C. J.
, &
Gordon
,
E.
(
2006
).
Mode of functional connectivity in amygdala pathways dissociates level of awareness for signals of fear.
Journal of Neuroscience
,
26
,
9264
9271
.
Zald
,
D. H.
(
2003
).
The human amygdala and the emotional evaluation of sensory stimuli.
Brain Research. Brain Research Reviews
,
41
,
88
123
.