- Email: [email protected]
Local contextual processing in major depressive disorder
Clinical Neurophysiology 125 (2014) 476–483
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Local contextual processing in major depressive disorder Noa Fogelson a,⇑, Avi Peled b,d, Sarah Marmor c,d, Miguel Fernandez-del-Olmo e, Ehud Klein c,d a
Department of Psychology, University of A Coruña, La Coruña, Spain Institute for Psychiatric Studies, Sha’ar Menashe Mental Health Center, Hadera, Israel c Department of Psychiatry, Rambam Medical Center, Technion, Haifa, Israel d B Rappaport Faculty of Medicine, Technion, Haifa, Israel e Department of Physical Education, University of A Coruña, La Coruña, Spain b
a r t i c l e
i n f o
Article history: Accepted 4 September 2013 Available online 26 September 2013 Keywords: Context Major depression EEG P3b N1
h i g h l i g h t s Reaction times and P3b latencies of predicted targets were prolonged in depression. P3b amplitudes were attenuated in major depressive disorder during processing of predictive
contextual information, as well as target N1 amplitudes. Processing of local contextual processing is altered in major depressive disorder.
a b s t r a c t Objective: The study investigated local contextual processing in patients with major depressive disorder (MDD). This was defined as the ability to utilize predictive contextual information to facilitate detection of predictable versus random targets. Method: We recorded EEG in 15 MDD patients and 14 age-matched controls. Recording blocks consisted of targets preceded by randomized sequences of standards and by sequences of standards that included a predictive sequence signaling the occurrence of a subsequent target event. Results: Both MDD patients and age-matched controls demonstrated a significant reaction time (RT) and P3b latency differences between predicted and random targets. However, patients demonstrated a specific prolongation of these measures during processing of predicted targets, as well as an attenuation of P3b amplitudes for the predictive sequence. In addition, patients target N1 amplitudes were attenuated compared with controls. Conclusion: MDD patients were able to utilize predictive context in order to facilitate processing of deterministic targets, however, this ability was limited compared to controls, as demonstrated by contextdependent P3b deficits. Significance: These findings suggest that patients with major depression have altered processing of local contextual processing. Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Contextual processing is considered to be an executive function and a subcomponent of working memory (Cohen and ServanSchreiber, 1992; Barch et al., 2001). Processing of goal-directed contextual information is essential for the selection of appropriate task behavior and facilitates the detection of task-relevant stimuli (Fogelson et al., 2009b). Major depressive disorder (MDD) is the most prevalent of all psychiatric disorders and is characterized by a wide range of ⇑ Corresponding author. Address: Department of Psychology, Campus de Elviña, La Coruña 15071, Spain. Tel.: +34 981167000x1785; fax: +34 981105641. E-mail address: [email protected] (N. Fogelson).
behavioral, emotional and cognitive symptoms (reviewed by Gotlib and Joormann, 2010; Marazziti et al., 2010; Murrough et al., 2011). Performance on measures of executive function tend to be impaired in depressed patients, and working memory deficits are common (Pelosi et al., 2000; Rose and Ebmeier, 2006; Segrave et al., 2010; Marazziti et al., 2010; Sato et al., 2011) and associated with impairments in attentional control (Pelosi et al., 2000; Rose and Ebmeier, 2006; Segrave et al., 2010; Gotlib and Joormann, 2010; Murrough et al., 2011). However, other studies have demonstrated that there are no impairments in working memory functions in patients with depression (Barch et al., 2003; Holmes et al., 2005). One measure that has been linked to contextual processing is the event related potential (ERP) called the P300 (Squires et al.,
1388-2457/$36.00 Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clinph.2013.09.001
477
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
1976; Donchin and Coles, 1988). The target P300, known as the P3b, is elicited, among other tasks, by targets in the classical oddball paradigm and has a posterior-parietal scalp distribution (Squires et al., 1975). P3b is thought to be a measure of the evaluation of environmental signals including contextual information (Squires et al., 1976; Donchin and Coles 1988) and is also thought to reflect monitoring processes mediating perceptual analysis and response initiation (Verleger et al. 2005). Findings of P300 abnormalities in patients with depression are inconsistent. Although there is a general trend for patients with depression to show P3b amplitude reductions (Roth et al., 1981; Pfefferbaum et al., 1984; Vandoolaeghe et al., 1998; Urretavizcaya et al., 2003; Campanella et al., 2012), these findings depend on stimulus modality (auditory versus visual), the complexity of the paradigm (oddball versus more cognitively demanding tasks) and on the subtype and severity of the symptoms (e.g., reviewed in Bruder et al., 2012). Studies utilizing more cognitively demanding tasks tend to show more consistent reductions in P3b amplitudes in patients with depression (Bruder et al., 2012). In the present study, we focus on the ability to process goal-relevant local contextual information. Previous studies have shown that P3b is modulated as a function of local predictive context (Fogelson et al., 2009a,b, 2010). In these studies, targets were preceded by either randomized sequences of standards or by sequences including a predictive sequence, signaling the occurrence of a subsequent target event. Two main neural correlates of contextual processing were identified. First, a facilitated processing of predicted targets was observed, as indicated by faster P3b latencies and reaction times for predicted targets compared with random ones. Second, local contextual processing was associated with the generation of a robust P3b to the final most-informative stimulus of the predicting sequence. Local contextual processing has been shown to be altered by aging (Fogelson et al., 2010) and impaired in patients with lateral prefrontal cortex lesions (Fogelson et al., 2009a), Parkinson’s disease (Fogelson et al., 2011a) and schizophrenia (Fogelson et al., 2011b), demonstrating specific alterations in the behavioral measures and neural correlates of local contextual processing in each patient population. It remains inconclusive whether patients with depression have working memory impairments. Thus, the objective of the current study was to investigate whether patients with major depressive disorder are impaired in their ability to process contextual information, an important subcomponent of working memory, utilizing the previously described paradigm (Fogelson et al., 2009a, 2010, 2011a,b). We evaluated neural correlates of contextual processing based on both behavioral and ERP measures to determine whether
a more cognitively demanding version of the oddball task will induce changes in performance in patients compared with controls. Our hypothesis was that if patients with major depression do have contextual processing deficits, we should observe alterations in the behavioral and electrophysiological indices of local contextual processing. 2. Methods 2.1. Participants 15 MDD patients (mean age ± standard error of the mean = 43 ± 4 years, 9 females) and 14 age-matched controls (mean age = 43 ± 4 years, 8 females) participated in the study. Patients were diagnosed with MDD according to DMS-IV-Tr criteria, and were rated for symptom severity using the Hamilton Rating Scale for Primary Depressive Illness (HRSD, Hamilton, 1960). Subjects with past history of neurologic disorders, drug or alcohol abuse were excluded. Mean illness duration was 3 ± 1 years. Mean HRSD scores were 30.2 ± 2.5. Demographics and clinical details of the patients are shown in Table 1. All patients had normal or corrected-to-normal visual acuity. All patients were hospitalized at the time of the experiment due to an episode of major depression and took their regular medications on the day the recordings. Most patients were medicated with a selective serotonin reuptake inhibitor (SSRI), some patients were treated with a mood stabilizer and benzodiazepine. Medication dosages were within regular therapeutic levels, achieving average therapeutic dosage based on blood levels. Patients were matched by controls for age, gender and education (mean = 14.6 ± .5 and 14.4 ± 1 years of education for MDD and controls, respectively). Age-matched controls had normal or corrected-to-normal visual acuity and had no history of psychiatric or neurological problems. All the subjects were right handed as indicated by self-report. The experimental procedures were approved by the local ethics committees. Written informed consent was obtained from all subjects participating in the study following a complete explanation of the study and procedures. 2.2. Procedure Subjects were seated 110 cm in-front of a 21-inch PC-computer screen. Stimuli were presented to either the left or right visual field 6° from a central point of fixation. The stimuli consisted of black triangles on a gray background. Subjects were asked to centrally
Table 1 Clinical details of patients with major depression. Patient (sex)
Diagnosis
HRSD
Medication
1 (F) 2 (F) 3 (F)
1 2 5
Major depression Major depression Major depression
21 26 38
4 (F) 5 (F) 6 (F)
1 1 1
Major depression Major depression Major depression + borderline personality disorder Depression + anxiety Major depression Major depression + anorexia Major depression Major depression Major depression Major depression Major depression Major depression
18 27 19
Mirtazapine Lithium Venalafaxine carbamazepine clonazepam risperidone Escitalopram sertraline qutiapin clonazepam Paroxetine
26 17 27 44 33 34 43 45 35
Lithium Qutiapin lithium mirtazapine Venalafaxine mirtazapine Fluoxetine Escitalopram Escitalopram Fluoxetine Fluvoxamine zolpidem lithium Escitalopram
7 (F) 8 (M) 9 (M) 10 (F) 11 (M) 12 (M) 13 (M) 14 (F) 15 (M)
Disease duration (years)
3 5 1 10 10 7 0.5 1 1.5
HRSD, Hamilton rating scale for primary depressive illness (Hamilton, 1960).
478
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
2.4. Behavioral analysis Accuracy was defined as the percentage of targets for which a button press was detected. Reaction times were calculated by averaging correct trials for predicted and random targets in each subject. Misses (no button press 150–1150 ms post-stimulus onset) were excluded from reaction time analysis. Reaction times were analyzed using E-prime (Psychology Software Tools, Inc., Pittsburgh, USA). 2.5. ERP analysis Fig. 1. Task timeline. Sequences of standards S1, S2 and S3 with a predicted sequence (top) and in randomized order (bottom) preceding the target (T). The predictive sequence is always S1 followed by S2 and then S3 (n 1). Stimuli presented to the left or right visual field. Inter-trial intervals, including duration of stimulus presentation (150 ms) are displayed.
fixate throughout the recording. Stimuli consisted of 15% targets (downward facing triangle) and 85% of equal amounts of three types of standards (triangles facing left, upwards and right, at 90° increments). In each block a total of 78 stimuli (12 targets, 22 of each standard type) were presented each for 150 ms and interstimulus interval (ISI) of 1 s. Recording blocks consisted of targets preceded by either randomized sequences of standards or by sequences including a three-standard predictive sequence. The predictive sequence always consisted of the three standards of triangles facing left, up and right, always in that order. Fig. 1 illustrates an example of a target preceded by a randomized sequence of standards and a target preceded by the predictive sequence of standards. Each block consisted of 6 different randomized sequences of standards (3–8 standards long) preceding the target; and 6 sequences of standards (3–8 standards long) with a predictive sequence preceding the target in each. Each recording session consisted of 14 different blocks, displayed in randomized order, each approximately 1.6 min long. A single sequence of trials appeared on the right or left hemifield in a randomized order within each block. Blocks were counterbalanced so that there were equal amount of stimuli presented to the right and left visual hemi-field across the blocks. Within a single sequence of trials (predictive or random) all stimuli were presented in one hemifield. The predictive sequence was always followed by a target. Subjects performed a brief training session to ensure they were able to detect the target accurately. Subjects were then shown the predictive sequence and were told that it would be 100% predictive of a target, but that targets would also appear randomly throughout the block. Subjects were asked to press a button with their right index finger each time a target was presented and to pay attention and look for the predictive sequence. Subjects then performed another brief training session to ensure that they were confident in the detection of the predictive sequence as well as the targets, before the recording session began. Stimulus presentation and response recordings were controlled using E-prime (Psychology Software Tools, Inc., Pittsburgh, USA).
2.3. Data acquisition EEG was recorded from a 64 electrode array using the ActiveTwo system (Biosemi, The Netherlands). External electrodes above and below the right eye monitored vertical eye movements and electrodes placed laterally to the left and right eyes monitored horizontal eye movements. Signals were amplified and digitized at 512 Hz. All channels were re-referenced to averaged linked earlobes.
Post processing and ERP analysis of the data was performed using Brain Vision Analyzer version 1.05 (Brain Products GmbH, Germany). Prior to ERP analysis eye saccades and blinks were defined using ICA (64 EEG electrodes were included), and the identified components were removed using the linear derivation function in Brain Vision Analyzer. Epochs containing misses (no button press 150–1150 ms post-stimulus onset) were excluded from further analysis. EEG signals were filtered at 0.1–30 Hz, using Butterworth zero phase filters, for subsequent analysis. EEG signals were sorted and averaged relative to the stimulus onset, with epochs set from 200 to 1000 ms relative to stimulus onset. A baseline correction of 200 ms pre-stimulus onset ( 200 to 0 ms) was performed. The maximal and minimal permissible amplitudes at any electrode for EEG epochs that were included in the analysis were 75 and 75 lV, respectively. 2.6. P3b P3b was determined as the most positive point in the latency range of 300–700 ms. In order to restrict the number of comparisons and since no lateralization effects of hemisphere or visual field were observed, all P3b ERP data were collapsed across visual fields. In addition we concentrated on midline electrode sites (AFz, Fz, FCz, Cz, CPz, and Pz) to explore anterior versus posterior topographical differences of P3b for the different target conditions. Peak P3b amplitude (measured in lV) at AFz, Fz, FCz, Cz, CPz, and Pz were evaluated for 4 conditions: targets after predictive sequences (predicted), targets after non predictive random sequences (random), random preceding standards (standards excluding those comprising of the predicting sequence) and the last most-informative standard of the predicting sequence (n 1) for both groups. There were comparable number of trials in patients and controls for predicted (65 ± 4, MDD and 62 ± 4, controls), random (63 ± 4, MDD and 65 ± 3, controls), n 1 (68 ± 4, MDD and 62 ± 3, controls) and standard (542 ± 37, MDD and 529 ± 25, controls) conditions after removal of misses and artifacts. Peak P3b latencies (measured in ms) were evaluated at at AFz, Fz, FCz, Cz, CPz, and Pz for predicted and random targets. 2.7. N1 To assess the early perceptual processes between the two target conditions, peak N1 amplitudes (measured in lV) were determined at PO7 and PO8, for both predicted and random targets presented to the right or left visual fields. N1 was determined as the most negative peak in the latency range of 50–200 ms. 2.8. N2 To assess N2 amplitude differences between the two target conditions, peak N2 amplitudes (measured in lV) were determined at electrode sites AFz, Fz, FCz, Cz, CPz, and Pz, for predicted and random targets. N2 was defined as the largest negative peak in
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
the time window starting with the P2 peak and ending 350 ms after target onset. N2 amplitudes were measured against the amplitude of the preceding P2, which was determined as the largest positive peak from 150 ms after target onset until the N2 peak. 2.9. Statistical analysis Analysis of variance (ANOVA) was performed with the Greenhouse-Geisser correction, followed by post hoc parametric paired t-tests, Sidak corrected for multiple comparisons unless otherwise stated. Partial eta squared (g2p ) and epsilon values are reported where applicable. Pearson’s Product Moment correlation coefficient was used to calculate correlations. Mean values with ± standard error of the mean (SEM) are used throughout the text. 3. Results 3.1. Behavioral results There were no significant differences in accuracy between MDD patients and controls for predicted targets (mean accuracy = 97 ± 2% and 98 ± 1%, for MDD and controls, respectively, t(27) = .11, p = .912) and random targets (mean accuracy = 96 ± 2% and 99 ± .2%, for MDD and controls, respectively, t(27) = 1.41, p = .17). To compare the reaction times (RT) for the targets between the groups we performed an ANOVA with group (MDD patients, controls) as the between-subjects factor and condition (predicted, random targets) as the repeated measures factor. There was a main effect for condition (F(1,27) = 61.32, p < .0001, g2p = .69) and a significant condition group interaction (F(1,27) = 8.47, p = .007, g2p = .24). Post hoc t-tests showed that RTs for predicted targets (mean RT = 312 ± 21 ms and 423 ± 19 ms, for controls and MDD patients, respectively) were shorter than those for random targets (mean RT = 454 ± 17 ms, t(13) = 7.03, p < .0001, and 488 ± 20 ms, t(14) = 3.77, p = .002, for controls and MDD patients, respectively) in both controls and MDD patients. However, RTs for predicted targets were shorter in controls compared with patients (t(27) = 3.95, p = .001), while RTs for random targets were not significantly different between patients and controls (t(27) = 1.28, p = 0.213). Reaction time comparisons are displayed in Fig. 2A. 3.2. P3b Topographical maps and waveforms of the grand-averaged ERPs across the subjects in the control group and across the subjects in the patient group, at electrode site Cz elicited by predicted and random targets, standards and n 1, the last most-informative stimulus of the predicting sequence, are shown in Fig. 3. To compare peak P3b amplitudes we performed an ANOVA with group (MDD patients, controls) as the between-subjects factor, and with condition (predicted, random targets, n 1, and standards) and electrode (AFz, Fz, FCz, Cz, CPz, and Pz) as the repeated measures factors. There were main effects for condition (F(3,81) = 35.68, p < .0001, g2p = .57, epsilon = .55) and electrode (F(5,135) = 23.49, p < .0001, g2p = .47, epsilon = .39). There was a main effect of group (p = .016), demonstrating larger P3b amplitudes in controls compared with patients, and a significant electrode x group interaction (F(5,135) = 3.64, p = .033, g2p = .20). Both controls and patients showed maximal P3b amplitudes at electrode sites CPz and Cz, respectively. Both groups showed minimal P3b amplitudes at electrode AFz. Post-hoc tests, Sidak corrected for multiple comparisons, showed that in both controls and patients the peak P3b amplitude was larger for predicted and random targets compared with standards (p 6 .02), and there were no
479
significant differences in P3b amplitude between random and predicted targets. However, only controls displayed larger P3b amplitudes for the n 1 condition compared with standards (p < .0001), while in the patients this difference was non-significant (p = .161). In addition, independent t-tests corrected for multiple comparisons showed smaller P3b amplitudes for the n 1 condition (at Cz, CPz and Pz) in patients compared to controls (t(27) < 4.1, p < .002). To compare peak P3b latencies we performed an ANOVA with group (MDD patients, controls) as the between-subjects factor and with condition (predicted, random targets) and electrode (AFz, Fz, FCz, Cz, CPz, and Pz) as the repeated measures factors. There was a main effect for condition (F(1,27) = 28.73, p < .0001, g2p = .52) but no main effect for electrode (F(5,135) = .49, p = .597, g2p = .02, epsilon = .36). There was a significant condition x group interaction (F(1,27) = 5.67, p = .025, g2p = .17). Both patients and controls showed peak P3b latency to be shorter for predicted targets compared with random targets (t(14) < 3.3, p 6 .045 and t(13) < 4.6, p 6 .002 for patients and controls, respectively) across the 6 electrode sites. However, independent t-tests revealed P3b latency for predicted targets to be longer in patients compared to controls (at Pz, CPz, Fz and Cz, t(27) < 2.5, p 6 .042), while no significant P3b latency differences were found between the groups for random targets. P3b latency comparisons are displayed in Fig. 2B.
3.3. N1 We utilized an ANOVA with group (MDD patients, controls) as the between-subjects factor and electrode (PO7 and PO8), visual hemifield (left, right) and condition (predicted, random targets) as repeated measures factors, to compare the peak N1 amplitude between predicted and random targets. There were no significant main effects for condition, electrode or side. There was a significant interaction between extrastriate electrode location and visual field of stimuli presentation (F(1,27) = 17.77, p < .0001, g2p = .40), demonstrating that N1 ERPs were enhanced to contralaterally presented stimuli. In addition, there was an electrode visual hemifield group interaction (F(1,27) = 5.79, p = .023, g2p = .18) showing smaller N1 amplitudes to contralaterally presented stimuli in patients with depression compared with controls. Post-hoc comparisons revealed no significant N1 amplitude differences between predicted and random targets across groups. N1 waveforms and topographical maps are demonstrated in Fig. 4.
3.4. N2 To compare peak N2 amplitudes we performed an ANOVA with group (MDD patients, controls) as the between-subjects factor and with condition (predicted, random targets) and electrode (AFz, Fz, FCz, Cz, CPz, and Pz) as the repeated measures factors. There were main effects for condition (F(1,27) = 29.51, p < .0001, g2p = .52) and electrode (F(5,135) = 5.19, p = .007, g2p = .16, epsilon = .43). There was a significant condition group interaction (F(1,27) = 5.35, p = .029, g2p = .16). Post hoc tests corrected for multiple comparisons showed peak N2 amplitude to be maximal at FCz and minimal at AFz across the groups. However, the independent t-tests for each target condition (predicted and random targets) showed no significant N2 amplitude differences between the groups across the 6 electrode sites. Both controls (all 6 electrodes, t(13) < 6.5, p 6 .003) and patients (at Pz, CPz, FCz, Cz, t(14) < 2.8, p 6 .04) showed N2 amplitude to be larger for random targets compared with predicted targets.
480
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
Fig. 2. Reaction times (A) and peak P3b latency at Cz (B) for predicted and random targets in patients with depression and age-matched controls, demonstrating specific prolongation of these measures in the predicted targets condition in patients compared with controls. Bars = SEM.
3.5. Correlations
4. Discussion
In order to determine the possible effects of clinical symptoms on the behavioral and electrophysiological measures associated with local contextual processing in the MDD patients, we calculated correlations between daily HRSD scores and illness duration and the following measures: RT for predicted and random targets, peak P3b amplitude and latency for predicted and random targets, peak P3b latency shift, n 1 peak P3b amplitude (all at Cz) and N1 amplitude for predicted and random targets (at PO7), across the 15 patients. There were no significant correlations between HRSD scores and illness duration with any of the behavioral or electrophysiological measures. In order to determine the association between behavioral measures and the electrophysiological findings, reaction times for predicted and random targets were correlated with P3b amplitudes for the predicted sequence, P3b latencies and N1 amplitudes for the target conditions, in MDD and control subjects. RTs for predicted targets were significantly correlated with peak P3b latency for predicted targets in MDD patients (r = .581, p = .023) and in controls (r = .693, p = .006). Both groups showed no significant correlations between RT for random targets and any of the electrophysiological measures. No correlations were observed between N1 amplitudes for target conditions and any of the other variables.
The current study provides electrophysiological evidence that processing of local contextual information is altered in patients with major depression. This was demonstrated by the following findings. First, although both patients and controls showed a RT difference and P3b latency shift between predicted and non-predictive targets, there was a specific prolongation of these measures for predicted targets in the patients. Second, patients demonstrated an attenuation of peak P3b to the last most-informative stimulus (n 1) of the predictive sequence compared with controls. Finally, there was an attenuation of N1 amplitudes for target conditions in the patients.
4.1. Depression and local contextual processing effects on ERPs In controls we replicated previous findings (Fogelson et al., 2009a, b, 2010, 2011a,b) showing that P3b amplitude increased with task-informative stimuli, so that a significant P3b is generated by the last and most-informative stimulus of the predicting sequence (n 1) compared with randomized standards, and P3b amplitude then reaches a maximum for predicted and random targets. Thus, the predictive sequence becomes a secondary target for the subjects with the n 1 P3b amplitude indicating how much
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
481
Fig. 3. Grand average at Cz, displaying P3b for the 4 conditions: targets after random non predictive (Random) and predictive sequences (Predicted), the last most informative standard comprising the predicting sequence (n 1) and random preceding standards (Standard) for age-matched controls (A) and patients with depression (B). Vertical dotted lines indicate time of stimulus presentation onset. Note attenuation of n 1 P3b amplitude in patients compared with controls. Topographical maps illustrate individual minima and maxima P3b amplitudes for each of the two target conditions and n 1.
Fig. 4. Grand average at PO7 displaying N1 for random and predicted targets presented to the right visual field, for age-matched controls (A) and patients with depression (B). Vertical dotted lines indicate time of stimulus presentation onset. Note N1 amplitude attenuations in patients compared with controls. Topographical maps illustrate individual minima and maxima N1 amplitudes for each of the two target conditions.
attention was allocated to the sequence and whether it was detected (Johnson, 1986). In the MDD patients we also observe maximal P3b amplitudes for the two target conditions compared with randomized standards. However, MDD patients demonstrated an attenuation of peak P3b amplitudes for the n 1 condition compared with the control group, suggesting that patients were limited in their ability to either allocate attention to or maintain the contextual information provided by the predictive sequence. This is supported by evidence showing working memory online maintenance deficits in depression (Segrave et al., 2010). Alternatively,
reduced decision confidence during the detection of the predictive sequence (Squires et al., 1973, 1975) can also explain the pronounced attenuation of n 1 P3b amplitudes in the MDD patients. This may explain the specific prolongation of RT and P3b latency for predicted targets that was observed in the patients. P3b amplitude reductions observed in the MDD patients are consistent with other ERP studies in depression (Vandoolaeghe et al., 1998; Urretavizcaya et al., 2003; Campanella et al., 2012). In the present study, a general decrease in P3b amplitude was observed, however, this attenuation was specifically pronounced during processing of the
482
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
predictive local context. This finding supports the proposition that patients with depression are particularly vulnerable during the performance of tasks requiring effortful, complex information processing (Pelosi et al., 2000; Hammar et al., 2003; Bruder et al., 2012; Murrough et al., 2011; Campanella et al., 2012), as the detection of the predictive sequence required more motivation and effort. In both MDD patients and controls, RT and P3b latency was shorter for predicted targets compared with random targets, suggesting a facilitation of the processing speed of predicted targets (Kutas et al., 1977; McCarthy and Donchin, 1981; Duncan-Johnson, 1981; Duncan-Johnson and Donchin, 1982; Hillyard and Kutas, 1983; Fogelson et al., 2009b). However, patients demonstrated prolonged RT and P3b latency compared with controls specifically for predicted targets, while there was no difference between the groups of these measures for random targets. These findings suggest that while patients with depression are largely intact in their ability to utilize predictive contextual information, they are limited in their ability to facilitate processing of deterministic targets and in the translation of the predictive contextual information into a faster motor response. Other studies have also shown prolonged RTs (Hammar et al., 2003; Rose and Ebmeier, 2006) and P3b latencies (Vandoolaeghe et al., 1998; Urretavizcaya et al., 2003) in depression. However, our study suggests that this slowing is specific to context-dependent stimuli. In addition, correlations between behavioral and ERP data in both groups demonstrate that for predicted targets the faster the reaction time the shorter the P3b latency (replicating findings in Fogelson et al., 2011b). However, these correlations were not observed for random targets. Perceptual processing of the target stimuli (Hillyard and Kutas, 1983) was similar for the two target conditions in both controls and MDD patients, since there were no significant N1 amplitude differences between predicted and random targets in both groups (Fogelson et al., 2009a, b). However, N1 amplitude attenuations were observed for both target conditions in the MDD patients compared with age-matched controls. This is in line with other findings in patients with depression during the performance of a working memory task, showing an attenuation of N1 that was unrelated to working memory load (Pelosi et al., 2000). This is further supported by the current study, showing a general N1 attenuation across target conditions, that was not correlated with the prolongation of the processing speed of predicted targets observed in the MDD patients. These findings suggest that depression may be associated with an impaired allocation of attentional resources to the appropriate perceptual components within the attentional control network (Pelosi et al., 2000; Gotlib and Joormann, 2010; Murrough et al., 2011). We found no differences in N2 amplitudes between MDD patients and controls, both groups displaying an attenuation of N2 amplitudes during the detection of a predictive target relative to the random condition, suggesting that less attentional resources were required for processing predicted targets in the visual cortex compared to that of random targets (Fitzgerald and Picton, 1983; Suwazono et al., 2000; Folstein and Van Petten, 2008; Fogelson et al., 2011a). 4.2. Altered processing of predictive local context in depression The findings of a specific prolongation of RT and P3b latency during processing of predictable targets and a pronounced attenuation of P3b amplitude to the predictive sequence in MDD patients, provides evidence for altered local contextual processing in depression. This supports evidence of an executive dysfunction and working memory deficits in depression (Pelosi et al., 2000; Rose and Ebmeier, 2006; Segrave et al., 2010; Marazziti et al., 2010;; Murrough et al., 2011; Sato et al., 2011). Our findings
suggest that although MDD patients were able to utilize predictive context in order to facilitate processing of deterministic targets (significant RT and P3b latency differences between predicted and random targets), this ability was limited compared with agematched controls. This limited ability may be related to deficits in the online maintenance of information provided by the predictive sequence, or in the processing of this information, which required effort and motivation, and may have been lacking in the patients (Pelosi et al., 2000; Hammar et al., 2003; Murrough et al., 2011; Campanella et al., 2012). Findings of local contextual processing deficits were also observed in patients with schizophrenia (Fogelson et al., 2011b) and in patients with prefrontal lobe damage (Fogelson et al., 2009a), suggesting a key role of dorsolateral prefrontal cortex in this function (Cohen and Servan-Schreiber, 1992; MacDonald et al., 2000; Miller and Cohen, 2001; Huettel et al., 2005; Fogelson et al., 2009a). In these patient populations prolonged processing speed of predictive targets and attenuated P3b amplitudes of n 1, compared with age-matched controls, were also observed. However, in contrast to the MDD patients, schizophrenia and prefrontal-lesioned patients showed pronounced attenuation of the P3b latency shift and the differential processing between the two target conditions, demonstrating an impaired ability to utilize predictive local context (Fogelson et al., 2009a, 2011b). There is evidence for the role of the prefrontal cortex and prefrontal networks in depression (Bench et al., 1992; Pascual-Leone et al., 1996), and prefrontal network dysfunction in depressed patients has been associated with working memory deficits (Pu et al., 2011; Sato et al., 2011; Price and Drevets, 2012). Thus, it is likely that the deficits in processing local contextual information observed in the MDD patients in the present study may also be related to prefrontal or prefrontal network dysfunction. Finally, this study has several limitations and thus results should be interpreted accordingly. First, the sample size used in the current study was small, although other similar studies used comparable size effects (Pelosi et al., 2000; Segrave et al., 2010; Campanella et al., 2012), and findings need to be replicated in a larger sample, allowing for comparison of gender differences, which were not possible in the present study. Second, the conclusions of the present study may have been confounded by medication effects. Since the patients were hospitalized, acute patients we could not obtain medication-free recordings for ethical reasons. Antidepressant treatment has been shown to either improve working memory and attention (e.g., Herrera-Guzmán et al., 2010), increase N1 amplitudes (Normann et al., 2007) and to either alter or normalize P3b parameters (Karaaslan et al., 2003; Veltmeyer et al., 2009). Accuracy has also been shown to be affected by antidepressant treatment during a working memory task (Veltmeyer et al., 2009). The fact that in the current study patients displayed attenuated N1 amplitudes and that P3b changes were specific to context-dependent stimuli, suggests that these changes are likely related to the illness itself rather than medication effects, although the behavioral and ERP findings were not correlated with the clinical scores. In addition, accuracy rates in patients were similar to controls, suggesting that the patients were alert during the performance of the task. However, it would be of importance to replicate the present findings in a larger cohort of un-medicated MDD patients. In conclusion, the current study provides evidence of altered contextual processing in patients with depression, by demonstrating specific alterations in the behavioral and neural correlates of local contextual processing. MDD patients were able to utilize predictive context in order to facilitate processing of deterministic targets, however, this ability was limited compared to age-matched controls, as demonstrated by context-dependent P3b deficits.
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
Acknowledgments This work was supported by the Spanish Ministry of Science and Innovation [PSI2012-34212 to NF] and [DEP2011-22466 and Xunta de Galicia CN 2012/170 to MFdO]. References Barch DM, Carter CS, Braver TS, Sabb FW, MacDonald A, Noll DC, et al. Selective deficits in prefrontal cortex function in medication-naive patients with schizophrenia. Arch Gen Psychiatr 2001;58:280–8. Barch DM, Sheline YI, Csernansky JG, Snyder AZ. Working memory and prefrontal cortex dysfunction: specificity to schizophrenia compared with major depression. Biol Psychiatry 2003;53:376–84. Bench CJ, Friston KJ, Brown RG, Scott LC, Frackowiak RS, Dolan RJ. The anatomy of melancholia-focal abnormalities of cerebral blood flow in major depression. Psychol Med 1992;22:607–15. Bruder GE, Kayser J, Tenke CE. Event-related brain potentials in depression: clinical, cognitive and neurophysiologic implications. In: Luck SJ, Kappenman ES, editors. The oxford handbook of event-related potential components. New York: Oxford University Press; 2012. p. 563–92. Campanella S, Delle-Vigne D, Kornreich C, Verbanck P. Greater sensitivity of the P300 component to bimodal stimulation in an event-related potentials oddball task. Clin Neurophysiol 2012;123:937–46. Cohen JD, Servan-Schreiber D. Context, cortex, and dopamine: a connectionist approach to behavior and biology in schizophrenia. Psychol Rev 1992;99:45–77. Donchin E, Coles MGH. Is the P300 component a manifestation of context updating? Behav Brain Sci 1988;11:357–74. Duncan-Johnson CC. P300 latency: a new metric of information processing. Psychophysiology 1981;18:207–15. Duncan-Johnson CC, Donchin E. The P300 component of the event-related brain potential as an index of information processing. Biol Psychol 1982;14:1–52. Fitzgerald PG, Picton TW. Event-related potentials recorded during the discrimination of improbable stimuli. Biol Psychol 1983;17:241–76. Fogelson N, Shah M, Scabini D, Knight RT. Prefrontal cortex is critical for contextual processing: evidence from brain lesions. Brain 2009a;132:3002–10. Fogelson N, Wang X, Lewis JB, Kishiyama MM, Ding M, Knight RT. Multimodal effects of local context on target detection: evidence from P3b. J Cogn Neurosci 2009b;21:1680–92. Fogelson N, Shah M, Bonnet-Brihault F, Knight RT. Electrophysiological evidence for aging effects on local contextual processing. Cortex 2010;46:498–506. Fogelson N, Fernandez-del-Olmo M, Santos-Garcia D. Contextual processing deficits in Parkinson’s disease: role of the frontostriatal system. Clin Neurophysiol 2011a;122:539–45. Fogelson N, Ribolsi M, Fernandez-del-Olmo M, Rubino IA, Romeo D, Koch G, et al. Neural correlates of local contextual processing deficits in schizophrenic patients. Psychophysiology 2011b;48:1217–26. Folstein JR, Van Petten C. Influence of cognitive control and mismatch on the N2 component ERP: a review. Psychophysiology 2008;45:152–70. Gotlib IH, Joormann J. Cognition and depression: current status and future directions. Annu Rev Clin Psychol 2010;6:285–312. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatr 1960;23:56–62. Hammar Å, Lund A, Hugdahl K. Automatic and effortful information processing in unipolar major depression. Scand J Psychol 2003;44:409–13. Herrera-Guzmán I, Herrera-Abarca JE, Gudayol-Ferré E, Herrera-Guzmán D, GómezCarbajal L, Peña-Olvira M, Villuendas-González E, Joan GO. Effects of selective serotonin reuptake and dual serotonergic-noradrenergic reuptake treatments on attention and executive functions in patients with major depressive disorder. Psychiatry Res 2010;177:323–9. Hillyard SA, Kutas M. Electrophysiology of cognitive processing. Annu Rev Psychol 1983;34:33–61. Holmes AJ, MacDonald A, Carter CS, Barch DM, Andrew Stenger V, Cohen JD. Prefrontal functioning during context processing in schizophrenia and major depression: an event-related fMRI study. Schizophr Res 2005;76:199–206. Huettel SA, Song AW, McCarthy G. Decisions under uncertainty: probabilistic context influences activation of prefrontal and parietal cortices. J Neurosci 2005;25:3304–11.
483
Johnson RJR. A triarchic model of P300 amplitude. Psychophysiology 1986;23:367–84. Karaaslan F, Gonul AS, Oguz A, Erdinc E, Esel E. P300 changes in major depressive disorders with and without psychotic features. J Affect Disord 2003;73:283–7. Kutas M, McCarthy G, Donchin E. Augmenting mental chronometry: the P300 as a measure of stimulus evaluation time. Science 1977;197:792–5. MacDonald AW, Cohen JD, Stenger VA, Carter CS. Dissociating the role of the dorsolateral prefrontal and anterior cingulated cortex in cognitive control. Science 2000;288:1835–8. Marazziti D, Consoli G, Picchetti M, Carlini M, Faravelli L. Cognitive impairment in major depression. Eur J Pharmacol 2010;626:83–6. McCarthy G, Donchin E. A metric for thought: a comparison of P300 latency and reaction time. Science 1981;211:77–80. Miller EK, Cohen JD. An integrative theory of prefrontal cortex function. Annu Rev Neurosci 2001;24:167–202. Murrough JW, Iacoviello B, Neumeister A, Charney DS, Iosifescu DV. Cognitive dysfunction in depression: neurocircuitry and new therapeutic strategies. Neurobiol Learn Mem 2011;96:553–63. Normann C, Schmitz D, Fürmaier A, Döing C, Bach M. Long-term plasticity of visually evoked potentials in humans is altered in major depression. Biol Psychiatry 2007;62:373–80. Pascual-Leone A, Rubio B, PallardoÁ F, CatalaÁ MD. Rapid-rate transcranial magnetic stimulation of left dorsolateral prefrontal cortex in drug-resistant depression. Lancet 1996;348:233–7. Pelosi L, Slade T, Blumhardt LD, Sharma VK. Working memory dysfunction in major depression: an event-related potential study. Clin Neurophysiol 2000;111:1531–43. Pfefferbaum A, Wenegrat B, Ford J, Roth W, Kopell B. Clinical application of the P3 component of event-related potentials. II. Dementia, depression and schizophrenia. Electroenceph Clin Neurophysiol 1984;59:104–24. Price JL, Drevets WC. Neural circuits underlying the pathophysiology of mood disorders. Trends Cogn Sci 2012;16:61–71. Pu S, Yamada T, Yokoyama K, Matsumura H, Kobayashi H, Sasaki N, et al. A multichannel near-infrared spectroscopy study of prefrontal cortex activation during working memory task in major depressive disorder. Neurosci Res 2011;70:91–7. Rose EJ, Ebmeier KP. Pattern of impaired working memory during major depression. Affect Disord 2006;90:149–61. Roth WT, Pfefferbaum A, Kelly AF, Berger PA, Kopell BS. Auditory event-related potentials in schizophrenia and depression. Psychiatr Res 1981;4:199–212. Sato H, Aoki R, Katura T, Matsuda R, Koizumi H. Correlation of within-individual fluctuation of depressed mood with prefrontal cortex activity during verbal working memory task: optical topography study. J Biomed Opt 2011;16:126007. Segrave RA, Thomson RH, Cooper NR, Croft RJ, Sheppard DM, Fitzgerald PB. Upper alpha activity during working memory processing reflects abnormal inhibition in major depression. J Affect Disord 2010;127:191–8. Squires KC, Hillyard SA, Lindsay PH. Vertex potentials evoked during auditory signal detection: relation to decision criteria. Percept Psychophys 1973;14:265–72. Squires KC, Squires NK, Hillyard SA. Decision-related cortical potentials during an auditory signal detection task with cued observation intervals. J Exp Psychol Hum Percept Perform 1975;1:268–79. Squires KC, Wickens C, Squires NK, Donchin E. The effect of stimulus sequence on the waveform of the cortical event-related potential. Science 1976;193:1142–6. Suwazono S, Machado L, Knight RT. Predictive value of novel stimuli modifies visual event-related potentials and behavior. Clin Neurophysiol 2000;111:29–39. Urretavizcaya M, Moreno I, Benlloch L, Cardoner N, Serrallonga J, Menchón JM, et al. Auditory event-related potentials in 50 melancholic patients: increased N100, N200 and P300 latencies and diminished P300 amplitude. J Affect Disord 2003;74:293–7. Vandoolaeghe E, Van Hunsel F, Nuyten D, Maes M. Auditory event related potentials in major depression: prolonged P300 latency and increased P200 amplitude. J Affect Disord 1998;48:105–13. Veltmeyer MD, Clark CR, McFarlane AC, Moores KA, Bryant RA, Gordon E. Working memory function in post-traumatic stress disorder: an event-related potential study. Clin Neurophysiol 2009;120:1096–106. Verleger R, Jas´kowski P, Wascher E. Evidence for an integrative role of P3b in linking reaction to perception. J Psychophysiol 2005;19:165–81.
Contents lists available at ScienceDirect
Clinical Neurophysiology journal homepage: www.elsevier.com/locate/clinph
Local contextual processing in major depressive disorder Noa Fogelson a,⇑, Avi Peled b,d, Sarah Marmor c,d, Miguel Fernandez-del-Olmo e, Ehud Klein c,d a
Department of Psychology, University of A Coruña, La Coruña, Spain Institute for Psychiatric Studies, Sha’ar Menashe Mental Health Center, Hadera, Israel c Department of Psychiatry, Rambam Medical Center, Technion, Haifa, Israel d B Rappaport Faculty of Medicine, Technion, Haifa, Israel e Department of Physical Education, University of A Coruña, La Coruña, Spain b
a r t i c l e
i n f o
Article history: Accepted 4 September 2013 Available online 26 September 2013 Keywords: Context Major depression EEG P3b N1
h i g h l i g h t s Reaction times and P3b latencies of predicted targets were prolonged in depression. P3b amplitudes were attenuated in major depressive disorder during processing of predictive
contextual information, as well as target N1 amplitudes. Processing of local contextual processing is altered in major depressive disorder.
a b s t r a c t Objective: The study investigated local contextual processing in patients with major depressive disorder (MDD). This was defined as the ability to utilize predictive contextual information to facilitate detection of predictable versus random targets. Method: We recorded EEG in 15 MDD patients and 14 age-matched controls. Recording blocks consisted of targets preceded by randomized sequences of standards and by sequences of standards that included a predictive sequence signaling the occurrence of a subsequent target event. Results: Both MDD patients and age-matched controls demonstrated a significant reaction time (RT) and P3b latency differences between predicted and random targets. However, patients demonstrated a specific prolongation of these measures during processing of predicted targets, as well as an attenuation of P3b amplitudes for the predictive sequence. In addition, patients target N1 amplitudes were attenuated compared with controls. Conclusion: MDD patients were able to utilize predictive context in order to facilitate processing of deterministic targets, however, this ability was limited compared to controls, as demonstrated by contextdependent P3b deficits. Significance: These findings suggest that patients with major depression have altered processing of local contextual processing. Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction Contextual processing is considered to be an executive function and a subcomponent of working memory (Cohen and ServanSchreiber, 1992; Barch et al., 2001). Processing of goal-directed contextual information is essential for the selection of appropriate task behavior and facilitates the detection of task-relevant stimuli (Fogelson et al., 2009b). Major depressive disorder (MDD) is the most prevalent of all psychiatric disorders and is characterized by a wide range of ⇑ Corresponding author. Address: Department of Psychology, Campus de Elviña, La Coruña 15071, Spain. Tel.: +34 981167000x1785; fax: +34 981105641. E-mail address: [email protected] (N. Fogelson).
behavioral, emotional and cognitive symptoms (reviewed by Gotlib and Joormann, 2010; Marazziti et al., 2010; Murrough et al., 2011). Performance on measures of executive function tend to be impaired in depressed patients, and working memory deficits are common (Pelosi et al., 2000; Rose and Ebmeier, 2006; Segrave et al., 2010; Marazziti et al., 2010; Sato et al., 2011) and associated with impairments in attentional control (Pelosi et al., 2000; Rose and Ebmeier, 2006; Segrave et al., 2010; Gotlib and Joormann, 2010; Murrough et al., 2011). However, other studies have demonstrated that there are no impairments in working memory functions in patients with depression (Barch et al., 2003; Holmes et al., 2005). One measure that has been linked to contextual processing is the event related potential (ERP) called the P300 (Squires et al.,
1388-2457/$36.00 Ó 2013 International Federation of Clinical Neurophysiology. Published by Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clinph.2013.09.001
477
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
1976; Donchin and Coles, 1988). The target P300, known as the P3b, is elicited, among other tasks, by targets in the classical oddball paradigm and has a posterior-parietal scalp distribution (Squires et al., 1975). P3b is thought to be a measure of the evaluation of environmental signals including contextual information (Squires et al., 1976; Donchin and Coles 1988) and is also thought to reflect monitoring processes mediating perceptual analysis and response initiation (Verleger et al. 2005). Findings of P300 abnormalities in patients with depression are inconsistent. Although there is a general trend for patients with depression to show P3b amplitude reductions (Roth et al., 1981; Pfefferbaum et al., 1984; Vandoolaeghe et al., 1998; Urretavizcaya et al., 2003; Campanella et al., 2012), these findings depend on stimulus modality (auditory versus visual), the complexity of the paradigm (oddball versus more cognitively demanding tasks) and on the subtype and severity of the symptoms (e.g., reviewed in Bruder et al., 2012). Studies utilizing more cognitively demanding tasks tend to show more consistent reductions in P3b amplitudes in patients with depression (Bruder et al., 2012). In the present study, we focus on the ability to process goal-relevant local contextual information. Previous studies have shown that P3b is modulated as a function of local predictive context (Fogelson et al., 2009a,b, 2010). In these studies, targets were preceded by either randomized sequences of standards or by sequences including a predictive sequence, signaling the occurrence of a subsequent target event. Two main neural correlates of contextual processing were identified. First, a facilitated processing of predicted targets was observed, as indicated by faster P3b latencies and reaction times for predicted targets compared with random ones. Second, local contextual processing was associated with the generation of a robust P3b to the final most-informative stimulus of the predicting sequence. Local contextual processing has been shown to be altered by aging (Fogelson et al., 2010) and impaired in patients with lateral prefrontal cortex lesions (Fogelson et al., 2009a), Parkinson’s disease (Fogelson et al., 2011a) and schizophrenia (Fogelson et al., 2011b), demonstrating specific alterations in the behavioral measures and neural correlates of local contextual processing in each patient population. It remains inconclusive whether patients with depression have working memory impairments. Thus, the objective of the current study was to investigate whether patients with major depressive disorder are impaired in their ability to process contextual information, an important subcomponent of working memory, utilizing the previously described paradigm (Fogelson et al., 2009a, 2010, 2011a,b). We evaluated neural correlates of contextual processing based on both behavioral and ERP measures to determine whether
a more cognitively demanding version of the oddball task will induce changes in performance in patients compared with controls. Our hypothesis was that if patients with major depression do have contextual processing deficits, we should observe alterations in the behavioral and electrophysiological indices of local contextual processing. 2. Methods 2.1. Participants 15 MDD patients (mean age ± standard error of the mean = 43 ± 4 years, 9 females) and 14 age-matched controls (mean age = 43 ± 4 years, 8 females) participated in the study. Patients were diagnosed with MDD according to DMS-IV-Tr criteria, and were rated for symptom severity using the Hamilton Rating Scale for Primary Depressive Illness (HRSD, Hamilton, 1960). Subjects with past history of neurologic disorders, drug or alcohol abuse were excluded. Mean illness duration was 3 ± 1 years. Mean HRSD scores were 30.2 ± 2.5. Demographics and clinical details of the patients are shown in Table 1. All patients had normal or corrected-to-normal visual acuity. All patients were hospitalized at the time of the experiment due to an episode of major depression and took their regular medications on the day the recordings. Most patients were medicated with a selective serotonin reuptake inhibitor (SSRI), some patients were treated with a mood stabilizer and benzodiazepine. Medication dosages were within regular therapeutic levels, achieving average therapeutic dosage based on blood levels. Patients were matched by controls for age, gender and education (mean = 14.6 ± .5 and 14.4 ± 1 years of education for MDD and controls, respectively). Age-matched controls had normal or corrected-to-normal visual acuity and had no history of psychiatric or neurological problems. All the subjects were right handed as indicated by self-report. The experimental procedures were approved by the local ethics committees. Written informed consent was obtained from all subjects participating in the study following a complete explanation of the study and procedures. 2.2. Procedure Subjects were seated 110 cm in-front of a 21-inch PC-computer screen. Stimuli were presented to either the left or right visual field 6° from a central point of fixation. The stimuli consisted of black triangles on a gray background. Subjects were asked to centrally
Table 1 Clinical details of patients with major depression. Patient (sex)
Diagnosis
HRSD
Medication
1 (F) 2 (F) 3 (F)
1 2 5
Major depression Major depression Major depression
21 26 38
4 (F) 5 (F) 6 (F)
1 1 1
Major depression Major depression Major depression + borderline personality disorder Depression + anxiety Major depression Major depression + anorexia Major depression Major depression Major depression Major depression Major depression Major depression
18 27 19
Mirtazapine Lithium Venalafaxine carbamazepine clonazepam risperidone Escitalopram sertraline qutiapin clonazepam Paroxetine
26 17 27 44 33 34 43 45 35
Lithium Qutiapin lithium mirtazapine Venalafaxine mirtazapine Fluoxetine Escitalopram Escitalopram Fluoxetine Fluvoxamine zolpidem lithium Escitalopram
7 (F) 8 (M) 9 (M) 10 (F) 11 (M) 12 (M) 13 (M) 14 (F) 15 (M)
Disease duration (years)
3 5 1 10 10 7 0.5 1 1.5
HRSD, Hamilton rating scale for primary depressive illness (Hamilton, 1960).
478
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
2.4. Behavioral analysis Accuracy was defined as the percentage of targets for which a button press was detected. Reaction times were calculated by averaging correct trials for predicted and random targets in each subject. Misses (no button press 150–1150 ms post-stimulus onset) were excluded from reaction time analysis. Reaction times were analyzed using E-prime (Psychology Software Tools, Inc., Pittsburgh, USA). 2.5. ERP analysis Fig. 1. Task timeline. Sequences of standards S1, S2 and S3 with a predicted sequence (top) and in randomized order (bottom) preceding the target (T). The predictive sequence is always S1 followed by S2 and then S3 (n 1). Stimuli presented to the left or right visual field. Inter-trial intervals, including duration of stimulus presentation (150 ms) are displayed.
fixate throughout the recording. Stimuli consisted of 15% targets (downward facing triangle) and 85% of equal amounts of three types of standards (triangles facing left, upwards and right, at 90° increments). In each block a total of 78 stimuli (12 targets, 22 of each standard type) were presented each for 150 ms and interstimulus interval (ISI) of 1 s. Recording blocks consisted of targets preceded by either randomized sequences of standards or by sequences including a three-standard predictive sequence. The predictive sequence always consisted of the three standards of triangles facing left, up and right, always in that order. Fig. 1 illustrates an example of a target preceded by a randomized sequence of standards and a target preceded by the predictive sequence of standards. Each block consisted of 6 different randomized sequences of standards (3–8 standards long) preceding the target; and 6 sequences of standards (3–8 standards long) with a predictive sequence preceding the target in each. Each recording session consisted of 14 different blocks, displayed in randomized order, each approximately 1.6 min long. A single sequence of trials appeared on the right or left hemifield in a randomized order within each block. Blocks were counterbalanced so that there were equal amount of stimuli presented to the right and left visual hemi-field across the blocks. Within a single sequence of trials (predictive or random) all stimuli were presented in one hemifield. The predictive sequence was always followed by a target. Subjects performed a brief training session to ensure they were able to detect the target accurately. Subjects were then shown the predictive sequence and were told that it would be 100% predictive of a target, but that targets would also appear randomly throughout the block. Subjects were asked to press a button with their right index finger each time a target was presented and to pay attention and look for the predictive sequence. Subjects then performed another brief training session to ensure that they were confident in the detection of the predictive sequence as well as the targets, before the recording session began. Stimulus presentation and response recordings were controlled using E-prime (Psychology Software Tools, Inc., Pittsburgh, USA).
2.3. Data acquisition EEG was recorded from a 64 electrode array using the ActiveTwo system (Biosemi, The Netherlands). External electrodes above and below the right eye monitored vertical eye movements and electrodes placed laterally to the left and right eyes monitored horizontal eye movements. Signals were amplified and digitized at 512 Hz. All channels were re-referenced to averaged linked earlobes.
Post processing and ERP analysis of the data was performed using Brain Vision Analyzer version 1.05 (Brain Products GmbH, Germany). Prior to ERP analysis eye saccades and blinks were defined using ICA (64 EEG electrodes were included), and the identified components were removed using the linear derivation function in Brain Vision Analyzer. Epochs containing misses (no button press 150–1150 ms post-stimulus onset) were excluded from further analysis. EEG signals were filtered at 0.1–30 Hz, using Butterworth zero phase filters, for subsequent analysis. EEG signals were sorted and averaged relative to the stimulus onset, with epochs set from 200 to 1000 ms relative to stimulus onset. A baseline correction of 200 ms pre-stimulus onset ( 200 to 0 ms) was performed. The maximal and minimal permissible amplitudes at any electrode for EEG epochs that were included in the analysis were 75 and 75 lV, respectively. 2.6. P3b P3b was determined as the most positive point in the latency range of 300–700 ms. In order to restrict the number of comparisons and since no lateralization effects of hemisphere or visual field were observed, all P3b ERP data were collapsed across visual fields. In addition we concentrated on midline electrode sites (AFz, Fz, FCz, Cz, CPz, and Pz) to explore anterior versus posterior topographical differences of P3b for the different target conditions. Peak P3b amplitude (measured in lV) at AFz, Fz, FCz, Cz, CPz, and Pz were evaluated for 4 conditions: targets after predictive sequences (predicted), targets after non predictive random sequences (random), random preceding standards (standards excluding those comprising of the predicting sequence) and the last most-informative standard of the predicting sequence (n 1) for both groups. There were comparable number of trials in patients and controls for predicted (65 ± 4, MDD and 62 ± 4, controls), random (63 ± 4, MDD and 65 ± 3, controls), n 1 (68 ± 4, MDD and 62 ± 3, controls) and standard (542 ± 37, MDD and 529 ± 25, controls) conditions after removal of misses and artifacts. Peak P3b latencies (measured in ms) were evaluated at at AFz, Fz, FCz, Cz, CPz, and Pz for predicted and random targets. 2.7. N1 To assess the early perceptual processes between the two target conditions, peak N1 amplitudes (measured in lV) were determined at PO7 and PO8, for both predicted and random targets presented to the right or left visual fields. N1 was determined as the most negative peak in the latency range of 50–200 ms. 2.8. N2 To assess N2 amplitude differences between the two target conditions, peak N2 amplitudes (measured in lV) were determined at electrode sites AFz, Fz, FCz, Cz, CPz, and Pz, for predicted and random targets. N2 was defined as the largest negative peak in
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
the time window starting with the P2 peak and ending 350 ms after target onset. N2 amplitudes were measured against the amplitude of the preceding P2, which was determined as the largest positive peak from 150 ms after target onset until the N2 peak. 2.9. Statistical analysis Analysis of variance (ANOVA) was performed with the Greenhouse-Geisser correction, followed by post hoc parametric paired t-tests, Sidak corrected for multiple comparisons unless otherwise stated. Partial eta squared (g2p ) and epsilon values are reported where applicable. Pearson’s Product Moment correlation coefficient was used to calculate correlations. Mean values with ± standard error of the mean (SEM) are used throughout the text. 3. Results 3.1. Behavioral results There were no significant differences in accuracy between MDD patients and controls for predicted targets (mean accuracy = 97 ± 2% and 98 ± 1%, for MDD and controls, respectively, t(27) = .11, p = .912) and random targets (mean accuracy = 96 ± 2% and 99 ± .2%, for MDD and controls, respectively, t(27) = 1.41, p = .17). To compare the reaction times (RT) for the targets between the groups we performed an ANOVA with group (MDD patients, controls) as the between-subjects factor and condition (predicted, random targets) as the repeated measures factor. There was a main effect for condition (F(1,27) = 61.32, p < .0001, g2p = .69) and a significant condition group interaction (F(1,27) = 8.47, p = .007, g2p = .24). Post hoc t-tests showed that RTs for predicted targets (mean RT = 312 ± 21 ms and 423 ± 19 ms, for controls and MDD patients, respectively) were shorter than those for random targets (mean RT = 454 ± 17 ms, t(13) = 7.03, p < .0001, and 488 ± 20 ms, t(14) = 3.77, p = .002, for controls and MDD patients, respectively) in both controls and MDD patients. However, RTs for predicted targets were shorter in controls compared with patients (t(27) = 3.95, p = .001), while RTs for random targets were not significantly different between patients and controls (t(27) = 1.28, p = 0.213). Reaction time comparisons are displayed in Fig. 2A. 3.2. P3b Topographical maps and waveforms of the grand-averaged ERPs across the subjects in the control group and across the subjects in the patient group, at electrode site Cz elicited by predicted and random targets, standards and n 1, the last most-informative stimulus of the predicting sequence, are shown in Fig. 3. To compare peak P3b amplitudes we performed an ANOVA with group (MDD patients, controls) as the between-subjects factor, and with condition (predicted, random targets, n 1, and standards) and electrode (AFz, Fz, FCz, Cz, CPz, and Pz) as the repeated measures factors. There were main effects for condition (F(3,81) = 35.68, p < .0001, g2p = .57, epsilon = .55) and electrode (F(5,135) = 23.49, p < .0001, g2p = .47, epsilon = .39). There was a main effect of group (p = .016), demonstrating larger P3b amplitudes in controls compared with patients, and a significant electrode x group interaction (F(5,135) = 3.64, p = .033, g2p = .20). Both controls and patients showed maximal P3b amplitudes at electrode sites CPz and Cz, respectively. Both groups showed minimal P3b amplitudes at electrode AFz. Post-hoc tests, Sidak corrected for multiple comparisons, showed that in both controls and patients the peak P3b amplitude was larger for predicted and random targets compared with standards (p 6 .02), and there were no
479
significant differences in P3b amplitude between random and predicted targets. However, only controls displayed larger P3b amplitudes for the n 1 condition compared with standards (p < .0001), while in the patients this difference was non-significant (p = .161). In addition, independent t-tests corrected for multiple comparisons showed smaller P3b amplitudes for the n 1 condition (at Cz, CPz and Pz) in patients compared to controls (t(27) < 4.1, p < .002). To compare peak P3b latencies we performed an ANOVA with group (MDD patients, controls) as the between-subjects factor and with condition (predicted, random targets) and electrode (AFz, Fz, FCz, Cz, CPz, and Pz) as the repeated measures factors. There was a main effect for condition (F(1,27) = 28.73, p < .0001, g2p = .52) but no main effect for electrode (F(5,135) = .49, p = .597, g2p = .02, epsilon = .36). There was a significant condition x group interaction (F(1,27) = 5.67, p = .025, g2p = .17). Both patients and controls showed peak P3b latency to be shorter for predicted targets compared with random targets (t(14) < 3.3, p 6 .045 and t(13) < 4.6, p 6 .002 for patients and controls, respectively) across the 6 electrode sites. However, independent t-tests revealed P3b latency for predicted targets to be longer in patients compared to controls (at Pz, CPz, Fz and Cz, t(27) < 2.5, p 6 .042), while no significant P3b latency differences were found between the groups for random targets. P3b latency comparisons are displayed in Fig. 2B.
3.3. N1 We utilized an ANOVA with group (MDD patients, controls) as the between-subjects factor and electrode (PO7 and PO8), visual hemifield (left, right) and condition (predicted, random targets) as repeated measures factors, to compare the peak N1 amplitude between predicted and random targets. There were no significant main effects for condition, electrode or side. There was a significant interaction between extrastriate electrode location and visual field of stimuli presentation (F(1,27) = 17.77, p < .0001, g2p = .40), demonstrating that N1 ERPs were enhanced to contralaterally presented stimuli. In addition, there was an electrode visual hemifield group interaction (F(1,27) = 5.79, p = .023, g2p = .18) showing smaller N1 amplitudes to contralaterally presented stimuli in patients with depression compared with controls. Post-hoc comparisons revealed no significant N1 amplitude differences between predicted and random targets across groups. N1 waveforms and topographical maps are demonstrated in Fig. 4.
3.4. N2 To compare peak N2 amplitudes we performed an ANOVA with group (MDD patients, controls) as the between-subjects factor and with condition (predicted, random targets) and electrode (AFz, Fz, FCz, Cz, CPz, and Pz) as the repeated measures factors. There were main effects for condition (F(1,27) = 29.51, p < .0001, g2p = .52) and electrode (F(5,135) = 5.19, p = .007, g2p = .16, epsilon = .43). There was a significant condition group interaction (F(1,27) = 5.35, p = .029, g2p = .16). Post hoc tests corrected for multiple comparisons showed peak N2 amplitude to be maximal at FCz and minimal at AFz across the groups. However, the independent t-tests for each target condition (predicted and random targets) showed no significant N2 amplitude differences between the groups across the 6 electrode sites. Both controls (all 6 electrodes, t(13) < 6.5, p 6 .003) and patients (at Pz, CPz, FCz, Cz, t(14) < 2.8, p 6 .04) showed N2 amplitude to be larger for random targets compared with predicted targets.
480
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
Fig. 2. Reaction times (A) and peak P3b latency at Cz (B) for predicted and random targets in patients with depression and age-matched controls, demonstrating specific prolongation of these measures in the predicted targets condition in patients compared with controls. Bars = SEM.
3.5. Correlations
4. Discussion
In order to determine the possible effects of clinical symptoms on the behavioral and electrophysiological measures associated with local contextual processing in the MDD patients, we calculated correlations between daily HRSD scores and illness duration and the following measures: RT for predicted and random targets, peak P3b amplitude and latency for predicted and random targets, peak P3b latency shift, n 1 peak P3b amplitude (all at Cz) and N1 amplitude for predicted and random targets (at PO7), across the 15 patients. There were no significant correlations between HRSD scores and illness duration with any of the behavioral or electrophysiological measures. In order to determine the association between behavioral measures and the electrophysiological findings, reaction times for predicted and random targets were correlated with P3b amplitudes for the predicted sequence, P3b latencies and N1 amplitudes for the target conditions, in MDD and control subjects. RTs for predicted targets were significantly correlated with peak P3b latency for predicted targets in MDD patients (r = .581, p = .023) and in controls (r = .693, p = .006). Both groups showed no significant correlations between RT for random targets and any of the electrophysiological measures. No correlations were observed between N1 amplitudes for target conditions and any of the other variables.
The current study provides electrophysiological evidence that processing of local contextual information is altered in patients with major depression. This was demonstrated by the following findings. First, although both patients and controls showed a RT difference and P3b latency shift between predicted and non-predictive targets, there was a specific prolongation of these measures for predicted targets in the patients. Second, patients demonstrated an attenuation of peak P3b to the last most-informative stimulus (n 1) of the predictive sequence compared with controls. Finally, there was an attenuation of N1 amplitudes for target conditions in the patients.
4.1. Depression and local contextual processing effects on ERPs In controls we replicated previous findings (Fogelson et al., 2009a, b, 2010, 2011a,b) showing that P3b amplitude increased with task-informative stimuli, so that a significant P3b is generated by the last and most-informative stimulus of the predicting sequence (n 1) compared with randomized standards, and P3b amplitude then reaches a maximum for predicted and random targets. Thus, the predictive sequence becomes a secondary target for the subjects with the n 1 P3b amplitude indicating how much
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
481
Fig. 3. Grand average at Cz, displaying P3b for the 4 conditions: targets after random non predictive (Random) and predictive sequences (Predicted), the last most informative standard comprising the predicting sequence (n 1) and random preceding standards (Standard) for age-matched controls (A) and patients with depression (B). Vertical dotted lines indicate time of stimulus presentation onset. Note attenuation of n 1 P3b amplitude in patients compared with controls. Topographical maps illustrate individual minima and maxima P3b amplitudes for each of the two target conditions and n 1.
Fig. 4. Grand average at PO7 displaying N1 for random and predicted targets presented to the right visual field, for age-matched controls (A) and patients with depression (B). Vertical dotted lines indicate time of stimulus presentation onset. Note N1 amplitude attenuations in patients compared with controls. Topographical maps illustrate individual minima and maxima N1 amplitudes for each of the two target conditions.
attention was allocated to the sequence and whether it was detected (Johnson, 1986). In the MDD patients we also observe maximal P3b amplitudes for the two target conditions compared with randomized standards. However, MDD patients demonstrated an attenuation of peak P3b amplitudes for the n 1 condition compared with the control group, suggesting that patients were limited in their ability to either allocate attention to or maintain the contextual information provided by the predictive sequence. This is supported by evidence showing working memory online maintenance deficits in depression (Segrave et al., 2010). Alternatively,
reduced decision confidence during the detection of the predictive sequence (Squires et al., 1973, 1975) can also explain the pronounced attenuation of n 1 P3b amplitudes in the MDD patients. This may explain the specific prolongation of RT and P3b latency for predicted targets that was observed in the patients. P3b amplitude reductions observed in the MDD patients are consistent with other ERP studies in depression (Vandoolaeghe et al., 1998; Urretavizcaya et al., 2003; Campanella et al., 2012). In the present study, a general decrease in P3b amplitude was observed, however, this attenuation was specifically pronounced during processing of the
482
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
predictive local context. This finding supports the proposition that patients with depression are particularly vulnerable during the performance of tasks requiring effortful, complex information processing (Pelosi et al., 2000; Hammar et al., 2003; Bruder et al., 2012; Murrough et al., 2011; Campanella et al., 2012), as the detection of the predictive sequence required more motivation and effort. In both MDD patients and controls, RT and P3b latency was shorter for predicted targets compared with random targets, suggesting a facilitation of the processing speed of predicted targets (Kutas et al., 1977; McCarthy and Donchin, 1981; Duncan-Johnson, 1981; Duncan-Johnson and Donchin, 1982; Hillyard and Kutas, 1983; Fogelson et al., 2009b). However, patients demonstrated prolonged RT and P3b latency compared with controls specifically for predicted targets, while there was no difference between the groups of these measures for random targets. These findings suggest that while patients with depression are largely intact in their ability to utilize predictive contextual information, they are limited in their ability to facilitate processing of deterministic targets and in the translation of the predictive contextual information into a faster motor response. Other studies have also shown prolonged RTs (Hammar et al., 2003; Rose and Ebmeier, 2006) and P3b latencies (Vandoolaeghe et al., 1998; Urretavizcaya et al., 2003) in depression. However, our study suggests that this slowing is specific to context-dependent stimuli. In addition, correlations between behavioral and ERP data in both groups demonstrate that for predicted targets the faster the reaction time the shorter the P3b latency (replicating findings in Fogelson et al., 2011b). However, these correlations were not observed for random targets. Perceptual processing of the target stimuli (Hillyard and Kutas, 1983) was similar for the two target conditions in both controls and MDD patients, since there were no significant N1 amplitude differences between predicted and random targets in both groups (Fogelson et al., 2009a, b). However, N1 amplitude attenuations were observed for both target conditions in the MDD patients compared with age-matched controls. This is in line with other findings in patients with depression during the performance of a working memory task, showing an attenuation of N1 that was unrelated to working memory load (Pelosi et al., 2000). This is further supported by the current study, showing a general N1 attenuation across target conditions, that was not correlated with the prolongation of the processing speed of predicted targets observed in the MDD patients. These findings suggest that depression may be associated with an impaired allocation of attentional resources to the appropriate perceptual components within the attentional control network (Pelosi et al., 2000; Gotlib and Joormann, 2010; Murrough et al., 2011). We found no differences in N2 amplitudes between MDD patients and controls, both groups displaying an attenuation of N2 amplitudes during the detection of a predictive target relative to the random condition, suggesting that less attentional resources were required for processing predicted targets in the visual cortex compared to that of random targets (Fitzgerald and Picton, 1983; Suwazono et al., 2000; Folstein and Van Petten, 2008; Fogelson et al., 2011a). 4.2. Altered processing of predictive local context in depression The findings of a specific prolongation of RT and P3b latency during processing of predictable targets and a pronounced attenuation of P3b amplitude to the predictive sequence in MDD patients, provides evidence for altered local contextual processing in depression. This supports evidence of an executive dysfunction and working memory deficits in depression (Pelosi et al., 2000; Rose and Ebmeier, 2006; Segrave et al., 2010; Marazziti et al., 2010;; Murrough et al., 2011; Sato et al., 2011). Our findings
suggest that although MDD patients were able to utilize predictive context in order to facilitate processing of deterministic targets (significant RT and P3b latency differences between predicted and random targets), this ability was limited compared with agematched controls. This limited ability may be related to deficits in the online maintenance of information provided by the predictive sequence, or in the processing of this information, which required effort and motivation, and may have been lacking in the patients (Pelosi et al., 2000; Hammar et al., 2003; Murrough et al., 2011; Campanella et al., 2012). Findings of local contextual processing deficits were also observed in patients with schizophrenia (Fogelson et al., 2011b) and in patients with prefrontal lobe damage (Fogelson et al., 2009a), suggesting a key role of dorsolateral prefrontal cortex in this function (Cohen and Servan-Schreiber, 1992; MacDonald et al., 2000; Miller and Cohen, 2001; Huettel et al., 2005; Fogelson et al., 2009a). In these patient populations prolonged processing speed of predictive targets and attenuated P3b amplitudes of n 1, compared with age-matched controls, were also observed. However, in contrast to the MDD patients, schizophrenia and prefrontal-lesioned patients showed pronounced attenuation of the P3b latency shift and the differential processing between the two target conditions, demonstrating an impaired ability to utilize predictive local context (Fogelson et al., 2009a, 2011b). There is evidence for the role of the prefrontal cortex and prefrontal networks in depression (Bench et al., 1992; Pascual-Leone et al., 1996), and prefrontal network dysfunction in depressed patients has been associated with working memory deficits (Pu et al., 2011; Sato et al., 2011; Price and Drevets, 2012). Thus, it is likely that the deficits in processing local contextual information observed in the MDD patients in the present study may also be related to prefrontal or prefrontal network dysfunction. Finally, this study has several limitations and thus results should be interpreted accordingly. First, the sample size used in the current study was small, although other similar studies used comparable size effects (Pelosi et al., 2000; Segrave et al., 2010; Campanella et al., 2012), and findings need to be replicated in a larger sample, allowing for comparison of gender differences, which were not possible in the present study. Second, the conclusions of the present study may have been confounded by medication effects. Since the patients were hospitalized, acute patients we could not obtain medication-free recordings for ethical reasons. Antidepressant treatment has been shown to either improve working memory and attention (e.g., Herrera-Guzmán et al., 2010), increase N1 amplitudes (Normann et al., 2007) and to either alter or normalize P3b parameters (Karaaslan et al., 2003; Veltmeyer et al., 2009). Accuracy has also been shown to be affected by antidepressant treatment during a working memory task (Veltmeyer et al., 2009). The fact that in the current study patients displayed attenuated N1 amplitudes and that P3b changes were specific to context-dependent stimuli, suggests that these changes are likely related to the illness itself rather than medication effects, although the behavioral and ERP findings were not correlated with the clinical scores. In addition, accuracy rates in patients were similar to controls, suggesting that the patients were alert during the performance of the task. However, it would be of importance to replicate the present findings in a larger cohort of un-medicated MDD patients. In conclusion, the current study provides evidence of altered contextual processing in patients with depression, by demonstrating specific alterations in the behavioral and neural correlates of local contextual processing. MDD patients were able to utilize predictive context in order to facilitate processing of deterministic targets, however, this ability was limited compared to age-matched controls, as demonstrated by context-dependent P3b deficits.
N. Fogelson et al. / Clinical Neurophysiology 125 (2014) 476–483
Acknowledgments This work was supported by the Spanish Ministry of Science and Innovation [PSI2012-34212 to NF] and [DEP2011-22466 and Xunta de Galicia CN 2012/170 to MFdO]. References Barch DM, Carter CS, Braver TS, Sabb FW, MacDonald A, Noll DC, et al. Selective deficits in prefrontal cortex function in medication-naive patients with schizophrenia. Arch Gen Psychiatr 2001;58:280–8. Barch DM, Sheline YI, Csernansky JG, Snyder AZ. Working memory and prefrontal cortex dysfunction: specificity to schizophrenia compared with major depression. Biol Psychiatry 2003;53:376–84. Bench CJ, Friston KJ, Brown RG, Scott LC, Frackowiak RS, Dolan RJ. The anatomy of melancholia-focal abnormalities of cerebral blood flow in major depression. Psychol Med 1992;22:607–15. Bruder GE, Kayser J, Tenke CE. Event-related brain potentials in depression: clinical, cognitive and neurophysiologic implications. In: Luck SJ, Kappenman ES, editors. The oxford handbook of event-related potential components. New York: Oxford University Press; 2012. p. 563–92. Campanella S, Delle-Vigne D, Kornreich C, Verbanck P. Greater sensitivity of the P300 component to bimodal stimulation in an event-related potentials oddball task. Clin Neurophysiol 2012;123:937–46. Cohen JD, Servan-Schreiber D. Context, cortex, and dopamine: a connectionist approach to behavior and biology in schizophrenia. Psychol Rev 1992;99:45–77. Donchin E, Coles MGH. Is the P300 component a manifestation of context updating? Behav Brain Sci 1988;11:357–74. Duncan-Johnson CC. P300 latency: a new metric of information processing. Psychophysiology 1981;18:207–15. Duncan-Johnson CC, Donchin E. The P300 component of the event-related brain potential as an index of information processing. Biol Psychol 1982;14:1–52. Fitzgerald PG, Picton TW. Event-related potentials recorded during the discrimination of improbable stimuli. Biol Psychol 1983;17:241–76. Fogelson N, Shah M, Scabini D, Knight RT. Prefrontal cortex is critical for contextual processing: evidence from brain lesions. Brain 2009a;132:3002–10. Fogelson N, Wang X, Lewis JB, Kishiyama MM, Ding M, Knight RT. Multimodal effects of local context on target detection: evidence from P3b. J Cogn Neurosci 2009b;21:1680–92. Fogelson N, Shah M, Bonnet-Brihault F, Knight RT. Electrophysiological evidence for aging effects on local contextual processing. Cortex 2010;46:498–506. Fogelson N, Fernandez-del-Olmo M, Santos-Garcia D. Contextual processing deficits in Parkinson’s disease: role of the frontostriatal system. Clin Neurophysiol 2011a;122:539–45. Fogelson N, Ribolsi M, Fernandez-del-Olmo M, Rubino IA, Romeo D, Koch G, et al. Neural correlates of local contextual processing deficits in schizophrenic patients. Psychophysiology 2011b;48:1217–26. Folstein JR, Van Petten C. Influence of cognitive control and mismatch on the N2 component ERP: a review. Psychophysiology 2008;45:152–70. Gotlib IH, Joormann J. Cognition and depression: current status and future directions. Annu Rev Clin Psychol 2010;6:285–312. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatr 1960;23:56–62. Hammar Å, Lund A, Hugdahl K. Automatic and effortful information processing in unipolar major depression. Scand J Psychol 2003;44:409–13. Herrera-Guzmán I, Herrera-Abarca JE, Gudayol-Ferré E, Herrera-Guzmán D, GómezCarbajal L, Peña-Olvira M, Villuendas-González E, Joan GO. Effects of selective serotonin reuptake and dual serotonergic-noradrenergic reuptake treatments on attention and executive functions in patients with major depressive disorder. Psychiatry Res 2010;177:323–9. Hillyard SA, Kutas M. Electrophysiology of cognitive processing. Annu Rev Psychol 1983;34:33–61. Holmes AJ, MacDonald A, Carter CS, Barch DM, Andrew Stenger V, Cohen JD. Prefrontal functioning during context processing in schizophrenia and major depression: an event-related fMRI study. Schizophr Res 2005;76:199–206. Huettel SA, Song AW, McCarthy G. Decisions under uncertainty: probabilistic context influences activation of prefrontal and parietal cortices. J Neurosci 2005;25:3304–11.
483
Johnson RJR. A triarchic model of P300 amplitude. Psychophysiology 1986;23:367–84. Karaaslan F, Gonul AS, Oguz A, Erdinc E, Esel E. P300 changes in major depressive disorders with and without psychotic features. J Affect Disord 2003;73:283–7. Kutas M, McCarthy G, Donchin E. Augmenting mental chronometry: the P300 as a measure of stimulus evaluation time. Science 1977;197:792–5. MacDonald AW, Cohen JD, Stenger VA, Carter CS. Dissociating the role of the dorsolateral prefrontal and anterior cingulated cortex in cognitive control. Science 2000;288:1835–8. Marazziti D, Consoli G, Picchetti M, Carlini M, Faravelli L. Cognitive impairment in major depression. Eur J Pharmacol 2010;626:83–6. McCarthy G, Donchin E. A metric for thought: a comparison of P300 latency and reaction time. Science 1981;211:77–80. Miller EK, Cohen JD. An integrative theory of prefrontal cortex function. Annu Rev Neurosci 2001;24:167–202. Murrough JW, Iacoviello B, Neumeister A, Charney DS, Iosifescu DV. Cognitive dysfunction in depression: neurocircuitry and new therapeutic strategies. Neurobiol Learn Mem 2011;96:553–63. Normann C, Schmitz D, Fürmaier A, Döing C, Bach M. Long-term plasticity of visually evoked potentials in humans is altered in major depression. Biol Psychiatry 2007;62:373–80. Pascual-Leone A, Rubio B, PallardoÁ F, CatalaÁ MD. Rapid-rate transcranial magnetic stimulation of left dorsolateral prefrontal cortex in drug-resistant depression. Lancet 1996;348:233–7. Pelosi L, Slade T, Blumhardt LD, Sharma VK. Working memory dysfunction in major depression: an event-related potential study. Clin Neurophysiol 2000;111:1531–43. Pfefferbaum A, Wenegrat B, Ford J, Roth W, Kopell B. Clinical application of the P3 component of event-related potentials. II. Dementia, depression and schizophrenia. Electroenceph Clin Neurophysiol 1984;59:104–24. Price JL, Drevets WC. Neural circuits underlying the pathophysiology of mood disorders. Trends Cogn Sci 2012;16:61–71. Pu S, Yamada T, Yokoyama K, Matsumura H, Kobayashi H, Sasaki N, et al. A multichannel near-infrared spectroscopy study of prefrontal cortex activation during working memory task in major depressive disorder. Neurosci Res 2011;70:91–7. Rose EJ, Ebmeier KP. Pattern of impaired working memory during major depression. Affect Disord 2006;90:149–61. Roth WT, Pfefferbaum A, Kelly AF, Berger PA, Kopell BS. Auditory event-related potentials in schizophrenia and depression. Psychiatr Res 1981;4:199–212. Sato H, Aoki R, Katura T, Matsuda R, Koizumi H. Correlation of within-individual fluctuation of depressed mood with prefrontal cortex activity during verbal working memory task: optical topography study. J Biomed Opt 2011;16:126007. Segrave RA, Thomson RH, Cooper NR, Croft RJ, Sheppard DM, Fitzgerald PB. Upper alpha activity during working memory processing reflects abnormal inhibition in major depression. J Affect Disord 2010;127:191–8. Squires KC, Hillyard SA, Lindsay PH. Vertex potentials evoked during auditory signal detection: relation to decision criteria. Percept Psychophys 1973;14:265–72. Squires KC, Squires NK, Hillyard SA. Decision-related cortical potentials during an auditory signal detection task with cued observation intervals. J Exp Psychol Hum Percept Perform 1975;1:268–79. Squires KC, Wickens C, Squires NK, Donchin E. The effect of stimulus sequence on the waveform of the cortical event-related potential. Science 1976;193:1142–6. Suwazono S, Machado L, Knight RT. Predictive value of novel stimuli modifies visual event-related potentials and behavior. Clin Neurophysiol 2000;111:29–39. Urretavizcaya M, Moreno I, Benlloch L, Cardoner N, Serrallonga J, Menchón JM, et al. Auditory event-related potentials in 50 melancholic patients: increased N100, N200 and P300 latencies and diminished P300 amplitude. J Affect Disord 2003;74:293–7. Vandoolaeghe E, Van Hunsel F, Nuyten D, Maes M. Auditory event related potentials in major depression: prolonged P300 latency and increased P200 amplitude. J Affect Disord 1998;48:105–13. Veltmeyer MD, Clark CR, McFarlane AC, Moores KA, Bryant RA, Gordon E. Working memory function in post-traumatic stress disorder: an event-related potential study. Clin Neurophysiol 2009;120:1096–106. Verleger R, Jas´kowski P, Wascher E. Evidence for an integrative role of P3b in linking reaction to perception. J Psychophysiol 2005;19:165–81.