Access the full text.
Sign up today, get DeepDyve free for 14 days.
Loading next page...
/lp/springer-journals/facial-and-semantic-emotional-interference-a-pilot-study-on-the-7LXlen0Wn3
- Publisher
- Springer Journals
- Copyright
- Copyright © 2011 by Ibáñez et al; licensee BioMed Central Ltd.
- Subject
- Biomedicine; Neurosciences; Neurology; Behavioral Therapy; Psychiatry
- eISSN
- 1744-9081
- DOI
- 10.1186/1744-9081-7-8
- pmid
- 21489277
- Publisher site
- See Article on Publisher Site
Abstract
Background: Integration of compatible or incompatible emotional valence and semantic information is an essential aspect of complex social interactions. A modified version of the Implicit Association Test (IAT) called Dual Valence Association Task (DVAT) was designed in order to measure conflict resolution processing from compatibility/incompatibly of semantic and facial valence. The DVAT involves two emotional valence evaluative tasks which elicits two forms of emotional compatible/incompatible associations (facial and semantic). Methods: Behavioural measures and Event Related Potentials were recorded while participants performed the DVAT. Results: Behavioural data showed a robust effect that distinguished compatible/incompatible tasks. The effects of valence and contextual association (between facial and semantic stimuli) showed early discrimination in N170 of faces. The LPP component was modulated by the compatibility of the DVAT. Conclusions: Results suggest that DVAT is a robust paradigm for studying the emotional interference effect in the processing of simultaneous information from semantic and facial stimuli. Keywords: ERP N170, LPP, IAT, DVAT, interference effects, valence, word, face Background conflict. In cognitive sciences, several paradigms are Integrating information about emotional valence from considered robust indexes of the degree of conflict, such face expressions and semantic information is an essen- as Simon effect, or interference between routes of diver- tial aspect of social interactions. In particular, the inte- gent/convergent emotional information, such as Emo- gration of emotional cues in a highly associative context tional Stroop effect. Conflict tasks, also known as (e.g., face to face communication) is critical for under- interference tasks, presentto the subjecttwo or more standing complex social cues. For example, to under- tasks to be performed simultaneously. Each task requires stand an irony, one benefits from integrating semantic the implementation of a limited number of maneuvers, information with facial clues that orient the listener to which produces interference or conflict when one task the opposite meaning. Language modulates the informa- is incongruent with another one. Here we present behavioural and neural correlates of tion presented in facial expressions [1], and in turn, emotion modulates semantic understanding [2]. In cer- an interference task, triggered by incongruent emotional tain situations, the incompatibility of emotional cues discrimination, in a similarveinthanthe emotional regarding semantic information in an associative context Stroop task. Nevertheless, in the DVAT the interference requires cognitive processes in order to solve this depends on the multimodal integration of (a) configura- tive and emotional aspects of face processing on one hand, and (b) semantic effects on the other. * Correspondence: aibanez@neurologiacognitiva.org Institute of Cognitive Neurology (INECO) and Institute of Neuroscience, Favaloro University, Argentina Full list of author information is available at the end of the article © 2011 Ibáñez et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 2 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 Implicit Association Task (IAT) the recording of ongoing electrophysiological activity The IAT [3] is an experimental method that measures the using electroencephalography (EEG). ERPs result from strength of associations between two categories, and it has the synchronous activation of neural subpopulations that been extensively used in social psychology [4,5]. The IAT occur in response to events (sensory, motor or cognitive). is a simultaneous stimulus categorization task that works ERPs are the sum of the activity of excitatory postsynap- by comparing subjects’ reaction times when classifying a tic potential and inhibitory postsynaptic potential acti- word or image shown on the computer screen into one of vated in response to each new stimulus. ERPs have a very two categories of response. The IAT is implicit in the low spatial resolution compared to neuroimaging techni- sense that subjects are not directly asked about the asso- ques. Nevertheless, they are useful not only for their ciations, but their reaction times to these associations are excellent temporal resolution but because recent measured. The IAT has been proven useful in assessing advances (e.g., dense arrays) provides multiple sources of social bias by considering valence categories (i.e., varying brain activity in response to cognitive events. from positive to negative) as well as social categories (e.g., The current research of ERPs in different models of Black and White people) and measuring whether experi- attention and emotion processes has highlighted the role mental blocks that require “compatible” associations (i.e., of early and late cortical dynamics. Early responses (eg., in prejudice: Black people/negative and White people/ 100-200 m ms after stimulus onset) index bottom-up sen- positive) show higher or lower reaction times relative to sory mechanisms sensitive to stimulus salience. In this “incompatible” associations (i.e., Black people/positive and regard, early modulation refers to the facilitation of early White people/negative). automatic and pre-attentional discrimination of salient stimuli. Later stages (300-800 ms) may be considered a Dual Valence Association Task (DVAT): An IAT measure of marker of top-down control mechanisms that support the emotional interference processing of task-relevant stimuli. The late process can In this investigation, we designed a modified version of be understood as a correlate of arousal and awareness trig- the IAT called the Dual Valence Association Task gered by the emotional content of stimuli. Moreover, it (DVAT). The DVAT measures interference effects pre- has been related to the integration of emotional stimuli sented in a double associative categorization valence features into the current context via controlled processing. tasks (that is, congruency/incongruence between emo- Thus, the early/late processes can be understood as an tional information presented in faces expressions and early-automatic and late-controlled parallel process trig- semantic information presented in words). In the gered by the stimuli salience and relevance. DVAT,wedonot useanevaluativetask(e.g.,positive Recent ERP paradigms using stimuli such as words vs. negative) and a categorization task (e.g., Black people and faces have elicited both early and late emotional vs. White people), but instead, we use two emotional effects. ERP research has demonstrated that early cere- valence evaluative tasks. Participants are asked to cate- bral modulation (130-170 ms) elicited by images of faces gorize the emotional valence of positive/negative words and other objects occurs in the occipito-temporal scalp or the emotional valence of positive/negative faces. Both area (N170; [6,7]). Other stimuli, unlike faces and attributes, the semantic dimension (pleasant or unplea- words, do not typically trigger the N170 effect [8-10]. sant words) and the facial dimension (happy or angry N170 is localized to the right hemisphere for faces, and facial expressions), must be categorized in blocks for to the left hemisphere for words ( [11,12], see also [13]). one dimension of compatible/incompatible valence. In Other studies have pointed out that N170 is modu- this way, the incompatible blocks imply either an asso- lated by faces and emotions [14-17]; but the specific ciative contextual effect of semantic interference with direction of emotion effects is not consistent between facial evaluation or an associative contextual effect of those studies, depending on the different emotions facial interference with semantic evaluation. included, tasks, and conditions. Nevertheless, when posi- The goal of this paper consists of providing a beha- tive vs. negative valence is considered, an enhancement vioral and ERP task for facial and semantic valence of positive valence amplitude has been reported with effects. The interaction between semantic and facial cues emotional pictures in early ERPs [18] and with emo- is an emerging research agenda (see review: Barret et al. tional faces paradigms in N170 [19]. [1]). Our paradigm provides a method borrowed from Contextual information (i.e., emotional background attitudes research (implicit association task) applied to congruent or incongruent with emotional expression of basic facial-semantic valence interference effects. faces) can also modulate N170 amplitude [20-22]. In summary, N170 is a neural marker of early face-selective Early and Late ERP processing processing which is not influenced by other objects pro- The technique of ERPs is a precise tool regarding time cessing. N170 is modulated by emotional context, affec- resolution (in the order of milliseconds) that incorporates tive valence and contextual cues. Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 3 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 Another early ERP component (which is occasionally participated voluntarily and signed an informed consent present along with N170), is characterized by a negative in agreement with the Helsinki declaration. All experi- voltage pattern at temporo-occipital areas of the scalp mental procedures were approved by the University (at 200-300 ms). This component is lateralized in the Ethics Committee. A questionnaire was given to all par- left hemisphere and specifically triggered by word sti- ticipants to rule out hearing, visual, psychiatric or neu- muli. This negativity is sensitive to the orthographic nat- rological deficits. ure of the stimuli, their lexical/phonologic properties [13,23], but also to their semantic effects [24]. Since its Stimuli validation semantic properties, this component would be modu- Pictorial stimuli for happy and angry faces were taken lated by emotional words and by semantic interference from a previous dataset used in previous studies effects in the DVAT. [31,36,37]. Happy and angry facial expressions, opposite in Late positive potential (LPP) have been related to eva- terms of valence dimension, were selected [38]. A set of 20 luative categorization and stimulus valence [25,26]. Sev- happyand 20 angrypicturescontrolledfor intensity, eral studies have suggested that factors related to brightness, colour and contrast was chosen. The happy perceptual processing of emotional stimuli are likely to and angry sets of pictures depicted the same people. We influence early ERP components, but not LPP in terms consider happy and angry facial emotion since we have of valence [27]. In line with these findings, several stu- previously reported consistent effects of N170 amplitude dies have demonstrated that LPP is, in fact, more closely modulation indexed by those both emotions [36,37] associated to arousal and higher cognitive processes Opposed valence words controlled for arousal, con- relative to the specific valence of an emotional sign [28]. tent, length and frequencies were also selected from a Similarly, previous ERP studies of comparable para- dataset reported in other study [31]. To validate word digms (e.g., lexical decision tasks with emotional stimuli, content, a questionnaire was used to gauge the pleasant- IAT tasks) have indicated an early modulation based on ness or unpleasantness of a list of 150 words with a valence and contextual cues (N170) and a late modula- moderate frequency use as selected using the Lifcach tion responding to incongruent blocks [18,29-32]. To frequency software. A sample of 50 psychology students sum up, early and late effects that have been reported participated in the validation. All participants were at using emotional stimuli which are characterized by an secondary school education or higher (average time at early valence effect (140-300 ms) and a later LPP modu- school 17 years, SD = 3.75), and 42 were right-handed- lation (300-700 ms) based on arousal and higher order ness. The average age was 19.62 (SD = 3.33), and 67.3% cognitive process, such as evaluations between cate- were female, with no visual deficits. Participants rated gories [33-35]. the set of words using a Likert scale where 1 repre- In this study, our aim was to present the DVAT as a sented a very positive valence and 7 represented a very new task designed to assess emotional interference effect negative valence. A repeated measure Analysis of Var- using facial and semantic stimuli. We measured the iance (ANOVA) was used to contrast categorizations for behavioural and electrophysiological correlates of the the list of pleasant and unpleasant words. Important dif- DVAT. We expected to find a behavioural effect in the ferences were obtained for the categorization of both DVAT (i.e., longer RTs for incompatible blocks relative lists [F(1, 73) = 25.16, p < 0.0001]. From the list of plea- to compatible blocks). Following previous result sant words, only those that were ranked between 1 and described above, we expected to find specific early and 3 were chosen (72 positive words were chosen, 3 late ERP effects, that is, (a) modulation of N170 for rejected). From the list of unpleasant words only those faces by valence and possible contextual association ratedbetween5and 7werechosen(71negative words effects, but not compatibility; b) modulation of the were chosen, 4 rejected). N240-280 component for word valence in the left pos- For faces, arousal mean value (range: 0 to 10) was 6.01 terior area of the scalp upon presentations of words; (SD = 1.23) and valence (range:-5 to 5) 0.21 (SD= 3.83). and (c) modulation of a late frontal LPP effect by com- For words, arousal mean values (range: 0 to 10) was patibility, but not by the specific valence of the target. 4.94(SD = 1.83) and valence (range: -5 to 5) 0.42 (SD= 4.28) respectively. For training blocks (see below) other Methods facial and semantic stimuli was considered. Participants The participants were 20 healthy volunteers, aged DVAT procedure between 19 and 26 years [M = 23.06, SD = 0.65], com- The task consisted of classifying faces and words as prising of 11 males and 9 females. The sample was either positive or negative in valence. Behavioral obtained from undergraduate student volunteers from responses were generated by pressing one of two keys the Cognitive Neuroscience Laboratory. All subjects with each forefinger (1 vs. 2). Therefore, each valence Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 4 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 category was assigned to a different side (left vs. right). analogous to the one made for the IAT [3-5,40,41]. The stimuli set had 20 pictures of happy faces, 20 pic- Since stimuli are the same in different test blocks, cate- tures of angry faces, 71 pleasant words and 71 unplea- gorization is subject only to contextual manipulation sant words. A greater number of word stimuli relative through category assignment to sides of the screen. to faces were selected to reduce the repetition effect of Training blocks included facial and semantic stimuli dif- words [39,32], (see discussion, ‘ERP results’ section). ferent than those presented during tests blocks. Initially, participants were informed that the study The task was organized in four different test blocks. In assessed emotions and classification processes regarding each block, trials were presented one by one with strict faces and words. They were told that they would see alternation between words and faces. Four labels were always visible at the top corners of the screen, two on words and faces and their task would consist of classify- each side, to remind the participants which answer cor- ing them as having either a positive or negative valence. responded to which response category. Each corner had Instructions indicated that participant should respond as a face category label (either “happy” or “angry”)anda quickly as possible. Participants sat in front of a compu- word category label (either “pleasant” or “unpleasant”). ter with electrodes placed on their heads and completed Test blocks only differed with respect to which side of the task as the stimuli were displayed on the computer the screen each category was assigned. Each test block screen. On-screen instructions preceded each block to was preceded by three practice blocks: one each for indicate what kind of stimuli would be presented and training face and word categorization and one consisting which button to press for each response category. The of a task identical to the corresponding test block but ERPs were stimuli locked to face and word onset (see shorter in length. Only one pair of labels was displayed the section on EEG recordings). Once the experiment at the top corners when only one type of stimulus was had finished, participants were thanked, and the presented in a training block. research goal was thoroughly explained. In block 1 (faces training), participants categorized Practice blocks involved 20 stimuli, consisting of faces either as happy or angry. In block 2 (words train- 10 face and 10 word stimuli, and test blocks comprised ing), they were asked to categorize words either as plea- 100 stimuli, that is, 50 face stimuli and 50 word stimuli. sant or unpleasant. For block 3 (faces/words training), Based on previous reports [31,32], and to ensure con- evaluative categories of blocks 1 and 2 were pooled scious access to faces and words, the time window during together in a single task of combined classification, such which stimuli were presented was set at 100 ms (faces) and 300 ms (words). Research on behavioral and ERP that stimuli were categorized as Angry-Unpleasant (left) assessment of word processing usually presents stimuli or Happy-Pleasant (right). The discrimination task that was carried out in practice block 3 was repeated but for 300 ms or more (e.g., [42-45]). When words are pre- with more stimuli in block 4 (test block, “Compatible”). sented for less than 300 ms, early effects are difficult to In block 5 (words training), words were categorized in obtain and an increase in error rates is expected. A recent thesamemanner, butthistimethe word categories report found an early (N170) effect when stimuli dura- were assigned to the opposite side of the screen (face tion was as long as response time [46]. On the contrary, expression labels were next to incompatible semantic facial stimuli requires less time for conscious detection labels). In block 6 (faces training), faces were categorized and processing (e.g., around 100 ms: [16,47,48]. without switching the side of the category presentation. Stimuli (faces and words) were repeated in order to The combined classification tasks followed in blocks 7 obtain an enough number of trials for ERP estimation. (faces/words training) and 8 (test block, “Incompatible”), Stimuli (faces and words) were repeated in order to with categories labelled on the same sides as in the pre- obtain a sufficient number of trials for ERP estimation ceding blocks 5 and 6: Angry-Pleasant (left) and Happy- [32,39]. The repetition effect of words is a robust modu- Unpleasant (right). To counterbalance the assignment of lator of ERPs [49-58]. On the contrary, facial ERP mod- hands to categories, the series of 8 blocks was repeated, ulation can be found even in a reduced number of faces switching sides for all labels, thus producing a total of (e.g., [53-58]. Moreover, the habituation effects (e.g., 16 blocks. attenuation of N170 amplitude and non-lateralization Blocks 4 and 12 (200 trials in total) were considered effects) are mostly frequent when the stimuli consist compatible since both negative categories were on one only of faces. When facial stimuli is presented together side (left) and both positive categories were on the other with other non facial stimulus (e.g., objects, words) the habituation and repetition effects disappear [58]. Given side (right). Hence, blocks 8 and 16 (also 200 total the mixed and counterbalanced presentation of faces trials), in which each side of the screen combined posi- and words in our paradigm, we chose a small number tive and negative classifications, were considered incom- patible and were expected to be more difficult tasks. of faces (40) because we do not expect habituation This compatible versus incompatible distinction is effects. Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 5 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 Figure 1 illustrates an example of a compatible test 0.5 and 30 Hz was applied to remove unwanted fre- block. Each trial began with a fixation trial (1000 ms), quency components. During recording the reference was followed by a blank screen (400 ms) and then a word set by default to vertex but then was re-referenced off- (300 ms) or a face (100 ms) in strict alternation. Follow- line to average electrodes. Two bipolar derivations were ing face or word presentation, participants had to cate- designed to monitor vertical and horizontal ocular gorize the valence of the stimulus displayed on the movements (EOG). ERPs were trigged from stimulus computer screen into one of two response categories onset (faces and words).Continuous EEG data were seg- (positive or negative). Incorrect responses were indicated mented from 200 ms prior to the stimulus to 800 ms with an ‘X’ (400 ms) in the center of the screen immedi- after the stimulus. All segments with eye movement ately after the response (however participants did not contamination were removed from further analysis using have to correct the error made after the ‘X’ appeared). an automatic (Gratton, Coles, and Donchin method for The negative feedback was introduced to ensure the removing eye-blink artifacts) and visual procedure. Arti- subject paid attention throughout the task. Then, an ISI fact-free segments were averaged to obtain ERPs. of 1000 ms was introduced. The stimuli were presented in the center of a black background of a 17-in color Data analysis monitor. Each stimulus was centered horizontally and DVAT score computing vertically on the screen subtending a visual angle of 4.5° A DVAT score was calculated for each subject based on × 3.15° at a viewing distance of approximately 80 cm. reaction times obtained from the test. This numeric value provides an index of difference between compati- EEG Recordings ble and incompatible tasks. Greenwald, Nosek, & Banaji EEG signals were sampled at 500 Hz from 129-channel [59] proposed a method to calculate this rate for social system with HydroCel Sensors from an Electrical Geo- categorization IATs, which involves eliminating extreme desic amplifier. Data outside the frequency band that reaction times, special management of wrong responses ranged from 0.1 Hz to 100 Hz were filtered out during and standardization of resulting reaction times based on the recording. Later, a band pass digital filter between the standard of each subject. In brief, this method Figure 1 Sequence representation for compatible test block. A. The schema shows a face and a word sequence presentation and the participant’s response. Face and word trials are alternately presented for a short time, strictly. Positive and negative faces, along with words of positive and negative valence, are present in the stimuli set and are presented in a randomized sequence. The participant is required to classify each stimulus to the left or to the right according to labels displayed on top of the screen. When a classification response error is made, negative feedback is given (i.e., in the word sequence illustrated in the figure). B. During the incompatible block test the valence classification of a face (e.g., happy-left vs. angry-right) must be classified in the opposite valence direction of the word (unpleasant-left vs. pleasant-right). When items from compatible categories (e.g. happy + pleasant) share a response key, performance is faster and more accurate than when items from incongruent categories (e.g. angry + pleasant) share a key Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 6 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 eliminates reaction times over 10000 ms; however, since factor. This factor involves two levels of word-face asso- our task was a rapid forced-choice task, we eliminated ciation: a compatible task (happiness/pleasant associa- responses with more than the 2000 ms from the analy- tion and anger/unpleasant association) and an sis. Finally the method recalculates wrong answers incompatible task (happiness/unpleasant association and (error trials) by adding 600 ms to their real value. Miss anger/pleasant association). trials (trials in which the subject does not respond), are For each stimulus type (face and word), Task was not included in the analysis. A subject’sscore is defined as follow: Compatible tasks for faces (faces asso- ciated with words of the same valence–[face positive/ obtained by measuring the reaction time difference word positive] and [face negative/word negative]) were between compatible and incompatible blocks, standar- dized according to the standard deviation (See additional compared to incompatible tasks (face associated with file 1 for a detailed explanation of the method).This words of the opposite valence–[face negative/word posi- method constitutes the best procedure for two-choice tive] and [face positive/word negative]). Following the compatible/incompatible tasks since it reduces the same procedure, compatible task for words (words asso- method-specific variance [60]. Hence, the measurement ciated with faces of the same valence–[word positive/ procedure was evaluated for bias using a scale that face positive] and [word negative/face negative]) were enables a comparison between different subjects. The compared to incompatible tasks(wordsassociatedwith result of this procedure was a number with an expected faces of the opposite valence–[word negative/face posi- value close to zero for subjects who did not show any tive] and [word positive/face negative]). bias in the test. Negative values corresponded to the To compare the specific effects of the target valence as detection of bias in favour of the compatible task. In well as the associated valence of the contextual affects, addition, as reported in the supplementary data (addi- ERP waveforms for early and late ERP effects were aver- tional file2),wedid thesameanalysisexcluding the aged for faces and words separately and analyzed using error trials and the penalties. Similar effects were a repeated measure ANOVA with the following within- obtained with this procedure. Finally, in the supplemen- subject factors: valence (positive vs. negative) and con- tary data (additional file 2) the accuracy as well as the textual association, which involves two possible context reaction times of each category was included. associations (positive and negative) for both faces and ERP analysis words. Faces (targets) were categorized in positive or Matlab software, EEGLab toolbox and T-BESP software negative valences of word contextual association (con- (http://www.neuro.udp.cl/) were used for off-line proces- text of the task). At the same time, words (targets) were sing and analysis of EEG data. Regions of interest (ROIs) categorized in positive or negative valences of facial con- textual association (context of the task). Interactions were used to analyze the scalp topography of the ERP components [61], which is recommended for dense between valence and contextual association produce arrays since it improves statistical power. ROIs were four categories (for both faces and words targets): posi- chosen by visual inspection of each component. Each tive target in a positive context; positive target in a N170 ROI (and for the left occipito-temporal negativity) negative context; negative target in a positive context; consisted of three adjacent electrodes (see additional file and negative target in a negative context. Finally, for 3 for those channel locations): the N170 ROIs were 65, each component, the factor Hemisphere (N170 left and 69, and 70 for the left and 83, 89, and 90 for the right. right locations) was considered. Frontal LPP was estimated based on previously pub- Partial eta squared (h ) effect size measures were lished electrode localizations [25]: Left (22, 23, 25, 26) computed for significant effects and relevant trends. and Right (2, 3, 8, 9). Although signal plots show the Univariate comparisons were done whenever necessary. overall averages of ERPs for each data cell, statistical Results were corrected with the Greenhouse-Geisser and tests were performed separately considering data for Bonferroni’s methods to adjust the univariate output of each participant using R software (http://www.r-project. repeated measure ANOVAs for violations of the com- org). For ERP analysis, mean average amplitudes were pound symmetry assumption. considered. The 140-190 ms time window for N170; 240-280 ms time window for the left occipito-temporal Results negativity; and the 400-700 ms time window for LPP Behavioural data: performance and DVAT scores were visually selected for mean amplitude analysis. Accuracy and reaction times The main factors submitted for analysis were task, No participants were eliminated based on Greenwald valence, and contextual association.ERP waveforms et al.’ 2003 criterion of 75% accuracy on each block were averaged for faces and words separately and ana- [59]. This percentage was obtained considering only cor- lyzed using a repeated measure ANOVA with Task rect responses within the 2000 ms (excluding errors and (compatible and incompatible) as the within-subject absent responses). Participants performed over >80% of Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 7 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 Table 1 Descriptive statistics of DVAT (accuracy and N170 and early effects reaction times) For the N170 elicited by faces, a main effect of Valence Category Accuracy (%) Reaction Times (ms) was observed [F(1, 19) = 12.23; p = 0.005; partial h = 0.39]. The N170 showed a larger amplitude in the posi- MSD M SD tive valence condition [M = -8.63, SD = 1.84] compared face- positive + context positive 93 3,6 890 221 to the negative valence condition [M = -7.51,SD = 1.72]. face- negative + context positive 85 6,7 970 243 In line with this observation, an interaction effect face- positive + context negative 91 5,6 1187 283 between Valence and Hemisphere was observed [F(1, face- negative + context negative 87 3,8 1140 212 19) = 5.80; p = 0.03; partial h = 0.23]. Post-hoc com- word- positive + context positive 94 4,2 1220 287 parisons [HSD test, MS = 2.42; df = 12,00] showed that word- negative + context positive 83 5,3 1662 287 positive valence faces [M = -9.86, SD = 1.85] elicited word- positive + context negative 81 5,6 1634 298 more amplitude than negative valence faces [M = word- negative + context negative 92 4,4 1303 245 -8,007, SD = 1.72] in the right hemisphere [p < 0.005]. Total 88,25 4,9 1250 258 In addition, an interaction between the Valence and Contextual Association was observed [F(1, 19) = 7.43, p accuracy in all categories. Regarding reaction times, no < 0.05; partial h = 0.28]. Post-hoc comparisons carried effects of valence or contextual association were found out for this interaction (see figure 2.b) suggested that (see table 1 for accuracy and RTs descriptive statistic). only the category with the largest modulation, that is Nevertheless, facial and semantic stimuli elicited a task happy faces associated with unpleasant words [M = effect: Compatible blocks produced shorter RTs than -8.93, SD = 1.94], was distinguished from angry faces incompatible ones. Accuracy and RTS effects are pre- associated with unpleasant words [M = -7.06, SD = 1.66, sented in the supplementary data (additional file 2). p = 0.02;HSD test, MS = 4.15, df = 12.00]. No effects of DVAT scores Task compatibility were found. Mean DVAT score of the subjects was -1.67 [SD = The N170 elicited by words was not significantly 1.03], a value significantly less than zero [t(20) = -7.39, affected by any of the relevant factors (see figure 2.a). p = 0.0003, Cohen’s d=-1.61]. DVAT score provides an However, in the left region of the scalp a second nega- index of difference between compatible and incompati- tive peak was observed (aprox. 270 ms). This compo- ble tasks (a value close to zero for subjects indicated no nent appeared to behave partially like the N170 effect in DVAT effects of compatibility). Only one subject the right hemisphere observed with facial stimuli. None- obtained a non-negative DVAT score (value= 0.07). theless, no main effects of Valence were found and only Thus, the reaction times of the subjects were longer on a trend between Valence and Contextual association average in the incompatible blocks relative to the com- interaction was observed [F(1, 19) = 4.12, p = 0.06; par- patible blocks. Our results show, as expected, that the tial h = 0.18]. Post-hoc comparisons [HSD test, MS = compatible valence blocks (happy/pleasant and anger/ 3.95, df = 12.00] carried out for this interaction sug- unpleasant association) were facilitated, compared to the gested that only the category with the largest modula- incompatible valence blocks (happy/unpleasant and tion, unpleasant words associated with happy faces, [M anger/pleasant association). = -2.98, SD = 0.56], was distinguished from unpleasant In our sample, DVAT scores had a Spearman-Brown words associated with angry faces [M = -1.23, SD = corrected split-half reliability of r = 0.82, yielding a 0.47, p = 0.02]. As with the N170 component, no effects SB standard error of s = 0.44. Split-half reliability was of Task compatibility were found in this time window. err computed considering each picture trial and the follow- ing word trial as an atomic unit. All units were used in LPP late effects an odd-even reliability estimation. In other words, each Regarding faces and words, no effects were found for half had words and pictures, the same structure, and Valence, Contextual association or Hemisphere. Never- was the same for every subject. This result suggests that theless, a Task compatibility effect was present for both the DVAT yields high reliability measurements, and pro- faces and words. vides an estimation of the standard error of the test In the Task comparison for faces, a main effect of making it useful for distinguishing actual differences Hemisphere was found [F(1, 19) = 20.73, p = 0.0006; from measurement error. We performed an additional partial h = 0.52], and only a trend for the effect of DVAT score analysis, this time excluding errors and Task was found [F(1, 19) = 3.51, p = 0.085; partial h = penalties (see supplementary data in additional file 2). 0.16]. Nevertheless, an interaction between Task and Those results were similar to the DVAT scores calcula- Hemisphere was observed [F(1,19) = 9.31, p < 0.01; par- tion presented here. tial h = 0.33]. Post-hoc comparisons for this interaction Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 8 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 Figure 2 N170 waveform and topography. N170 component elicited by emotional words and faces. The N170 modulation is predominant in the left hemisphere for words and in the right for faces. Boxes in ERPs figures are indicative of regions where statistical comparison shown significant differences. Right side of the figure: Word and Face voltage maps averaged on 140-190 ms. Abbreviations: Word P-A (pleasant words associated to anger face); Word U-A (unpleasant words associated to anger face); Face H-P (happy faces associated to pleasant word); Face H-U (happy faces associated to unpleasant word); Face A-U (anger faces associated to unpleasant word); Face A-P (anger faces associated to pleasant word). [HSD test, MS = 0.44 df = 12.000] found larger ampli- bimodal (faces and words) emotional conflict task. This tudes and differences between compatible [M = 1.15, SD paradigm evaluates the emotional conflict and facilita- = 0.44] and incompatible [M = 3.48, SD = 0.82] blocks tion of incompatible and compatible bimodal valence in the right hemisphere [p = 0.0002]. In the left hemi- associations, suggesting that this modified IAT is suita- sphere, a small but still significant difference was also ble for emotional interference research on normal and observed between compatible [M = -0.34, SD = 0.62] psychiatric populations. and incompatible [M = 0.98, SD = 0.76] blocks [p < 0.05]. In both hemispheres, the incompatible blocks eli- cited larger amplitudes relative to compatible blocks (see figure 3.a). For words, a main effect of Hemisphere was found [F(1, 19)= 9.99, p = 0.008; partial h = 0.35], but no main effect of Task was found. Nevertheless, an interaction between Task and Hemisphere, similar to our faces effect but slightly stronger, was observed [F(1, 19) = 8.81, p = 0.01; partial h = 0.32]. Post-hoc comparisons for this interac- tion [HSD test, MS = 1.34, df = 12.00] found only differ- ences between the compatible [M = 0.49, SD = 0.82] and incompatible [M = 2.82, SD = 0.65] blocks in the right hemisphere [p = 0.0002]. No differences were found in the left hemisphere (see figure 3.b). Synthesizing the ERP main results, the effects of valence and contextual association were observed at an early processing (N170 component) but not later. Com- patibility effects were observed only at later process as indexed by the LPP component. Discussion Figure 3 Frontal LPP. Selected ROIs of LPP elicited by faces and The aim of this study was to examine performance in a words in response to Incompatible (continuous line) and Compatible tasks (dashed line). Top: ERPs of Faces. Bottom: ERPs of DVAT with two forms of emotional compatible/incom- Words. In agreement with the behavioral results, we found a LPP patible associations (facial and semantic). This novel modulation related to incompatible blocks. Note the lateralization task, although similar to IAT, does not involve an eva- effects for word stimuli and the bilateral effects for faces. luative dimension and a categorization task but a Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 9 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 The Novel Relevance of DVAT: Implicit cognition and studies were comprised of incongruent trials in which a Conflict tasks word referring to a colour was presented (e. g., red), but Several recent studies investigating IAT have criticized the actual color in which the word was written was behavioural measurements based on classical bias attri- incongruent (e. g., blue), provoking longer reaction butes in incompatibility effects. There is an ongoing times. This reaction time effect accounts for the inter- debate about its significance as an attitude index, largely ference provoked by the conflict emerging from the due to theoretical assumptions about the cognitive and incongruence of the semantic meaning of the word and behavioural processes involvedinanIAT task the colour in which it is written [68]. The Stroop task [40,41,62-65]. IAT may not measure only social bias, has undergone several modifications since it was created but also several other cognitive processes [4], for exam- to examine and explain the interference phenomenon ple, interference effects based on stimuli. Although our and relevant variables [69,70]. Consistent with our results do not provide data to directly address this issue, results, this electrophysiological interference effect they support this hypothesis. Since DVAT uses two eva- appears to be modulated at later times. Recent studies luative judgments, the incompatibility effects might be suggest that the interference process occurs closer to explained by an interference effect that forces a complex the response stage than to the stimulus-processing stage strategy: for incompatible blocks, a dual strategy (i.e., [71]. The Simon effect [72] is another classic interfer- positive word valence in one direction and positive face ence paradigm (although with no emotional content) valence in another direction) is expected to demand that is observed in information processing during more effort and time in comparison to a simple strategy response selection. In an electrophysiological study of (ie., words and pictures valence in the same direction). Simon effect [73], although an early modulation (N1, This effect is relevant to the hypothesis that the IAT is N2) was observed, the reaction times associated with a dual task (a simple and a complex task with interfer- incongruence have been found to correlate with the ence), given that no social bias effect or similar effect is P300 component. In summary, the processing of inter- present in our task. Furthermore, our behavioural results ference stimulated by a pair of incongruent stimuli is a show that the pattern of the IAT effect is present in late process [68,74] that occurs during the response social bias paradigms. In race-IAT, this process affects selection stage, which explains the longer times that test IAT scores. For example, when participants showing subjects need to respond to incongruent stimuli. The implicit racial bias performed a race-IAT, in the compa- DVAT results are in agreement with previous reports of tible blocks (White-positive task), they simply responded conflict tasks, an additionally elicits electrophysiological to bothfacesand wordsonthe basisofwhether they early (valence modulation) and a late (conflict modula- liked the presented stimuli. Because task switching leads tion) stages of emotional integration and interference. to performance costs, performance is diminished in the This is the first report of an implicit association task Black-positive task compared to the White-positive task based on interference effects (triggered by facial and (See [4] for a detailed discussion on this topic). Thus, semantic valences) being more suitable for basic emo- task switching in the more complex task (incompatible) tion research than IAT. Compared to the emotional would be mainly triggered by this interference. Stroop effect, the interference is not triggered by colour In both IAT and DVAT, the “implicit” property relies and emotion, but by facial processing and semantic in the fact that the association between faces and word valence. Emotional inference of facial clues is one of the content is not explicit. Nevertheless, DVAT’s implicit- most important steps in the development of complex ness property has to be carefully considered. We have social behavior [75]. Faces are multi-dimensional stimuli adopted the De Houwer definition of implicit measure directly related to important social incentives [76]. Emo- as “outcome of a measurement procedure that results tional face expression gives an automatic and fast short- from automatic processes by which the to-be measured cut to alarm signals, metallization and inter-subjective attribute causally determines the outcome” [66]. The communication. Thus, the DVAT opens a new area of DVAT can only partially assess the normative criterion research on emotional interference effects better suited of this definition (the “What” criterion, the “How” cri- for social cognition research than the emotional stroop terion and the implicitness criterion; see De Houwer et task. DVAT considers stimuli (emotional faces) with al. 66] for more details on this definition). high relevance for social cognition. Our report has Our results suggest that the DVAT can be used as a shown cortical markers of early valence modulation and conflict or interference measure, with early and late late compatibility effect suggesting that emotional mar- neural correlates. A classic conflict or interference para- kers and evaluative comparison are processed in two digm that has been thoroughly studied is the Stroop temporal brain dynamics. Valence and emotional inter- effect, which resembles the classic IAT regarding com- ference would be automatically processed in the struc- patible/incompatible tasks [41,67]. The first Stroop tural stage (eg., fusiform gyrus and superior temporal Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 10 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 sulcus indexed by N170). Compatibility should be pro- stimuli. The emotional content of the target (either a cessed in a more controlled and delayed process, word or picture) is contextually affected by the compat- indexed by prefrontal cortices, as LPP topography sug- ibility or incompatibility of the subsidiary categorization gests. Early emotional decoding possibly subserves adap- (either word or picture) and is independent of its speci- tive social behavior; and later LPP processing indexes fic content (positive or negative). Therefore, regardless cognitive control and behavioral performance. Those of the specific valence of the target, the responses results are consistent with recent reports of context- (DVAT scores) are modulated by the interference/asso- dependant effects on cognitive processing [29-32,77-86]. ciation or facilitation of the subsidiary category valence. Considering that interference and conflict studies have had consistent applicability in psychiatry, the DVAT The relation of our ERP results and previous research might be a sensitive paradigm for pathologies such as Confirming our hypothesis, ERP results indicated the schizophrenia and ADHD, in which emotional and inhi- occurrence of early and late modulations. When pre- bition/interference deficits are simultaneously observed. sented with face stimuli, the N170 component was Since this paradigm demonstrates the early effects of mainly modulated by valence (positive > negative) and valence and the late effects of interference in electrophy- showed independence from compatibility effects. Only a siological measures, it may be used to investigate the trend of contextual association was observed in the right time course of both processes in these psychiatric popu- scalp, which occurred for positive valence faces (happi- lations. Moreover, early and late ERPs to emotional pro- ness) associated with a negative semantic valence cessing are modulated by psychiatric medication [87]. (unpleasant). This finding suggests an early effect on the Future studies would assess the possible role of psychia- salient categories of contextual incongruence. Neverthe- tric medication (antidepressants, mood stabilizers, sti- less, there was no effect of this component on mulants, etc) on emotional interference effects indexed incompatibility. by the DVAT. Word ERPs were not modulated as a function of any A well-known effect of blocked presentation in inter- of the variables previously described for the N170 win- ference tasks (e.g., Stroop effect) has been reported else- dow. However, a specific trend of word stimuli in the where (for a recent report, see [42]. In our design, the left hemisphere was later observed. In the 240-280 ms main reason for selecting a blocked presentation (here time window, a second negative peak was observed, as well as in multiple IAT tasks) is the complexity of which was modulated as a function of valence and con- the task: if participants have to constantly change the textual association (unpleasant words associated with dual categorization of faces and words, as well as their happy stimuli > unpleasant words associated with anger corresponding positive and negative valence, the para- stimuli). Once more, there was no compatibility effect. digm becomes difficult to perform. Using a blocked pre- Of interest, we only observed significant discrimina- sentation allows the learning of the task during the tion in the frontal LPP component window based on practice trials. Future studies, however, with a well- compatible and incompatible categories, independent of designed paradigm, should test the compatibility effects the specific valence, both for words (a restricted effect in mixed blocks. in the right scalp) and faces (both hemispheres). The LPP was affected neither by main effects of valence for Behavioural pattern face and word targets, nor by contextual association. The accuracy was relatively high in all categories of Importantly, when compared to reaction times, LPP DVAT (mean: 88%) and the reaction times were modu- effects presented a similar pattern (modulation by task) lated only by task compatibility. Our DVAT scores confirming the association of later ERP components to showed a robust effect that distinguished compatible/ behavioral reposes and arousal [33-35] incompatible tasks. Incompatible task blocks generated a In summary, the effects of valence and contextual longer latency in responses, supporting an interference association showed early discrimination in ERPs of pic- effect (of words on pictures and vice versa) produced by tures and a tendency for words (N170 and the second two opposed pairs in the same dimension of emotional peak around 240-280 ms, respectively). LPP component valence. Indeed, behavioural data showed a very strong corresponded with behavioural measures, accounting for DVAT effect. The mean DVAT score was negative with late modulation by the compatibility of the DVAT. large statistical significance, which was expected because This study is in concordance with several previous subjects were faster when their task consisted of an findings. First, N170 was larger for faces than for words association in the same valence dimension (positive: [8-10,88], and this effect was lateralized to the right happy/pleasant vs. negative: angry/unpleasant). This evi- hemisphere when pictures were used as stimuli, which has been previously observed especially when the experi- dence suggests that the DVAT is sensitive to interfer- mental paradigm includes faces and words or faces and ence in valence associations of facial and semantic Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 11 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 objects [11,12,89]. Only contextual effects of salient sti- The time course of the stimuli (the ISI and the fixation mulus associations were observed [20-22], but not of by one side, and the delay in the response by other) incompatibility [29]. Finally, early components has been avoids any feedback overlap with the preceding and fol- previously reported as modulated by emotional effects, lowing stimulus. showing increased amplitude from positive stimuli over neutral stimuli (EPN: [18]; N170 for faces:[19]). Early and Late effects of DVAT In other hand, an early posterior negativity (EPN) has Various studies with emotional stimuli have indicated been described in paradigms of affective picture proces- that modulation of early and late components (i.e., the sing as an index of emotional discrimination [18,90]. An direction of amplitude differences) is not consistent EPN-like has been recently reported to be elicited by amongst different types of paradigms and stimulus fac- faces and words [19]. We did not find such effect; how- tors. The task appears to affect differentially early and ever, different studies with emotional faces (for a review, late components. Indeed, several previous ERPs studies [91] or words: for a review see, [92]) have not reported of emotional stimuli showed different effects and altera- an EPN either. Probably, task dependent factors can tions attributed to aspects of the stimuli and experimen- explain the presence or absence of EPN. tal setting, including picture size, presentation time, Words showed no N170 modulation of valence content, gender, inter-picture interval, inclusion of distracters, in agreement with previously reported results [19]. How- stimulus relevance, spatial frequency, colour, complexity ever, we find a delayed (N230-280 ms) emotional modula- and feature composition of pictures [27,35,100-103]. tion tendency. Its modulation in our study partially The fact that different stimuli and tasks produce differ- resembles the N170 effects elicited by faces, but a delayed ent results suggests that ERP results are paradigm- response, maybe caused by the delayed word presentation. dependent [104]. More importantly, it is still uncertain This early semantic modulation, probably facilitated by how various attributes of stimuli interact and how they emotional content, has been recently reported in several affect the overall affective process [34,101]. Future stu- paradigms [93-95]. Further studies with a manipulation of dies should assess whether the direction of the early time window stimuli presentation should asses if this early effects of valence and LPP components are paradigm- modulation is part of a N170-like component. dependent aspects or specific properties associated with In later stages, consistent with reaction times, frontal the categorization of the dual-valence task. LPPs were exclusively modulated by the effect of incom- Despite paradigm-dependent differences in previous patible blocks (more amplitude) in comparison to com- reports and the novelty of our results, the overall pattern patible ones. This effect was more pronounced in the confirms emotional effects previously reported with right hemisphere when word stimuli where used. These other paradigms: an early modulation based on valence results are consistent with previous studies showing a and a late effect of arousal and higher order cognitive modulation of the LPP in a valence judgment task based processes [33-35]; these findings contribute to the clarifi- on compatibility[31] and arousal stimuli [25]. The fron- cation of these discrepancies. The configuration aspects tal effects are compatible with the role attributed to of stimuli seem to be discriminated early, but their effects anterior structures in interference. The anterior cingu- are combined during later phases. These results are parti- late cortex (ACC) has been associated with the detection cularly similar to reported effects, where early modula- of interference and its functional relationship with the tion has been observed for configuration categories (i.e., prefrontal cortex [74]. Studies using fMRI have indicated positive valence) and associated contextual effects that in Stroop tasks, interference activates frontal areas, together with subsequent late modulation based on com- such as the ACC and dorsolateral prefrontal cortex [69]. patibility. However, unlike those reports, in the DVAT Following classical assessment of the two-choice task the early effects are related to the discrimination of [96-99], negative feedback (an “X” appeared on the valence, and incompatibility is recognized later as evi- screen) was given immediately after incorrect categoriza- denced by non-lateralized frontal LPPs upon presentation tion of a stimulus. The negative feedback was intro- of faces. Future studies must examine whether early duced to increase the attention of the participants effects (valence and context) and late effects (task com- during the task. The ISI (1000 ms) and the fixation patibility/incompatibility) are caused by intrinsic emo- (1000 ms) following the negative feedback prevented tional biases of stimuli or only dependent on the any overlap effect of the feedback and the following face interference effect presented in a dual categorization task. andwordstimuli.Atthe same time,the response (and consequently the negative feedback) overlapped after the Limitations and future testing ERP window locked to the stimulus (M = 1250 ms, SD Extensive behavioural assessment is required in order to = 258). Thus, feedback does not affect the previous face confirm the validity of DVAT. For future neuropsychia- or word ERP window (0 to 700 ms, from onset to LPP). tric evaluation, a multivariate analysis including Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 12 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 3. Greenwald AG, McGhee DE, Schwartz JL: Measuring individual differences neuropsychological performance in combination with in implicit cognition: the implicit association test. J Pers Soc Psychol 1998, DVAT is desired. A larger sample would allow the study 74: 1464-1480. of individual differences related to the performance of 4. De Houwer J, Teige-Mocigemba S, Spruyt A, Moors A: Implicit measures: A normative analysis and review. Psychol Bull 2009, 135: 347-368. this paradigm. As a new emotional interference mea- 5. Greenwald AG, Poehlman TA, Uhlmann EL, Banaji MR: Understanding and sure, a comprehensive testing of their psychometric using the Implicit Association Test: III. Meta-analysis of predictive properties is needed (reliability, validity), as well as ERP validity. J Pers Soc Psychol 2009, 97: 17-41. 6. Bentin S, Allison T, Puce A, Perez E, McCarthy G: Electrophysiological source localization studies, and the examination of the studies of face perception in humans. J Cognitive Neuroscience 1996, 8: paradigm on affective disorders. 551-565. 7. Joyce C, Rossion B: The face-sensitive N170 and VPP components manifest the same brain processes: the effect of reference electrode Conclusions site. Clin Neurophysiol 2005, 116: 2613-2631. Our results suggest that the DVAT is a robust beha- 8. Rebai M, Poiroux S, Bernard C, Lalonde R: Event-related potentials for vioural and electrophysiological paradigm for studying category-specific information during passive viewing of faces and objects. Int J Neurosci 2001, 106: 209-226. the emotional interference effect in the processing of 9. Rossion B, Gauthier I, Tarr MJ, Despland P, Bruyer R, Linotte S, simultaneous information from semantic and facial sti- Crommelinck M: The N170 occipito-temporal component is delayed and muli. The effects of valence (more in faces than in enhanced to inverted faces but not to inverted objects: an electrophysiological account of face-specific processes in the human words) are discriminated early, and the effects of com- brain. Neuroreport 2000, 11: 69-74. patibility, regardless of the specific valence, are observed 10. Rossion B, Curran T, Gauthier I: A defense of the subordinate-level at a later time (consistent with reaction time effects). expertise account for the N170 component. Cognition 2002, 85: 189-196. 11. Brandeis D, Lehmann D, Michel CM, Mingrone W: Mapping event-related Future studies are required to determine the applicabil- brain potential microstates to sentence endings. Brain Topogr 1995, 8: ity of this paradigm as a conflict marker of socio-emo- 145-159. tional skills in normal and psychiatric populations. 12. Rossion B, Joyce CA, Cottrell GW, Tarr MJ: Early lateralization and orientation tuning for face, word, and object processing in the visual cortex. Neuroimage 2003, 20: 1609-1624. Additional material 13. Bentin S, Mouchetant-Rostaing Y, Giard MH, Echallier JF, Pernier J: ERP manifestations of processing printed words at different psycholinguistic levels: time course and scalp distribution. J Cogn Neurosci 1999, 11: Additional file 1: Supplementary data on DVAT algorithm. 235-260. Additional file 2: Supplementary data on behavioral measures. 14. Ashley V, Vuilleumier P, Swick D: Time course and specificity of event- Additional file 3: Supplementary data on Channel Locations. related potentials to emotional expressions. Neuroreport 2004, 15: 211-216. 15. Batty M, Taylor MJ: Early processing of the six basic facial emotional expressions. Brain Res Cogn Brain Res 2003, 17: 613-620. 16. Blau VC, Maurer U, Tottenham N, McCandliss BD: The face-specific N170 Acknowledgements component is modulated by emotional facial expression. Behav Brain This research was partially supported by grants from CONICET, FINECO and Funct 2007, 3:7. CG090325006 (Diego Portales University). 17. Pizzagalli DA, Lehmann D, Hendrick AM, Regard M, Pascual-Marqui RD, Davidson RJ: Affective judgments of faces modulate early activity Author details (approximately 160 ms) within the fusiform gyri. Neuroimage 2002, 16: Institute of Cognitive Neurology (INECO) and Institute of Neuroscience, 663-677. Favaloro University, Argentina. National Scientific and Technical Research 18. Dufey M, Hurtado E, Fernandez AM, Ibañez A: Exploring the relationship Council (CONICET), Buenos Aires, Argentina. Laboratory of Cognitive between vagal tone and event-related potentials in response to an Neuroscience, Universidad Diego Portales, Santiago, Chile. Pontificia affective picture task. Social Neuroscience 2010, 6: 48-62. Universidad Católica de Chile. Laboratory of Epistemology and History of 19. Schacht A, Sommer W: Emotions in word and face processing: early and Medicine (LEPHIM), Instituto Universitario del Hospital Italiano de Buenos late cortical responses. Brain Cogn 2009, 69: 538-550. Aires, Argentina. Integrative Neuroscience Laboratory, Physics Department, 20. Fruhholz S, Fehr T, Herrmann M: Early and late temporo-spatial effects of University of Buenos Aires, Argentina. contextual interference during perception of facial affect. Int J Psychophysiol 2009, 74: 1-13. Authors’ contributions 21. Gonzalez-Garrido AA, Ramos-Loyo J, Lopez-Franco AL, Gomez-Velazquez FR: The study was conceived and designed by AI. Data were captured and Visual processing in a facial emotional context: an ERP study. Int J analyzed by AI, RE AL and EH. EH, AP, RR, AL, SB and AI participated in Psychophysiol 2009, 71: 25-30. discussions about analysis and interpretation of data and wrote parts of the 22. Righart R, De Gelder B: Recognition of facial expressions is influenced by article. FM and EH revised the manuscript, critically contributing to its emotional scene gist. Cogn Affect Behav Neurosci 2008, 8: 264-272. generation. All authors read and approved the final manuscript. 23. Simon G, Bernard C, Largy P, Lalonde R, Rebai M: Chronometry of visual word recognition during passive and lexical decision tasks: an ERP Competing interests investigation. Int J Neurosci 2004, 114: 1401-1432. The authors declare that they have no competing interests. 24. Scott GG, O’Donnell PJ, Leuthold H, Sereno SC: Early emotion word processing: evidence from event-related potentials. Biol Psychol 2009, 80: Received: 15 October 2010 Accepted: 13 April 2011 95-104. Published: 13 April 2011 25. Cunningham WA, Espinet SD, DeYoung CG, Zelazo PD: Attitudes to the right- and left: frontal ERP asymmetries associated with stimulus valence References and processing goals. Neuroimage 2005, 28: 827-834. 1. Barrett LF, Lindquist KA, Gendron M: Language as context for the 26. Werheid K, Alpay G, Jentzsch I, Sommer W: Priming emotional facial perception of emotion. Trends Cogn Sci 2007, 11: 327-332. expressions as evidenced by event-related brain potentials. Int J 2. Havas DA, Glenberg AM, Rink M: Emotion simulation during language Psychophysiol 2005, 55: 209-219. comprehension. Psychon Bull & Rev 2007, 14: 436-441. Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 13 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 27. Codispoti M, Ferrari V, Junghofer M, Schupp HT: The categorization of and in patients with Alzheimer-type dementia. Neuropsychologia 1994, 32: natural scenes: brain attention networks revealed by dense sensor ERPs. 381-398. Neuroimage 2006, 32: 583-591. 52. Rugg MD, Mark RE, Gilchrist J, Roberts RC: ERP repetition effects in 28. Dillon DG, Cooper JJ, Grent-’t-Jong T, Woldorff MG, LaBar KS: indirect and direct tasks: effects of age and interitem lag. Dissociation of event-related potentials indexing arousal and Psychophysiology 1997, 34: 572-586. semantic cohesion during emotional word encoding. Brain Cogn 2006, 53. Astikainen P, Hietanen JK: Event-related potentials to task-irrelevant 62: 43-57. changes in facial expressions. Behav Brain Funct 2009, 5: 30. 29. Gonzalez R, Lopez V, Haye A, Hurtado E, Navarro A, Ibañez A: N170 and 54. Cornejo C, Simonetti F, Ibanez A, Aldunate N, Ceric F, Lopez V, et al: LPP discrimination of same raze versus other raze facial stimuli and Gesture and metaphor comprehension: electrophysiological evidence of positive and negative words in indigenous and non-indigenous cross-modal coordination by audiovisual stimulation. Brain Cogn 2009, participants. Clinical Neurophysiology 2008, 119: e155. 70: 42-52. 30. Guex R, Ceric F, Hurtado E, Navarro A, Gonzalez R, Manes F, Ibanez A: 55. Ibáñez A, Manes F, Escobar J, Trujillo N, Andreucci P, Hurtado E: Gesture Performance errors of ingroup/outgroup stimuli and valence association influences the processing of figurative language in non-native speakers: in the implicit association task: Brain bias of ingroup favoritism. Open ERP evidence. Neurosci Lett 2010, 471: 48-52. Neurosci J 2011. 56. Wong TK, Fung PC, McAlonan GM, Chua SE: Spatiotemporal dipole source 31. Hurtado E, Haye A, Gonzalez R, Manes F, Ibañez AM: Contextual blending localization of face processing ERPs in adolescents: a preliminary study. of ingroup/outgroup face stimuli and word valence: LPP modulation Behav Brain Funct 2009, 5: 16. and convergence of measures. BMC Neurosci 2009, 10: 69. 57. Lopez V, Lopez-Calderon J, Ortega R, Kreither J, Carrasco X, Rothhammer P, 32. Ibáñez A, Gleichgerrcht E, Hurtado E, González R, Haye A, Manes F: Early et al: Attention-deficit hyperactivity disorder involves differential cortical Neural Markers of Implicit Attitudes: N170 Modulated by Intergroup and processing in a visual spatial attention paradigm. Clin Neurophysiol 2006, Evaluative Contexts in IAT. Front Human Neurosci 4: 188. 117: 2540-2548. 33. Codispoti M, Ferrari V, Bradley MM: Repetition and event-related 58. Maurer U, Bucher K, Brem S, Benz R, Kranz F, Schulz E, et al: potentials: distinguishing early and late processes in affective picture Neurophysiology in preschool improves behavioral prediction of reading perception. J Cogn Neurosci 2007, 19: 577-586. ability throughout primary school. Biol Psychiatry 2009, 66: 341-348. 34. Olofsson JK, Nordin S, Sequeira H, Polich J: Affective picture processing: an 59. Greenwald AG, Nosek BA, Banaji MR: Understanding and using the integrative review of ERP findings. Biol Psychol 2008, 77: 247-265. implicit association test: I. An improved scoring algorithm. J Pers Soc 35. Rozenkrants B, Polich J: Affective ERP processing in a visual oddball task: Psychol 2003, 85: 197-216. arousal, valence, and gender. Clin Neurophysiol 2008, 119: 2260-2265. 60. Mierke J, Klauer KC: Method-specific variance in the implicit association 36. Petroni A, Urquina H, Guex R, Hurtado E, Manes F, Sigman M, Ibañez A: test. J Pers Soc Psychol 2003, 85: 1180-1192. Early cortical measures of valence, stimulus type discrimination and 61. Oken BS, Chiappa KH: Statistical issues concerning computerized analysis interference: association to executive function and social cognition.. of brainwave topography. Ann Neurol 1986, 19: 493-497. 37. Ibáñez A, Petroni A, Urquina H, Torralva T, Torrente F, Hurtado E, Guex R, 62. Arkes HR, Tetlock PE: Attributions of Implicite Prejudice, or “Would Jesse Sigman M, Lischinsky A, Manes F: Cortical deficits in emotion processing Jackson ‘Fail’ the Implicit Association Test?”. Psychological Inquiry 2004, 15: for faces in adults with ADHD: Its relation to social cognition and 257-278. executive functioning.. 63. Conrey FR, Sherman JW, Gawronski B, Hugenberg K, Groom CJ: Separating 38. Ekman P, Friesen WV, Ellsworth P: What emotion categories or dimensions multiple processes in implicit social cognition: the quad model of can observers judge from facial behavior? In Emotion in the human face. implicit task performance. J Pers Soc Psychol 2005, 89: 469-487. Edited by: Ekman P. New York: Cambridge University Press; 1982: 39-55. 64. Hamann S, Canli T: Individual differences in emotion processing. Curr 39. Bentin S, Peled BS: The contribution of task-related factors to ERP Opin Neurobiol 2004, 14: 233-238. repetition effects at short and long lags. Mem Cognit 1990, 18: 359-366. 65. Kinoshita S, Peek-O’Leary M: Does the compatibility effect in the race 40. De Houwer J: A Structural and Process Analysis of the Implicit Implicit Association Test reflect familiarity or affect? Psychon Bull Rev Association Test. J Exp Soc Psychol 2001, 37: 443-451. 2005, 12: 442-452. 41. Rothermund K, Wentura D: Underlying processes in the implicit 66. De Houwer J: The propositional approach to associative learning as an association test: dissociating salience from associations. J Exp Psychol Gen alternative for association formation models. Learn Behav 2009, 37: 1-20. 2004, 133: 139-165. 67. De Houwer J: A structural analysis of indirect measures of attitudes. In 42. Ashley V, Swick D: Consequences of emotional stimuli: Age differences The Psychology of Evaluation: Affective Processes in Cognition and Emotion. on pure and mixed blocks of the emotional Stroop. Behav Brain Funct Edited by: Musch J, Klauer KC. Mahwah,NJ: Lawrence Erlbaum; 2003: 2009, 5: 14. 219-244. 43. Dong G, Hu Y, Zhou H: Event-related potential measures of the intending 68. Zurron M, Pouso M, Lindin M, Galdo S, Diaz F: Event-related potentials process: time course and related ERP components. Behav.Brain Funct with the Stroop colour-word task: timing of semantic conflict. Int J 2010, 6: 15. Psychophysiol 2009, 72: 246-252. 44. Maurer U, Brandeis D, McCandliss B: Fast, visual specialization for reading 69. Carter CS, Van Veen V: Anterior cingulate cortex and conflict detection: in English revealed by the topography of the N170 ERP response. Behav an update of theory and data. Cogn Affect Behav Neurosci 2007, 7: Brain Funct 2005, 1: 1-13. 367-379. 45. Simon G, Petit L, Bernard C, Rebai M: N170 ERPs could represent a 70. Egner T, Hirsch J: The neural correlates and functional integration of logographic processing strategy in visual word recognition. Behav Brain cognitive control in a Stroop task. Neuroimage 2005, 24: 539-547. Funct 2007, 3: 21. 71. Markela-Lerenc J, Ille N, Kaiser S, Fiedler P, Mundt C, Weisbrod M: 46. Zhu XR, Zhang HJ, Wu TT, Luo WB, Luo YJ: Emotional conflict occurs at an Prefrontal-cingulate activation during executive control: which comes early stage: evidence from the emotional face-word Stroop task. Neurosci first? Brain Res Cogn Brain Res 2004, 18: 278-287. Lett 2010, 478: 1-4. 72. Simon JR, Rudell AP: Auditory S-R compatibility: the effect of an 47. Ito TA, Bartholow BD: The neural correlates of race. Trends Cogn Sci 2009, irrelevant cue on information processing. J Appl Psychol 1967, 51: 300-304. 13: 524-531. 73. Melara RD, Wang H, Vu KP, Proctor RW: Attentional origins of the Simon 48. Tsao DY, Livingstone MS: Mechanisms of face perception. Annu Rev effect: behavioral and electrophysiological evidence. Brain Res 2008, Neurosci 2008, 31: 411-437. 1215: 147-159. 49. Doyle MC, Rugg MD: Word repetition within- and across-visual fields: 74. Hanslmayr S, Pastotter B, Bauml KH, Gruber S, Wimber M, Klimesch W: The an event-related potential study. Neuropsychologia 1998, 36: electrophysiological dynamics of interference during the Stroop task. J 1403-1415. Cogn Neurosci 2008, 20: 215-225. 50. Rugg MD: Event-related brain potentials dissociate repetition effects of 75. Grossmann T: The development of emotion perception in face and voice high- and low-frequency words. Mem Cognit 1990, 18: 367-379. during infancy. Restor Neurol Neurosci 2010, 28: 219-236. 51. Rugg MD, Pearl S, Walker P, Roberts RC, Holdstock JS: Word repetition 76. Ohman A, Mineka S: Fears, phobias, and preparedness: toward an effects on event-related potentials in healthy young and old subjects, evolved module of fear and fear learning. Psychol Rev 2001, 108: 483-522. Ibáñez et al. Behavioral and Brain Functions 2011, 7:8 Page 14 of 14 http://www.behavioralandbrainfunctions.com/content/7/1/8 77. Aravena P, Hurtado E, Riveros R, Cardona JF, Manes F, Ibáñez A: 100. Bradley MM, Hamby S, Low A, Lang PJ: Brain potentials in perception: Applauding with closed hands: neural signature of action-sentence picture complexity and emotional arousal. Psychophysiology 2007, 44: compatibility effects. PLoS One 2010, 5: e11751. 364-373. 78. Barutta J, Cornejo C, Ibáñez A: Theories and theorizers: a contextual 101. Cano ME, Class QA, Polich J: Affective valence, stimulus attributes, and approach to theories of cognition. Integr Psychol Behav Sci 2011. P300: color vs. black/white and normal vs. scrambled images. Int J 79. Cosmelli D, Ibáñez A: Human cognition in context: on the biologic, Psychophysiol 2009, 71: 17-24. cognitive and social reconsideration of meaning as making sense of 102. Delplanque S, N’diaye K, Scherer K, Grandjean D: Spatial frequencies or action. Integr Psychol Behav Sci 2008, 42: 233-244. emotional effects? A systematic measure of spatial frequencies for IAPS 80. Gleichgerrcht E, Ibanez A, Roca M, Torralva T, Manes F: Decision Making pictures by a discrete wavelet analysis. J Neurosci Methods 2007, 165: Cognition in Neurodegenerative Diseases. Nat Rev Neurol 2010, 6: 144-150. 611-623. 103. Rozenkrants B, Olofsson JK, Polich J: Affective visual event-related 81. Ibáñez A, Lopez V, Cornejo C: ERPs and contextual semantic potentials: arousal, valence, and repetition effects for normal and discrimination: degrees of congruence in wakefulness and sleep. Brain distorted pictures. Int J Psychophysiol 2008, 67: 114-123. Lang 2006, 98: 264-275. 104. Bradley MM, Codispoti M, Cuthbert BN, Lang PJ: Emotion and motivation I: 82. Ibáñez A, Haye A, González R, Hurtado E, Henríquez R: Multi-level analysis defensive and appetitive reactions in picture processing. Emotion 2001, of cultural phenomena: The role of ERP approach to prejudice. J Theor 1: 276-298. Soc Behav 2010, 39: 81-110. doi:10.1186/1744-9081-7-8 83. Ibáñez A, Riveros R, Aravena P, Vergara V, Cardona JF, Garcia L, et al: When Cite this article as: Ibáñez et al.: Facial and semantic emotional context is difficult to integrate: Cortical measures of congruency in interference: A pilot study on the behavioral and cortical responses to schizophrenics and healthy relatives from multiplex families. Schizophr the dual valence association task. Behavioral and Brain Functions 2011 7:8. Res 2011, 126: 303-305. 84. Ibáñez A, Toro P, Cornejo C, Hurquina H, Manes F, Weisbrod M, et al: High contextual sensitivity of metaphorical expressions and gesture blending: A video event-related potential design. Psychiatry Res 2011, 191: 68-75. 85. Riveros R, Manes F, Hurtado E, Escobar M, Martin RM, Cetkovich M, et al: Context-sensitive social cognition is impaired in schizophrenic patients and their healthy relatives. Schizophr Res 2010, 116: 297-298. 86. San Martin R, Manes F, Hurtado E, Isla P, Ibáñez A: Size and probability of rewards modulate the feedback error-related negativity associated with wins but not losses in a monetarily rewarded gambling task. Neuroimage 2010, 51: 1194-1204. 87. Labuschagne I, Croft RJ, Phan KL, Nathan PJ: Augmenting serotonin neurotransmission with citalopram modulates emotional expression decoding but not structural encoding of moderate intensity sad facial emotional stimuli: an event-related potential (ERP) investigation. J Psychopharmacol 2010, 24: 1153-64. 88. Ito TA, Thompson E, Cacioppo JT: Tracking the timecourse of social perception: the effects of racial cues on event-related brain potentials. Pers Soc Psychol Bull 2004, 30: 1267-1280. 89. Maurer U, Rossion B, McCandliss BD: Category specificity in early perception: face and word n170 responses differ in both lateralization and habituation properties. Front Hum Neurosci 2008, 2: 18. 90. Schupp HT, Junghofer M, Weike AI, Hamm AO: The selective processing of briefly presented affective pictures: an ERP analysis. Psychophysiology 2004, 41: 441-449. 91. Vuilleumier P, Pourtois G: Distributed and interactive brain mechanisms during emotion face perception: evidence from functional neuroimaging. Neuropsychologia 2007, 45: 174-194. 92. Fischler I, Bradley M: Event-related potential studies of language and emotion: words, phrases, and task effects. Prog Brain Res 2006, 156: 185-203. 93. Dien J, Spencer KM, Donchin E: Localization of the event-related potential novelty response as defined by principal components analysis. Brain Res Cogn Brain Res 2003, 17: 637-650. 94. Proverbio AM, Del ZM, Zani A: The emergence of semantic categorization in early visual processing: ERP indices of animal vs. artifact recognition. BMC Neurosci 2007, 8: 24. 95. Segalowitz SJ, Zheng X: An ERP study of category priming: evidence of early lexical semantic access. Biol Psychol 2009, 80: 122-129. Submit your next manuscript to BioMed Central 96. Brendan M, Aidman E: Individual Differences in Implicit and Declared and take full advantage of: Self-Esteem as Predictors of Response to Negative Performance Evaluation: Validating Implicit Association Test as a Measure of Self- • Convenient online submission Attitudes. Int J Test 2004, 4: 19-42. 97. De Houwer J, Teige-Mocigemba S, Spruyt A, Moors A: Implicit measures: A • Thorough peer review normative analysis and review. Psychol Bull 2009, 135: 347-368. • No space constraints or color figure charges 98. Greenwald AG, McGhee DE, Schwartz JL: Measuring individual differences • Immediate publication on acceptance in implicit cognition: the implicit association test. J Pers Soc Psychol 1998, 74: 1464-1480. • Inclusion in PubMed, CAS, Scopus and Google Scholar 99. Greenwald AG, Poehlman TA, Uhlmann EL, Banaji MR: Understanding and • Research which is freely available for redistribution using the Implicit Association Test: III. Meta-analysis of predictive validity. J Pers Soc Psychol 2009, 97: 17-41. Submit your manuscript at www.biomedcentral.com/submit
Journal
Behavioral and Brain Functions – Springer Journals
Published: Apr 13, 2011