QEEG and LORETA in Teenagers With Conduct Disorder and Psychopathic Traits

Abstract

Background. Few studies have investigated the impact of the psychopathic traits on the EEG of teenagers with conduct disorder (CD). To date, there is no other research studying low-resolution brain electromagnetic tomography (LORETA) technique using quantitative EEG (QEEG) analysis in adolescents with CD and psychopathic traits. Objective. To find electrophysiological differences specifically related to the psychopathic traits. The current investigation compares the QEEG and the current source density measures between adolescents with CD and psychopathic traits and adolescents with CD without psychopathic traits. Methods. The resting EEG activity and LORETA for the EEG fast spectral bands were evaluated in 42 teenagers with CD, 25 with and 17 without psychopathic traits according to the Antisocial Process Screening Device. All adolescents were assessed using the DSM-IV-TR criteria. The EEG visual inspection characteristics and the use of frequency domain quantitative analysis techniques (narrow band spectral parameters) are described. Results. QEEG analysis showed a pattern of beta activity excess on the bilateral frontal-temporal regions and decreases of alpha band power on the left central-temporal and right frontal-central-temporal regions in the psychopathic traits group. Current source density calculated at 17.18 Hz showed an increase within fronto-temporo-striatal regions in the psychopathic relative to the nonpsychopathic traits group. Conclusions. These findings indicate that QEEG analysis and techniques of source localization may reveal differences in brain electrical activity among teenagers with CD and psychopathic traits, which was not obvious to visual inspection. Taken together, these results suggest that abnormalities in a fronto-temporo-striatal network play a relevant role in the neurobiological basis of psychopathic behavior.

Keywords

EEG QEEG LORETA psychopathic traits conduct disorder

Introduction

Conduct disorder (CD) is one of the psychiatric illness frequently assessed by child mental services.¹ CD is a diagnosis typically assigned to individuals younger than 18 years, who habitually violate the rights of others, and whose behavior does not conform to the law or social norms appropriate for their age.² The annual prevalence of CD is 2% to 10%. It is more frequently diagnosed in boys, at least in part due to boys’ tendencies to act out violently.²

One form of CD is marked by the presence of psychopathic traits. These traits have a core callous-unemotional component and an impulsive-antisocial component.² These components are related to 2 essential cognitive impairments. The first is associated with reductions in specific forms of emotional empathy. This functional alteration is associated with reduced amygdala and ventromedial prefrontal cortex responsiveness to distress cues. The second cognitive impairment is related to aspects of decision making, specifically in reinforcement learning and the representation of reinforcement expectancies. This impairment may relate more to the impulsive-antisocial component of psychopathic traits.³

Boys with callous-unemotional conduct problems present impairments on neuropsychological tasks tapping the functionality of the ventrolateral/orbitofrontal cortex and the amygdala functioning.⁴ Consistent with this finding, researchers using functional magnetic resonance imaging (fMRI) techniques have identified limbic-prefrontal circuit abnormalities in children and adolescents with high psychopathic traits^5-7 and reduced activity within regions associated with empathic pain, including rostral anterior cingulate cortex, ventral striatum (putamen), and amygdala.⁸ In the past decade, several studies using MRI have described brain structural differences in boys with CD. They found

reduced volumes in the temporal lobe,^9,10 insula,¹¹ amygdala,^10,11 hippocampus, orbitofrontal cortex, ventromedial prefrontal cortex,¹⁰ right middle frontal cortex, and left parietal lobule¹²

increased brain tissue volumes in the left putamen, left ansa peduncularis, right superiomedial prefrontal cortex, left inferior frontal cortex, and right medial temporal, right orbitofrontal cortex regions¹²

increased gray matter and decreased white matter concentrations in the right superior frontal lobe (subgyral), right dorsal anterior cingulate (limbic lobe), right superior temporal gyrus and left precuneus^13,14

decreased overall mean cortical thickness associate with thinning of the cingulate, prefrontal, and insular cortices in children with disruptive behavior disorders.¹⁵

Two recent voxel-based morphometry studies reported paralimbic gray matter reductions in incarcerated male adolescent with psychopathic traits.^16,17

Few studies have investigated the impact of the psychopathic traits on the EEG of teenagers with CD. The construct of psychopathy as a clinical syndrome has been applied as a discriminating variable in the identification and classification of children who will develop chronic offending careers.^18-21 Nevertheless, there is considerable controversy regarding the assessment and diagnosis of psychopathy in juveniles. Some have argued that the early identification of psychopathic traits might allow clinicians to intervene sooner and more effectively with youths who presumably are at high risk to continue antisocial behavior into adulthood.²²

Abnormal EEG activity had been described in delinquent and antisocial adolescents.²³ Earlier studies generally based on qualitative analysis reported a high amount of slow EEG activity in children with CD²⁴ and paroxysmal abnormalities in repeatedly violent delinquent boys.²⁵ Alpha slowing among children and adolescents had been associated with later delinquency^26,27 while the beta activity with externalizing spectrum disorders,^28,29 such as conduct disorder and antisocial behavior. A number of studies have reported increase of beta activity and decrease of alpha power in adult psychopaths,³⁰ in impulsive subjects,³¹ and in subjects with psychiatric disorders considered important precursors of psychopathy as moderate intermittent explosive disorder (mIED),³² and attention deficit/hyperactivity disorder (ADHD).^33-35

EEG provides an excellent index of brain activation level, particularly in cortical regions.¹⁸ Quantitative EEG (QEEG) analysis moved away from the traditional analysis of graphoelements at certain electrodes to a comprehensive analysis of the brain’s electric field at the scalp. A deeper understanding of cortical coupling is possible by studying the correlations between current sources derived from the surface EEG using an inverse method.³⁶ Recent advances in EEG recording technology and EEG analysis methods make the window into the brain much more transparent, and the signal source relationship has become clearer.³⁷

Data from electrophysiology will be used to clarify the neurobiological basis of children with presumable psychopathy and their risk of being violent offenders later during their adult lives. In a previous study of EEG in psychopathic offenders,³⁰ we described a certain pattern of EEG abnormalities when compared with nonpsychopathic offenders used as controls. The aim of this study was to investigate whether EEG differences can be found between adolescents with CD with comorbid psychopathy and CD adolescents without this comorbidity and to quantify the nature of these differences. It is hypothesized that adolescents with psychopathic traits will have increased beta activity with reduced levels of alpha compared with nonpsychopathic traits adolescents. We also examined whether the QEEG abnormalities are associated differentially with traits of psychopathy assessed by the Antisocial Process Screening Device (APSD) scale.

Materials and Methods

Subjects

The study included 42 male CD teenagers, mean age was 15.7 years (SD = 1.02) from a special conduct disorder school located in Havana City. Assessment was conducted during a 2-year period, from January 2010 to December 2011.

Clinical assessment was performed by childhood psychiatrists directly related to this study. All they were well trained in the application of all structured, semistructured interviews, and diagnostic tools used. Diagnostic interviews were carried out separately with participants and caregivers. Clinical interviews incorporated information from as many sources as were available. These included a history given by a parent or guardian, school reports, data from any other health professionals and behavioral observations during the assessment. This study used DSM-IV-TR³⁸ criteria for the diagnosis of CD and APSD in order to evaluate psychopathic traits.³⁹ Twenty-five adolescents were diagnosed with CD and psychopathic traits (mean age 15.66 years, SD = 1.21). The control group consisted of 17 adolescents from the same school diagnosed with CD without psychopathic traits (mean age 15.79 years, SD = 0.82).

In order to be included in the CD with comorbid psychopathic traits group, adolescents were required to meet the full diagnostic criteria for psychopathic disorder according to APSD scale created by Frick and Hare.³⁹ This scale is an extension of the Psychopathy Checklist-Revised (PCL-R) for adults.⁴⁰ The APSD is a 20-item behavior rating scale typically completed by a parent and/or teacher. This instrument does not specify a particular amount of time necessary for the rater to have known the child. The APSD was designed to assess a lack of empathy and callous behavior toward others, which may reflect a distinct subtype of those with more severe externalizing behavior problems and poorer developmental outcomes. Factor analysis of the APSD uncovered 3 dimensions: Narcissism, Impulsivity, and Callous-Unemotional (CU). The CU dimension includes items such as “concerned about the feelings of others” and “does not show emotions.”⁴¹

Sociobiographical, criminological, medical, and psychopathological information was recorded on a standardized score sheet and entered into a computer system for further processing.

Exclusion criteria in this study were presence of any current or previous history of neurological disease that can influence results of EEG records, as the presence of an abnormal prenatal, perinatal or neonatal event, a disorder of consciousness, a head injury resulting in cognitive deficits, antecedents of central nervous system (CNS) diseases, convulsions, psychiatric abnormalities (other than CD, ADHD, or Oppositional Defiant Disorder) or substance abuse disorders. None of the adolescents was taking any medication at the time of testing. Participation in the study was voluntary, written informed consent was obtained from all parents. The study was approved by the Ethics Committee of the Cuban Center for Neurosciences.

EEG Procedure

EEG was recorded using a 21-channel digital EEG hardware and software package from Neuronic SA (MEDICID IV, Cuba). Electrodes were placed in 19-electrodes sites (Fp1, Fp2, Fz, F3, F4, F7, F8, Cz, C3, C4, T3, T4, T5, T6, Pz, P3, P4, O1, and O2) according to the international 10-20 system, using surface electrodes referenced to linked earlobes with a ground electrode attached to the forehead. Impedance for all electrodes was kept ≤5 kohm during the entire record. EEG was amplified by 10 600, with a bandpass at 0.5 to 30 Hz and sampled through a 12-bit analog-to-digital converter at 200 Hz.

Electrodes were fitted while subjects were familiarizing with the testing equipment and procedures. EEG record was carried out in a quiet, air-conditioned room with the experimenter and recording equipment present. All subjects were instructed to relax and to remain still during testing to minimize artifacts produced by ocular movements and to avoid excessive blinking. During the recording the subjects were awake with eyes closed, seated on a reclining chair. All recordings were performed in the morning.

Each resting EEG was obtained during 8 to 10 minutes with the subjects eyes closed. Subsequently, 2 minutes of alternation between closed and opened eyes were recorded, following by 3 minutes carrying out hyperventilation, 2 minutes of recovery were also recorded. Taking into account that sleepiness might have caused an enhancement of theta activity, the individual vigilance level was checked during EEG acquisition, seeking for slowing of the EEG background activity or for the appearance of typical sleep patterns. In addition, at the end of the recording process, individuals were asked about whether they were awake during the whole recording.

Visual Assessment of the EEG

A clinical neurophysiologist (first author of current research report) performed and interpreted the EEG analysis. Several bipolar montages were used for off line EEG interpretation. The EEG was considered NORMAL if it had an adequate organization of the background activity (according to the subject’s age), well-defined spatial differentiation, rhythmical alpha activity, and absence of paroxysmal activity. The SLOW EEG subgroup was characterized by the presence of persistent nonrhythmic theta-delta slow waves. PAROXYSMAL category included the EEG with activity such as spikes, sharp wave, and spike and wave. EEGs with both types of previously described abnormalities were considered in the SLOW and PAROXYSMAL category. Ratios and percentages in all categories were calculated.

Quantitative EEG Analysis

Tracings were visually inspected and edited off-line in order to eliminate epochs with movement artifacts, eye blinking, muscle activity, or drowsiness. For each subject, 20 to 24 EEG segments (without artifact) of 2.56 seconds at the closed eyes state were selected. Spectral analysis using fast Fourier transform (FFT) was carried out in order to obtain the cross spectral matrices estimation in all individual records.^42,43 Cross-spectral matrices were calculated for every 0.39 Hz, from 0.78 to 19.50 Hz.

EEG Source Estimation

The EEG tomography was conducted using The low-resolution electromagnetic tomography (LORETA).⁴⁴ Each LORETA file was computed from the frequency domain EEG cross spectra corresponding to the the selected EEG frequencies, one in the alpha range (9.37 Hz) and another in the beta band (17.18 Hz). These 2 specific frequencies were selected taking in consideration the results obtained from the comparison made between the Z cross spectra of the 2 groups (see below). Current density vectors (CDVs) were calculated for each individual from all the data segments using the Neuronic Source Localizer software (Neuronic SA).

Compute lead field (LF) was the first step for brain electrical tomography calculation. This software computes electric LF using the isotropic and piecewise homogeneous three spheres head model.^45,46 In this research current density distribution inside the head was calculated using a 3-dimensional grid including all brain volume. Anatomical information for LF computation was provided from a brain average template obtained at the Neurological Institute of Montreal.⁴⁷ In this study, the source space predefined was cortical gray matter plus basal ganglia and grids of 4 × 4 × 4. The number of generators used was 21 926.

The tomographic solutions were seen using the Neuronic Tomographic Viewer (Neuronic SA)

Statistical Analysis

In order to determine the dependence among categorical variables (demographic, behavioral, and visual inspection of EEG), Pearson chi-square test and 1-way analysis of variance test were used for analysis (Statistica 6.0 for Windows). The level of statistical significance was set at .05 for all the tests.

QEEG Spectrum Analysis

In order to compare the psychopathic and nonpsychopathic traits groups with normative values the Cuban normative database was used. Zsava et al⁴⁸ in 1997 reported the methodology for constructing the normative database. This was constructed from the EEG of 211 normal subjects (105 males, 106 females) with an age range from 5 to 97 years. The EEG was recorded using the same conditions in each subject. The Z standardized statistic was calculated for all spectral parameters

z = \frac{x - μ (age)}{σ (age)}

where µ and σ are the mean and standard deviation (SD), respectively, estimated in the normative group for the specific age of the subject been compared. The x value is the observation of each subject.

Statistical Methodology to Compare the Z Spectra Mean of Both Groups

In order to evaluate differences between the Z spectra of both groups, a permutation test was used.^49-52 The permutation test has the following advantages: free distribution, which controls the experiment wise error for simultaneous univariate comparisons, no assumption of an underlying correlation structure, providing exact P values for any number of individuals, frequency points, and recording sites.

Multivariate statistics were used to summarize and test differences between 2 Z spectra obtained from the method of nonparametric combination of dependent permutation tests.⁵³ The method of nonparametric combination methodology of a finite number of dependent permutation tests is a general tool for multivariate testing problems. These multivariate statistics are calculated according to combined univariate t test. The combine function used was the maximum. Two separate partial tests were performed of

the maximum of t statistics in each electrode for all frequencies

the maximum of t statistics in each frequency for all electrodes.

These methodologies were obtained as follows:

Step 1: The observations were repeatedly permutated between groups. Both statistics were calculated for every repetition.

Step 2: The empirical distribution was estimated using the statistics calculated in the step above for all permutations.

Step 3: Significance levels were set using the t and max(t) of the original samples in the empirical distribution estimated in step 2.^49-52

Inverse Solution Analysis

In order to identify significant regional differences between groups in current density (CDV) for beta and alpha EEG bands, a t test for independent samples with correction for multiple comparisons was performed (Neuronic Statistica software, Neuronic SA). The final outcome was a map of the t test values for each voxel thresholded at a false discovery rate (FDR) q = 0.1.

Coordinates of main activation were represented in Talairach space (Neuronic Tomographic Viewer, Neuronic SA).

Results

Demographic and Behavioral Results

Analyses of demographic data indicated no significant differences in age between the 2 groups (Table 1). Scores for APSD showed significant group differences. ADHD, dissocial conduct disorder, and oppositional defiant disorder as comorbility with CD diagnosis did not show significant group differences (Table 1).

Table 1.

Comparison Between the Psychopathic Traits and the Nonpsychopathic Traits Groups on Demographic and Behavioral Variables.

Variable	Psychopathic Traits Group	Nonpsychopathic Traits Group	Statistic
Age, years	15.66 (1.21)	15.79 (0.81)	F = 0.70
APSD score	27.76 (4.30)	13.63 (4.16)	χ² = 0.00^a
ADHD	11	5	χ² = 0.09
Dissocial conduct disorder	16	12	χ² = 0.38
Oppositional defiant disorder	5	2	χ² = 0.28

Abbreviations: ADHD, attention deficit/hyperactivity disorder; APSD, Antisocial Process Screening Device.

Analysis of variance or Pearson χ², P < .05.

EEG Visual Inspection

The resting EEG visual analyses revealed a high percentage of EEG abnormalities in both studied groups. The total amount of abnormal EEGs were of 24 adolescent (96.0%) in the psychopathic traits group and 12 subjects (70.58%) in the control group. The number of individuals with some type of EEG abnormality in the whole sample was of 36 (85.7%).

Table 2 presents details of the EEG visual analysis results. In this analysis, both groups (psychopathic and nonpsychopathic traits) results were very similar. PAROXYSMAL EEG was the category including the highest number of subjects (around 43%), followed by SLOW EEG category (28.6%). Twelve adolescents from both groups were included in the other 2 categories.

Table 2.

Classification of the Adolescents’ EEG in Both Groups by Visual Inspection.

Group	Normal	Slow	Paroxysmal	Slow and Paroxysmal
Psychopath	1 (4.0%)	9 (36.0%)	12 (48.0%)	3 (12.0%)
Nonpsychopath	5 (29.4%)	3 (17.6%)	6 (35.3%)	3 (17.6%)

The Pearson chi-square test comparison found significant differences between the 2 groups regarding the total EEG with abnormalities versus normal EEG by visual inspection (χ² = 5.45, df = 1, P = .02).

Table 3 shows the topographical distribution of the EEG abnormalities found in both groups. Frontal lobe was the single brain region most affected by EEG abnormalities in both groups. A comparison using the Pearson chi-square was made taking into consideration only the PAROXYSMAL category (because it was the EEG abnormality most frequent in both groups) and comparing frontal versus temporal localizations. There were differences between these topographical distributions of paroxysmal EEG activity between both groups (χ² = 4.02, df = 1, P = .04). Paroxysmal EEG was more frequent at frontal derivations in the psychopathic traits group while temporal regions were more affected by this EEG abnormality in the nonpsychopathic traits group.

Table 3.

Topographic Distribution of EEG Abnormalities.

Group	Slow		Paroxysmal		Slow and Paroxysmal
Group	Frontal	Widespread	Frontal	Temporal	Frontal	Widespread
Psychopathic	6 (24.0%)	3 (12.0%)	11 (44.0%)	1 (4.0%)	1 (4.0%)	2 (8.0%)
Nonpsychopathic	1 (5.9%)	2 (11.8%)	3 (17.6%)	3 (17.6%)	1 (5.9%)	2 (11.8%)

Quantitative EEG Analysis

The mean duration of EEG in the eyes-closed state for the psychopathic traits CD group was 50.08 seconds (SD ±21.86) and 49.35 seconds (SD ±12.56) for the nonpsychopathic traits CD group. No significant differences between the 2 groups were found (P = .90).

Significant statistical differences between the mean parameters of cross spectral measures of psychopathic and nonpsychopathic traits groups using the permutation test were found in the beta band within a frequency range of 13.67 to 18.75 Hz at bilateral frontal-temporal regions. The power value for this frequency was increased for the psychopathic traits group. In contrast, within the 9.37 to 9.76 Hz frequency range of the alpha band a decrease of the energy was found at the right frontal-central-temporal regions and the parietal-central lead for this same group.

Correlation Between EEG Frequencies (QEEG) and APSD Scores

Significant positive correlations were found between the “blames others for mistakes” (r = 0.38; P < .02), “lies easily and skillfully” (r =0.38; P < .01), “keeps promises” (r = 0.34; P < .03), “brags about accomplishments” (r = 0.32; P < .05) with the beta energy level at central parietal region. These correlations were not statistically significant after correction for multiple comparisons.

EEG Source Analysis

Comparison of Beta Activity Sources (LORETA) Between Groups

The sources of the increased beta activity at 17.18 Hz were localized on the following Brodmann’s areas: 11, 12 (orbitofrontal cortex); 6 medial frontal; 44 (opercular portions of the inferior frontal gyrus); 8 (superior frontal gyrus) and 13 (insula) in both hemispheres; 7 (postcentral area); 4 precentral, 22 (superior temporal gyrus), 21 (medial temporal gyrus); 38 (superior temporal pole), putamen and globus pallidus in the left hemisphere and 45 (tri portions of the inferior frontal gyrus) in the psychopathic traits group when they were compared with nonpsychopathic offenders (FDR q = 0.1) (Table 4; Figure 1).

Table 4.

Brain Areas Showing Statistical Differences (q = 0.1) in Beta Frequency (17.18 Hz) Between Psychopathic Traits and Nonpsychopathic Teenagers According to Voxel-by-Voxel Analysis With Low-Resolution Brain Electromagnetic Tomography (LORETA).^a

x (mm)	y (mm)	z (mm)	Brodmann Area(s)	Anatomical Locations
41	59	−8	11, 12	Orbitofrontal cortex
−29	57	−4	11, 12	Orbitofrontal cortex
37	51	12	6	Medial frontal
−29	54	12	6	Medial frontal
49	13	20	44	Opercular portions of the inferior frontal gyrus
−47	9	28	44	Opercular portions of the inferior frontal gyrus
19	65	12	8	Superior frontal gyrus
−32	62	12	8	Superior frontal gyrus
34	17	12	13	Insula
30	21	12	13	Insula
−61	−3	16	7	Postcentral area
−56	2	16	4	Precentral
−55	−6	−5	22	Superior temporal gyrus
−47	−1	−16	21	Medial temporal gyrus
−43	1	−20	38	Superior temporal pole
−23	4	8		Putamen
−10	5	−4		Globus pallidus
−54	21	12	45	Tri portions of the inferior frontal gyrus

Coordinates are given in millimeters, and the origin is at the anterior commissure. For x, negative values represent left, and positive values represent right. For y, negative values represent posterior, and positive values represent anterior. For z, negative values represent inferior, and positive values represent superior.

Figure 1.

The anatomical distribution of the sources of beta activity (17.18 Hz) increased in psychopathic traits group were localized a broad region within both hemispheres. The main region where the sources were found were orbitofrontal cortex, medial frontal; opercular portion of the inferior frontal gyrus, superior frontal gyrus and insula in both hemispheres, postcentral area, precentral, superior temporal gyrus, medial temporal gyrus, superior temporal pole, putamen and globus pallidus in the left hemisphere and tri portion of the inferior frontal gyrus in the psychopath group. R, right hemisphere; L, left hemisphere. Z is the z-axis value for each image of the brain template showed.

There were no significant group differences in alpha band.

Discussion

In the present study, the paroxysmal activity was the abnormality most frequently found through EEG visual inspection. The frontal lobe was the single brain region most affected by both types of EEG abnormalities (paroxysmal and slow) in both groups. There were group differences between topographical distributions of paroxysmal EEG activity. The frontal paroxysmal activity was higher in psychopathic traits group. In the nonpsychopathic traits group, this type of EEG abnormality was localized predominantly in the temporal lobe. Other EEG studies of CD adolescent have typically found similar frequency of EEG abnormality in qualitative EEG assessment as ours.^53,54

Significant differences in QEEG and regional current source density between psychopathic and nonpsychopathic traits groups were also found. The psychopathic traits group had more beta energy at bilateral frontal-temporal regions and less alpha activity in the right frontal-central-temporal regions and the midline parietal-central lead than the control group. LORETA showed differences especially in the frontal and temporal cortices.

Increase of beta activity and decrease of alpha power were found in the psychopathic traits group when QEEG results were compared with the control group. Similar findings have been demonstrated in children with ADHD,^33,34 in female delinquents,⁵⁵ in psychopathic offenders,³⁰ and adult delinquents with ADHD.³⁵

Beta activity is believed to result from cortical/cortical and thalamo/cortical interactions. Activity in the beta frequency band, a presumed index of the level of cortical arousal, has consistently shown associations with externalizing spectrum disorders^28,29,56 such as substance abuse disorders, conduct disorder, and antisocial behavior.⁵⁷ The increased beta may indicate cortical hyperarousal. Our findings could potentially be interpreted as evidence of cortical disinhibition in the psychopathic traits group as compared with nonpsychopathic traits group.

Decrease alpha activity in adolescents with disruptive behavior had been associated with a failure in functional cortical development (maturational retardation hypothesis).^58,59 This finding is consistent with research that demonstrates less alpha activity levels in delinquent children⁵⁹ and in institutionalized children.⁶⁰ Alpha is the normal rhythm seen in a subject who is awake with his or her eyes closed. It is maximal over the occipital and, to a lesser extent, the parietal regions, and is often less evident over the dominant hemisphere. The adult pattern of dominant alpha rhythm becomes evident in late adolescence and is referred to as the mature EEG.⁶¹ Nevertheless, by the age range of studied subject; we consider that this finding supports the existence of maturational lag of CNS in adolescents with psychopathic traits.

Studies in both animals and humans have shown the thalamus and regions of occipital and parieto-occipital cortices are involved in alpha rhythm generation.⁶¹ Previous research reveals that the modulations of alpha frequency among neurologically intact adults depend on the structural integrity of white matter tracts.⁶² Consistent with this idea, Sheridan et al⁶³ suggested that reduced alpha power observed in the EEG of children exposed to institutional rearing may be the result of delay in the development of cortical white matter (WM). Results from longitudinal studies have demonstrated that normal brain development shows a linear increase in white matter from childhood to adulthood. The WM increment makes the signal transduction become faster and more efficient, allowing increasingly higher frequency contributions to the overall signal. WM maturation is an essential element of brain development and is fundamental for normal function and cognitive maturity.^64,65 Ineffective signal transduction due to decreases in myelination could be one explanation for alpha power reduction.

An important finding was that the sources of increase of beta activity in psychopath traits adolescents, using LORETA were localized on the frontal cortex, the temporal cortex, insula, and the striatum (putamen and globus pallidus). These results support the idea that psychopathic traits could be associated with brain abnormalities in a frontal-temporo-striatal circuit that is involved in emotion, learning, reward processing and motivational aspects of behavior.⁶⁶

LORETA solves the inverse problem by constraining the electrical neuronal activity to fulfill with the property of maximal synchronization, that is, by assuming similar activation among neighboring neuronal sources. A prerequisite to being able to define the inverse solution results within the structural MRI is the co-registration between these 2 imaging modalities, that is, the EEG space and the MRI space.⁴⁴ This step requires that electrode positions are matched to the scalp surface defined by the MRI using some transformation (rotation and translations) operations. These parameters are usually obtained by measuring some “common” landmarks during both the EEG as well as the MRI acquisition. Most commonly, MRI-visible markers are placed on the skin that correspond to the position of the electrodes and/or some fiduciary landmarks (eg, nasion, inion, pre-auricular points and vertex). In order to avoid the labor-intensive measurement of the electrodes and fiduciary landmarks on every subject, many studies use one single MRI template (such as the average brain from the Montreal Neurological Institute) and assume a standard electrode coordinate system. In this case, predefined translation parameters are used to match the EEG to the MRI space. Individual differences in head size and electrode positions are thereby ignored, leading to limited accuracy with regard to the anatomical precision of the source locations. The use of individual anatomical MRI to calculate sources of localization eliminates these limitations and contributes to increasing the accuracy of LORETA localization as has been demonstrated by different research.^67,68 LORETA is a technique that facilitates a deeper understanding of the neurological findings present in the EEG.⁴⁴ In general, LORETA helped to define the structures related to the increase of beta activity as a sign of a dysfunctional (hyperexcitability) neuronal state, probably secondary to CNS injury. To our knowledge, no previous studies of LORETA using QEEG involving psychopath traits adolescent samples have been published.

While using generic minimum-norm methods, it has been shown that deeper sources may not be recovered properly because more superficial dipoles with smaller magnitudes would be privileged for the same EEG/MEG data set. That is the reason why lead field normalization can be operated to give all the sources, those close to the surface and deeper ones, the same opportunity of being nicely reconstructed with a minimum-norm technique.⁶⁹ LORETA combines the lead-field normalization with the Laplacian operator, thus, gives the depth-compensated inverse solution under the constraint of smoothly distributed sources. It is based on the maximum smoothness of the solution.^69,70 The LORETA source estimates work in the entire inner volume of the head. The source space is sampled according to a regular grid of voxel. Each grid point corresponds to a regional source made of 3 orthogonal dipoles. The LORETA priors correspond to a smooth distribution of the source intensities in the entire source space.

LORETA current density findings in the present study and the surface EEG findings are also consistent with MRI studies showing a reduced frontal and temporal lobes, striatal volume^71-73 and reduced cortical surface area in adolescents with CD.⁷⁴ Dysfunction within the striatum and prefrontal cortex had been observed in youths with psychopathic traits causing deficits in the capacity to relationship rewards or punishments with stimuli or responses.³ Moreover, there is evidence that the use of outcome information by the anterior insula, inferior frontal cortex and dorsomedial prefrontal cortex to guide the individual away from suboptimal behavioural choices^75,76 is disrupted in adolescents with psychopathic traits.⁷⁷

Different studies have revealed interesting and reproducible relations between current sources and network connectivity that provide a deeper understanding of the surface EEG dynamics.^36,78 Our finding could be caused by abnormal functional connectivity among the structures related with the increase of beta activity.

Our results support theoretical models of psychopathy that suggest it is a neurodevelopmental disorder.^4,79 A neurodevelopmental perspective of psychopathy is consistent with the fact that the disorder has its roots early in life.⁸⁰ It has been stated that psychopathic traits are present early in life and develop throughout childhood and adolescence.⁸¹ Its course since first manifestations is steady and progressive and is relatively resistant to conventional treatments.^82-84

Our results are consistent with other neurobiological perspectives that emphasize the role of frontal-temporal areas in psychopathy. The prefrontal cortex (PFC) regulates attention, behavior, and emotion through extensive network connections with other brain regions. According with frontal roles, Blair³ emphasizes that one of the core impairments associated with psychopathic traits, is a deficit in decision making and in reinforcement learning, which reflects dysfunction in the ventromedial prefrontal cortex and striatum.

There are 2 other significant models of psychopathy that are supported by findings of deficiencies in the prefrontal cortex: the somatic marker hypothesis⁸⁵ and the prefrontal dysfunction theory.⁸⁶ Damasio hypothesis⁸⁵ suggests that a psychopathic individual may have limited ability to consider emotions when making a decision. This impairment causes self-defeating behavior and violation of social norms with detrimental consequences, all behaviors indicative of the psychopathic profile. Gorenstein and Newman⁸⁶ reported significant correlations between human disinhibition and lesions in the prefrontal cortex. Prefrontal areas of the adolescent brain may be less efficient in generating inhibitory responses and the activation in these frontal areas may be linked to “disinhibition.”⁸⁶

Consistent with other research, our findings suggest poor decision making and emotional dysregulation associated with psychopathic traits may be due to, at least in part, functional anomalies in frontal and temporal regions whereas striatal structural variations may contribute to sensation-seeking and reward-driven behavior in psychopathic traits individuals.¹² The temporal lobe is involved in theory of mind and its dysfunction is often implicated in violent psychopathy. The anterior/middle temporal gyrus has a secondary role, being recruited during working memory and general cognitive processing. The superior temporal sulcus evaluates the intentionality and social complexity of moral actions.⁸⁷ The insula has been implicated in integrating interoceptive states into conscious feelings, decision making, and empathy and pain recognition in others^88-90 and thus may be important in understanding decision making deficits seen in adolescent with psychopathic traits. Our research confirms that CNS functional impairment of psychopathy concern to more than one brain region.

Several neuroimaging studies in adolescents with psychopathic traits have revealed reduction of gray matter in the prefrontal cortex, insula, striatum,^71-73 paralimbic regions, including orbitofrontal cortex, bilateral temporal poles and posterior cingulate cortex^16,17 and abnormal brain activation in limbic regions,^91-93 as well as in the prefrontal and temporal lobes.⁹⁴ A recent study employing diffusion tensor imaging found that adolescents with conduct disorder showed reduced fractional anisotropy of the uncinate fasciculus pathway connecting the amygdala and orbitofrontal cortex.⁹⁵

Summarizing, the results of this study support the hypothesis that fronto-temporal-striatal brain functions are impaired in psychopathic trait adolescents. QEEG and LORETA results indicate that adolescents with CD do not constitute a homogeneous group in terms of EEG profile. Both types of EEG assessment could be important tools for the identification of functional abnormalities of CNS that can to be evaluated as possible biomarkers into psychopathic traits.

Some important limitations of the present report are (a) in the current study only EEG visual inspection was used to remove artifacts. However, in some EEGs, the impact of biological artefacts may still represents a significant problem, the elimination the contaminated EEG segments can bear to the loss of information and viability for clinical diagnosis. Future studies should use automated artefact rejection procedure, as independent component analysis (ICA),^96-98 to remove residual artifacts not detected by visual inspection. (b) Some studies using EEG measurements have produced laterality effect; however, the majority of them have not evaluated the resting EEG or analyzed EEG frequency bands separately.^99-101 Unfortunately, in this study, handedness was not controlled. It would be advisable for future research to include the evaluation of handedness preference and its possible connection with the presence of lateralized abnormalities in CD teenager subgroups.

Footnotes

Acknowledgements

The authors gratefully acknowledge two anonymous referees for their constructive comments and suggestions that were very helpful in improving the manuscript.

Author Contributions

Ana Calzada-Reyes contributed to conception and design of the study, adquisition, analysis and interpretation of the data, drafted the manuscript. Alfredo Alvarez-Amador contributed to conception and design of the study, analysis and interpretation of the data, revised the manuscrip, gave final approval. LídiceGalán-García contributed to design of the study, analysis of the data, revised the manuscript. Mitchell Valdés-Sosa contributed to conception and design of the study, interpretation of the data, revised the manuscript.

Declaration of Conflicting Interests

The author(s) declared no conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Garland

Hough

McCabe

Yeh

Wood

Aarons

. Prevalence of psychiatric disorders in youths across five sectors of care. J Am Acad Child Adolesc Psychiatry. 2001;40:409-418.

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Association; 2013.

Blair

. The neurobiology of psychopathic traits in youths. Nat Rev Neurosci. 2013;14:786-799.

Blair

Newman

Mitchell

. Differentiating among prefrontal substrates in psychopathy: neuropsychological test findings. Neuropsychology. 2006;20:153-165.

Finger

Marsh

Mitchell

. Abnormal ventromedial prefrontal cortex function in children with psychopathic traits during reversal learning. Arch Gen Psychiatry. 2008;65:586-594.

Marsh

Finger

Mitchell

. Reduced amygdala response to fearful expressions in children and adolescents with callous-unemotional traits and disruptive behaviour disorders. Am J Psychiatry. 2008;165:712-720.

Jones

Laurens

Herba

Barker

Viding

. Amygdala hypoactivity to fearful faces in boys with conduct problems and callous-unemotional traits. Am J Psychiatry. 2009;166:95-102.

Marsh

Finger

Fowler

. Empathic responsiveness in amygdala and anterior cingulate cortex in youths with psychopathic traits. J Child Psychol Psychiatry. 2013;54:900-910.

Kruesi

Casanova

Mannheim

Johnson-Bilder

. Reduced temporal lobe volume in early onset conduct disorder. Psychiatry Res. 2004;132:1-11.

10.

Huebner

Vloet

Marx

. Morphometric brain abnormalities in boys with conduct disorder. J Am Acad Child Adolesc Psychiatry. 2008;47:540-547.

11.

Sterzer

Stadler

Poustka

Kleinschmidt

. A structural neural deficit in adolescents with conduct disorder and its association with lack of empathy. Neuroimage. 2007;37:335-342.

12.

Yang

Narr

Baker

. Frontal and striatal alterations associated with psychopathic traits in adolescents. Psychiatry Res. 2015;231:333-340.

13.

De Brito

Mechelli

Wilke

Laurens

Jones

Barker

. Size matters: increased grey matter in boys with conduct problems and callous unemotional traits. Brain. 2009;132:843-852.

14.

De Brito

McCrory

Mechelli

. Small, but not perfectly formed: decreased white matter concentration in boys with psychopathic tendencies. Mol Psychiatry. 2011;16:476-477.

15.

Fahim

Yoon

Chen

Evans

Perusse

. Neuroanatomy of childhood disruptive behaviour disorders. Aggress Behav. 2011;37:326-337.

16.

Ermer

Cope

Nyalakanti

Calhoun

Kiehl

. Aberrant paralimbic gray matter in incarcerated male adolescents with psychopathic traits. J Am Acad Child Adolesc Psychiatry. 2013;52:94-103.

17.

Cope

Ermer

Nyalakanti

Calhoun

Kiehl

. Paralimbic gray matter reductions in incarcerated adolescent females with psychopathic traits. J Abnorm Child Psychol. 2014;42:659-668.

18.

Lynam

. Early identification of chronic offenders: who is the fledgling psychopath? Psychol Bull. 1996;120:209-234.

19.

Lynam

. Pursuing the psychopath: capturing the fledgling psychopath in a nomological net. J Abnorm Psychol. 1997;106:425-438.

20.

Forth

Mailloux

. Psychopathy in youth: what do we know? In: Gacono

, ed. The Clinical and Forensic Assessment of Psychopathy: A Practitioner’s Guide. Mahwah, NJ: Lawrence Erlbaum; 2000:25-54.

21.

Frick

Bodin

Barry

. Psychopathic traits and conduct problems in community and clinic-referred samples of children: further development of the psychopathy screening device. Psychol Assess. 2000;12:382-393.

22.

Frick

. Conduct Disorders and Severe Antisocial Behaviour. New York, NY: Plenum; 1998.

23.

Volavka

. Aggression, electroencephalography, and evoked potentials: a critical review. Neuropsychiatry Neuropsychol Behav Neurol. 1990;3:249-259.

24.

Coble

Taska

Kupfer

Kazdin

Unis

French

. EEG sleep “abnormalities” in preadolescent boys with a diagnosis of conduct disorder. J Am Acad Child Psychiatry. 1984;23:438-447.

25.

Lewis

Pincus

Shanok

Glaser

. Psychomotor epilepsy and violence in a group of incarcerated adolescent boys. Am J Psychiatry. 1982;139:882-887.

26.

Mednick

Vka

Gabrielli

Jr Itil

. EEG as a predictor of antisocial behavior. Criminology. 1981;19:219-230.

27.

Petersén

K.G.I.

Matousek

Mednick

S.A.

Volavka

Pollock

. EEG antecedents of thievery. Acta Psychiatric Scand. 1982; 65: 331–338

28.

Iacono

Carlson

Malone

McGue

. P3 event-related potential amplitude and the risk for disinhibitory disorders in adolescent boys. Arch Gen Psychiatry. 2002;59:750-757.

29.

Porjesz

Rangaswamy

Kamarajan

Jones

Padmanabhapillai

Begleiter

. The utility of neurophysiological markers in the study of alcoholism. Clin Neurophysiol. 2005;116:993-1018.

30.

Calzada-Reyes

Alvarez-Amador

Galán-García

Valdés-Sosa

. EEG abnormalities in psychopath and non-psychopath violent offenders. J Forensic Leg Med. 2013;20:19-26.

31.

Houston

Stanford

. Electrophysiological substrates of impulsiveness: potential effects on aggressive behaviour. Prog Neuropsychopharmacol Biol Psychiatry. 2005;29:305-313.

32.

Koelsch

Sammler

Jentschke

Siebel

. EEG correlates of moderate intermittent explosive disorder. Clin Neurophysiol. 2008;119:151-162.

33.

Clarke

Barry

McCarthy

Selikowitz

. EEG-defined subtypes of children with attention deficit/hyperactivity disorder. Clin Neurophysiol. 2001;112:2098-2105.

34.

Clarke

Barry

McCarthy

Selikowitz

. Excess beta activity in children with attention deficit/hyperactivity disorder: an atypical electrophysiological group. Psychiatry Res. 2001;103:205-218.

35.

Meier

Perrig

Koenig

. Is excessive electroencephalography beta activity associated with delinquent behavior in men with attention-deficit hyperactivity disorder symptomatology. Neuropsychobiology. 2014;70:210-219.

36.

Thatcher

Biver

North

. Spatial-temporal current source. Correlations and cortical connectivity. Clin EEG Neurosci. 2007;38:35-48.

37.

Michel

Murray

. Towards the utilization of EEG as a brain imaging tool. Neuroimage. 2012;61:371-385.

38.

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed. Text Revision. Washington, DC: American Psychiatric Association; 2000.

39.

Frick

Hare

. The Antisocial Process Screening Device: Technical Manual. Toronto, Ontario, Canada: Multi-Health Systems; 2001.

40.

Hare

. The Psychopathy Checklist–Revised. Toronto, Ontario, Canada: Multi-Health Systems; 1991.

41.

Frick

Bodin

Barry

. Psychopathic traits and conduct problems in community and clinic-referred samples of children: further development of the Psychopathy Screening Device. Psychol Assess. 2000;12:382-393.

42.

John

Karmel

Coming

. Neurometrics: numerical taxonomy identifies different profiles of brain dysfunctions within groups of behaviourally similar people. Science. 1977;196:1393-1410.

43.

Ferrero

. Análisis computado del EEG. 1ra. ed. Buenos Aires, Argentian: Las Heras; 1995.

44.

Pascual-Marqui

Michel

Lehmann

. Low resolution electromagnetic tomography: a new method for localizing electrical activity in the brain. Int J Psychophysiol. 1994;18:49-65.

45.

Riera

Fuentes

Valdés

Oharriz

. Theoretical basis of the EEG spline inverse solutions for a spherical head model. Biomed Tech. 1997;42:219-222.

46.

Riera

Valdés

Fuentes

Oharris

. Explicit Backus and Gilbert EEG inverse solution for a spherical head model. Biomed Tech. 1997;42:216-218.

47.

Evans

Collins

Mills

Brown

Kelly

Peters

. 3D statistical neuroanatomical models from 305 MRI volumes. In: Proceedings of the IEEE Nuclear Science Symposium and Medical Imaging Conference. London, England: MTP Press; 1993:1813-1817.

48.

Szava

Valdés

Biscay

Bosch

Száva

Virués

. High resolution quantitative EEG analysis. Brain Topogr. 1994;6:211-219.

49.

Raz

. Analysis of repeated measurements using non-parametric smoothers and randomization tests. Biometrics. 1989;45:851-871.

50.

Blair

Karniski

. An alternative method for significance testing of waveform difference potential. Psychophysiology. 1993;30:518-524.

51.

Blair

Karniski

. Distribution-free statistical analyses of surface and volumetric maps. In: Thatcher

Hallett

Zeffiro

John

Huerta

, eds. Functional Neuroimaging. New York, NY: Academic Press; 1994:19-28.

52.

Galán

Biscay

Rodríguez

Pérez Abalo

Rodríguez

. Testing topographic differences between events related brain potentials by using nonparametric combinations of permutation test. Electroencephalogr Clin Neurophysiol. 1997;102:240-247.

53.

Pesarin

Salmaso

. Permutation Tests for Complex Data: Theory, Applications and Software. Hoboken, NJ: Wiley; 2010.

54.

Bennett

Korein

Kalmijn

Grega

Campbell

. Electroencephalogram and treatment of hospitalized aggressive children with haloperidol or lithium. Biol Psychiatry. 1983;18:1427-1440.

55.

Selin

Gottschalk

. Schizophrenia, conduct disorder and depressive disorder: neuropsychological, speech sample and EEC results. Percept Mot Skills. 1983;57:427-444.

56.

Liu

. EEG studies in female adolescent delinquents. Quantitative and visual analysis. Zhonghua Shen Jing Jing Shen Ke Za Zhi. 1991;24:379-381.

57.

Gilmore

Malone

Iacono

. Brain electrophysiological endophenotypes for externalizing psychopathology: a multivariate approach. Behav Genet. 2010;40:186-200.

58.

Matsuura

Okubo

Toru

. A cross-national EEG study of children with emotional and behavioral problems: a WHO collaborative study in the Western Pacific Region. Biol Psychiatry. 1993;34:59-65.

59.

Bauer

Hesselbrock

. Brain maturation and subtypes of conduct disorder: interactive effects on P00 amplitude and topography in male adolescents. J Am Acad Child Adolesc Psychiatry. 2003;42:106-115.

60.

Gabrielli

Mednick

. Intellectual performance in children of alcoholics. J Nerv Ment Dis. 1983;171:444-447.

61.

Éismont

Pritchenko

Pavlenko

. Peculiarities of the pattern of EEG activity in institutionalized children. Neurophysiology. 2014;46:415-421.

62.

Steriade

. Grouping of brain rhythms in corticothalamic systems. Neuroscience. 2006;137:1087-1106.

63.

Valdés-Hernández

Ojeda-González

Martínez-Montes

. White matter architecture rather than cortical surface area correlates with the EEG alpha rhythm. Neuroimage. 2010;49:2328-2339.

64.

Sheridan

M A

Fox

Zeanah

McLaughlin

Nelson

. Variation in neural development as a result of exposure to institutionalization early in childhood. Proc Natl Acad Sci U S A. 2012;109:12927-12932.

65.

Luna

Thulborn

Munoz

. Maturation of widely distributed brain function subserves cognitive development. Neuroimage. 2001;13:786-793.

66.

Paus

Zijdenbos

Worsley

. Structural maturation of neural pathways in children and adolescents: in vivo study. Science. 1999;283:1908-1911.

67.

O’Doherty

Dayan

Schultz

Deichmann

Friston

Dolan

. Dissociable roles of ventral and dorsal striatum in instrumental conditioning. Science. 2004;304:452-454.

68.

Guggisberg

Dalal

Zumer

. Localization of cortico-peripheral coherence with electroencephalography. Neuroimage. 2011;57:1348-1357.

69.

Michel

Murray

Lantz

Gonzalez

Spinelli

Grave

Peralta

. EEG source imaging. Clin Neurophysiol. 2004;115:2195-2222.

70.

Pascual-Marqui

. Review of methods for solving the EEG inverse problem. Int J Bioelectromagn. 1999;1:75-86.

71.

Baillet

. Toward Functional Brain Imaging of Cortical Electrophysiology Markovian Models for Magneto and Electroencephalogram Source Estimation and Experimental Assessments [dissertation]. Orsay, France: University of Paris XI; 1998.

72.

Fairchild

Passamonti

Hurford

. Brain structure abnormalities in early-onset and adolescent-onset conduct disorder. Am J Psychiatry. 2011;168:624-633.

73.

Fairchild

Hagan

Walsh

Passamonti

Calder

Goodyer

. Brain structure abnormalities in adolescent girls with conduct disorder. J Child Psychol Psychiatry. 2013;54:86-95.

74.

Wallace

White

Robustelli

. Cortical and subcortical abnormalities in youths with conduct disorder and elevated callous-unemotional traits. J Am Acad Child Adolesc Psychiatry. 2014;53:456-465.

75.

Sarkar

Craig

Dell’Acqua

. Prenatal stress and limbic-prefrontal white matter microstructure in children aged 6-9 years: a preliminary diffusion tensor imaging study. World J Biol Psychiatry. 2014;15:346-352.

76.

Budhani

Marsh

Pine

Blair

. Neural correlates of response reversal: considering acquisition. Neuroimage. 2007; 34:1754-1765.

77.

Kuhnen

Knutson

. The neural basis of financial risk-taking. Neuron. 2005;47:763-770.

78.

Blair

. The amygdala and ventromedial prefrontal cortex: functional contributions and dysfunction in psychopathy. Phil Trans R Soc B. 2008;363:2557-2565.

79.

Pascual-Marqui

Koukkou

Lehmann

Kochi

. Functional localization and functional connectivity with LORETA comparison of normal controls and first episode drug naïve schizophrenics. J Neurotherapy. 2001;4(4):35-37.

80.

Gao

Glenn

Schug

Yang

Raine

. The neurobiology of psychopathy: a neurodevelopmental perspective. Can J Psychiatry. 2009;54:813-823.

81.

Glenn

Raine

Venables

Mednick

. Early temperamental and psychophysiological precursors of adult psychopathic personality. J Abnorm Psychol. 2007;116:508-518.

82.

Lynam

Derefinko

. Psychopathy and personality. In Patrick

, ed. Handbook of Psychopathy. New York, NY: Guilford Press; 2006:133-155.

83.

Blonigen

Carlson

Krueger

Patrick

. A twin study of self-reported psychopathic personality traits. Pers Individ Diff. 2003;35:179-197.

84.

Taylor

Loney

Bobadilla

Iacono

McGue

. Genetic and environmental influences on psychopathy trait dimensions in a community sample of male twins. J Abnorm Child Psychol. 2003;31:633-645.

85.

Damasio

. The somatic marker hypothesis and the possible functions of the prefrontal cortex. Philos Trans R Soc Lond B Biol Sci. 1996;351:1413-1420.

86.

Gorenstein

Newman

. Disinhibitory psychopathology: a new perspective and a model for research. Psychol Res. 1980;87:301-315.

87.

McNamee

Dunfee

Luna

Clark

Eddy

Tarter

. Brain activation, response inhibition, and increased risk for substance use disorder. Alcohol Clin Exp Res. 2008;32:405-413.

88.

Fumagalli

Priori

. Functional and clinical neuroanatomy of morality. Brain. 2012;135:2006-2021.

89.

Craig

Catani

Deeley

. Altered connections on the road to psychopathy. Mol Psychiatry. 2009;14:946-953.

90.

Decety

Jackson

. A social-neuroscience perspective on empathy. Curr Dir Psychol Sci. 2006;15:54-58.

91.

Naqvi

Bechara

. The hidden island of addiction: the insula. Trends Neurosci. 2009;32:56-67.

92.

de Wied

Gispen-de Wied

van Boxtel

. Empathy dysfunction in children and adolescents with disruptive behaviour disorders. Eur J Pharmacol. 2010;626:97-103.

93.

White

Pope

Sinclair

. Disrupted expected value and prediction error signaling in youth with disruptive behavior disorders during a passive avoidance task. Am J Psychiatry. 2013;170:315-323.

94.

Finger

Marsh

Mitchell

. Abnormal ventromedial prefrontal cortex function in children with psychopathic traits during reversal learning. Arch Gen Psychiatry. 2008;65:586-594.

95.

Passamonti

Fairchild

Fornito

. Abnormal anatomical connectivity between the amygdala and orbitofrontal cortex in conduct disorder. PLoS One. 2012;7(11):e48789.

96.

Bell

Sejnowski

. An information-maximization approach to blind separation and blind deconvolution. Neural Comput. 1995;7:1129-1159.

97.

Makeig

Bell

Jung

Sejnowski

. Independent component analysis of electroencephalographic data. Adv Neural Inform Process Syst. 1996;8:145-151.

98.

Jung

Makeig

McKeown

Bell

Lee

Sejnowski

. Imaging brain dynamics using independent component analysis. Proc IEEE. 2001;89:1107-1122.

99.

Davidson

Chapman

Henriques

. Asymmetrical brain electrical activity discriminates between psychometrically-matched verbal and spatial cognitive tasks. Psychophysiology. 1990;27:528-543.

100.

Alexander

Sufka

. Cerebral lateralization in homosexual males: a preliminary EEG investigation. Int J Psychophysiol. 1993;15:269-274.

101.

Çiçek

Nalçacı

. Interhemispheric asymmetry of EEG alpha activity at rest and during the Wisconsin Card Sorting Test: relations with performance. Biol Psychol. 2001;58:75-88.