- Email: [email protected]
Transition from reactive control to proactive control across conflict adaptation: An sLORETA study
Brain and Cognition 100 (2015) 7–14
Contents lists available at ScienceDirect
Brain and Cognition journal homepage: www.elsevier.com/locate/b&c
Transition from reactive control to proactive control across conflict adaptation: An sLORETA study Kota Suzuki a,⇑, Haruo Shinoda b a Department of Developmental Disorders, National Institute of Mental Health, National Center of Neurology and Psychiatry (NCNP), 4-1-1, Ogawahigashi, Kodaira, Tokyo 187-8553, Japan b Department of Clinical Psychology, Faculty of Psychology, Rissho University, Osaki 4-2-16, Shinagawa, Tokyo 141-8602, Japan
a r t i c l e
i n f o
Article history: Received 23 June 2015 Revised 28 August 2015 Accepted 10 September 2015
Keywords: Conflict adaptation Standardized Low Resolution Brain Electromagnetic Tomography Alpha oscillation N1
a b s t r a c t In a flanker task, behavioral performance is modulated by previous trial compatibility (i.e., conflict adaptation); a longer response time (RT) is found for a compatible stimulus preceded by an incompatible stimulus than by a compatible stimulus, whereas a shorter RT is found for an incompatible stimulus preceded by an incompatible stimulus than by a compatible stimulus. We examined the temporal characteristics of cognitive control across conflict adaptation using prestimulus electroencephalogram oscillatory activity and an event-related potential component, N1. Prestimulus frontal (Fz) and posterior (O1 and O2) alpha1 (7–9 Hz) and alpha2 (10–13 Hz) activities were enhanced in trials preceded by incompatible stimuli more than those preceded by compatible stimuli. Furthermore, there were significant differences of alpha2 current densities between previous trial compatibilities in the superior/medial frontal cortex. We suggested that the modulation of alpha activity by previous trial compatibility was associated with proactive attentional control. N1 amplitude was decreased in trials preceded by incompatible stimuli more than in those preceded by compatible stimuli. N1 current densities in the right inferior frontal cortex were smaller for an incompatible stimulus preceded by an incompatible stimulus than those preceded by a compatible stimulus, suggesting that demands of transient cognitive control induced by an incompatible stimulus were decreased by the proactive control. Moreover, correlational analysis showed that participants with a larger increase in alpha2 activity tended to have a larger decrease in N1 in trials preceded by incompatible stimulus. These findings revealed that the manner of cognitive control for the incompatible stimulus was transited from reactive control to proactive control across conflict adaptation. Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction It is important for adaption to the environment to be able to flexibly change the manner of cognitive control according to the context. Types of cognitive control are temporally distinguished into at least two cognitive controls (Braver, 2012; Braver, Paxton, Locke, & Barch, 2009),1 i.e., ‘‘proactive” control and ‘‘reactive” control. Proactive control represents anticipatory and preparatory attentional control for the interference of the upcoming stimulus. Reactive control is postulated as transient recruitment of cognitive control after the stimulus presentation. ⇑ Corresponding author. E-mail addresses: [email protected] (K. Suzuki), [email protected] (H. Shinoda). In the context of studies regarding conflict adaptation (e.g., Correa, Rao, & Nobre, 2009), proactive control refers to cognitive control triggered by a cue informing the upcoming stimulus compatibility, whereas reactive control refers to cognitive control triggered by the previous trial compatibility. In this study, terms ‘‘proactive” and ‘‘reactive” control were used in the broader definition, in which we assumed that proactive and reactive control was triggered by previous trial compatibilities. 1
http://dx.doi.org/10.1016/j.bandc.2015.09.001 0278-2626/Ó 2015 Elsevier Inc. All rights reserved.
Cognitive control has been well-studied using interference tasks such as the flanker task (Eriksen & Eriksen, 1979), the Stroop task (Stroop, 1935), and the Simon task (Simon, 1969). For example, in the flanker task, participants are required to respond to targets presented in the central visual field, while flankers appear that aid (i.e., a compatible stimulus) or interfere with (i.e., an incompatible stimulus) participants’ accomplishment of the task. Gratton, Coles, and Donchin (1992) reported that shorter response times (RTs) and higher accuracies were observed for a compatible stimulus preceded by a compatible stimulus (cC) than by an incompatible stimulus (iC), whereas longer RTs and lower accuracies were observed for an incompatible stimulus preceded by a compatible stimulus (cI) than by an incompatible stimulus (iI). This phenomenon is called conflict adaptation (or the Gratton effect). Conflict has been used to refer to the competition between target information and irrelevant information induced by an incompatible stimulus (Botvinick, Braver, Barch, Carter, & Cohen, 2001). Researchers have interpreted conflict adaptation as a cognitive control mechanism driven by the degree of conflict
8
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
Table 1 Means (standard deviation) of correct RTs and incorrect response rates classified by present and previous trial compatibility. Present trial compatibility Previous trial compatibility
Compatible
Correct RT (ms)
363.41 (32.57) 4.38 (7.87)
Incorrect response rate (%)
Incompatible
Compatible Incompatible Compatible Incompatible 371.92 (40.96) 5.69 (12.80)
428.14 (65.94) 25.93 (8.25)
417.17 (48.18) 14.73 (7.05)
(Botvinick et al., 2001). Then, it is hypothesized that a high degree of conflict (i.e., induced by an incompatible stimulus) causes the attention to be focused on the central visual field in order to inhibit the interference of flankers on the next trial, whereas a low degree of conflict (i.e., induced by a compatible stimulus) makes the attentional area broaden in order to utilize information about flankers on the next trial (Botvinick et al., 2001; Kerns et al., 2004). In accordance with this hypothesis, cognitive control is transiently required to resolve a high degree of conflict when an incompatible stimulus is presented after a compatible stimulus (Wang et al., 2015). On the other hand, when responding to an incompatible stimulus preceded by an incompatible stimulus, cognitive control is preliminarily executed before the presentation of the stimulus. Thus, we postulate that conflict adaptation is associated with a transition from reactive control to proactive control. An oscillation of the electroencephalogram (EEG) or magnetoencephalogram (MEG) of approximately 10 Hz (i.e., alpha) is sensitive to the proactive attentional state before the stimulus presentation. Typically, enhanced alpha activity in a sensory area is interpreted as functional inhibition (Klimesch, Sauseng, & Hanslmayr, 2007). Consistent with this view, enhanced posterior alpha activity predicted a decrease in the visual perceptual performance (Babiloni, Vecchio, Bultrini, Luca Romani, & Rossini, 2006; Ergenoglu et al., 2004; Hanslmayr et al., 2007). Moreover, the change in alpha activity has been considered as reflecting topdown attentional control, which enhances relevant sensory inputs and inhibits irrelevant ones. Several previous studies have shown that posterior alpha activity is enhanced on the ipsilateral side of the visual field attended to, while it is attenuated at the contralateral side (Freunberger et al., 2008; Rihs, Michel, & Thut, 2007; Rihs, Michel, & Thut, 2009; Sauseng et al., 2005; Thut, Nietzel, Brandt, & Pascual-Leone, 2006; Worden, Foxe, Wang, & Simpson, 2000). In terms of interference tasks, Compton, Huber, Levinson, and Zheutlin (2012) reported that alpha activity is reduced in trials preceded by compatible stimuli in the Stroop task, as compared with neutral stimuli, suggesting that reduced alpha activity is related to the enhancement of information processing about the irrelevant aspect. In addition, Min and Park (2010) examined prestimulus alpha activity in color and shape discrimination tasks, in which participants discriminated colors or shapes of objects, ignoring other aspects. They reported that task performance was worse and posterior alpha activity was enhanced more for the shapediscrimination task than the color task, suggesting that alpha activity was enhanced for inhibition of irrelevant sensory processing. These findings suggest that proactive control for the inhibition of flankers is associated with an increase in posterior alpha activity during the prestimulus interval. Posterior alpha activity has been reported to be regulated by frontoparietal areas. Previous studies have shown that the reduced excitability in the frontal eye fields and intraparietal sulcus cause the loss of the modulation of posterior alpha activities by the attending visual field (Capotosto, Babiloni, Romani, & Corbetta, 2009; Sauseng, Feldheim, Freunberger, & Hummel, 2011). A simultaneous EEG and functional magnetic resonance imaging (fMRI)
study has shown that activity of the superior/medial frontal cortex is positively correlated with alpha lateralization by selective attention (Liu, Bengson, Huang, Mangun, & Ding, 2014). Rana and Vaina (2014) reported that alpha activity of frontoparietal areas was increased when controlling spatial attention. Thus, it is predicted that the alpha activity derived from these regions is also modulated by previous trial compatibility. We consider that N1 reflects the demands of the recruitment of cognitive control after stimulus presentation. N1 is an eventrelated potential (ERP) component observed as a negative peak from 100 to 200 ms at the occipital scalp electrode (Hillyard, Teder-Salejarvi, & Munte, 1998), preceding ERP components relevant to detection of conflict (i.e., N2 and ERN; Yeung, Botvinick, & Cohen, 2004). Typically, N1 is called a ‘‘mesogenous” component (Fabiani & Gratton, 2007). That is, both physical stimulus properties (e.g., size and luminance) and internal processing (e.g., attention) influence N1 (Johannes, Munte, Heinze, & Mangun, 1995). Some previous studies have reported the inverse relationship between N1 and the prestimulus alpha activity (Basar & Stampfer, 1985; Rahn & Basar, 1993), suggesting that N1 is reduced by the functional inhibition of sensory processing concerning flankers associated with alpha activity. Thus, it is predictable that N1 would be enhanced in trials preceded by compatible stimuli as compared with those preceded by incompatible stimuli, as the visual area of flankers is not inhibited before stimulus presentation. The main purpose of this study is to elucidate the change in cognitive control across conflict adaptation. Thus, we examine the difference of alpha activity and N1 between previous trial compatibilities. Furthermore, we estimate the source of alpha activity and N1 associated with conflict adaptation using standardized Low Resolution Brain Electromagnetic Tomography (sLORETA) (Pascual-Marqui, 2002). 2. Materials and methods 2.1. Participants Twenty-one graduate and undergraduate students (seven female; mean age of 24.16 ± 2.54 years, age range: 20–29 years) voluntarily participated in the experiment. All participants gave written informed consent. All except four were right-handed, and all had normal or corrected-to-normal vision. They did not report any neurological and/or psychiatric illness. This study was approved by the ethics committee of the graduate school of psychology at Rissho University. 2.2. Stimulus and procedure A modified arrow version of the flanker task was employed using STIM2 software (NeuroScan, Inc.). In the task, participants were required to press the right or left button with their thumb corresponding to the direction of a target arrow presented in the center of the screen. The target arrow was flanked by four arrows. The stimulus types were classified into compatible stimuli, where the directions of the central arrow and flanker arrows were the same (i.e., <<<<< and >>>>>), and incompatible stimuli, where they were different (i.e., >><>> and <<><<). The arrows were approximately 1° tall and 0.9° wide (total = 5.85°), and were presented for 100 ms. Stimulus onset asynchrony (SOA) varied randomly between 2500 and 3000 ms (in 100 ms increments). The same compatibility was repeated 3–5 times. Participants performed 24 blocks, after two practice blocks. The compatible stimulus was always presented in the beginning trial in each block. Response directions (left, right), and the number of times the same
9
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
(a)
Present trial compatibility
Incompatible
0
O2 Power (μ V^2 )
40
0
O1 Power (μ V^2 )
50
0
Fz Power (μ V^2 )
10
Compatible
0
5
10
15
20
25
30
Frequency (Hz)
0
5
10
iC - cC
iC - cC
iI - cI -0.5
20
25
30
Alpha2
Alpha1
(b)
15
Frequency (Hz) Previous trial compatibility Compatible Incompatible
0
0.5
(c)
iI - cI -2.5
0
2.5
Power (μ V^2 )
Fig. 1. Results of frequency analysis: (a) Grand averaged power spectrums for each condition at Fz, O1, and O2, (b) Grand averaged topographies of difference between power of each frequency band in trials preceded by the compatible stimulus and those preceded by the incompatible stimulus for both stimuli, and (c) sLORETA images focusing on the largest difference between current densities of alpha2 (10–13 Hz) in trials preceded by the compatible stimulus and those preceded by incompatible stimulus.
compatibility was repeated (3–5) were equally and randomly arranged in a block (72 trials). Participants were instructed to respond as fast as possible, while maintaining a low error rate. 2.3. EEG recordings Electroencephalogram (EEG) and electrooculograms (EOGs) were recorded using easycaps (EASYCAP Inc.) and NUAMPS
(NeuroScan Inc.). Data were collected from 29 scalp electrodes (Fz, FCz, Cz, CPz, Pz, Fp1, Fp2, F7, F3, F4, F8, FC5, FC1, FC2, FC6, T7, C3, C4, T8, CP5, CP1, CP2, CP6, P7, P3, P4, P8, O1, and O2), and electrodes placed on the left and right earlobes, above and below the left eye, and at the outer canthi of both eyes. The recordings were referenced to the nose tip, grounded at AFz, and were sampled at 500 Hz (0.1–80 Hz band-pass). Electrode impedance was less than 5 KX.
10
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
Table 2 Significant differences between alpha2 (10–13 Hz) current densities in trials preceded by compatible stimulus and those preceded by incompatible stimulus in sLORETA (compatible < incompatible, two-tailed p < .01). Region
X(MNI)
Superior frontal gyrus Superior frontal gyrus Medial frontal gyrus Superior frontal gyrus Medial frontal gyrus
15 10 10 20 15
Y(MNI)
Z(MNI)
BA
35 40 30 40 30
55 50 45 50 45
8 8 8 8 8
MNI: Montreal Neurological Institute, BA: Broadmann area.
between two conditions. For each comparison, sLORETA software computed the log of the ratio of averages with 5000 random permutations, and significance levels were corrected. 2.8. Correlational analysis To examine the relationship between oscillatory activity and N1, we calculated the indices of the change of activity between previous trial compatibilities as the sum of iC and iI divided by the sum of cC and cI of powers of each frequency band and GFP amplitudes of N1.
2.4. Data analysis
3. Results
The analyses were conducted with MATLAB2010Ra (Mathworks Inc.) and EEGLAB (Delorme & Makeig, 2004). Offline EEG was lowpass filtered at 50 Hz. Epochs were extracted from 1000 ms before to 1000 ms after stimulus onset. Baseline corrections were separately conducted for prestimulus oscillatory and ERP analysis (see below). We rejected epochs that included artifacts such as eye blinking and eye movement by visual inspection after epochs containing ±50 lV were automatically rejected. The remaining epochs were re-referenced to a common average reference.
3.1. Behavioral results
2.5. Prestimulus oscillatory analysis Epochs of prestimulus interval (1000 ms) were used for power spectrum analysis. A Hanning window with a 10% taper was applied to the interval. Then, a Fast Fourier Transform (FFT) was used to compute the power spectrum. The spectra were separately averaged according to present and previous trial compatibilities (iC, cC, cI, and iI) after epochs in the previous and present incorrect trials were excluded. The spectrum powers were integrated into two frequency bands, i.e., alpha1 (7–9 Hz), alpha2 (10–13 Hz). The mean spectrum powers of each band were then logtransformed (base 10). We used the log-transformed powers of each band at Fz, O1, and O2 for the statistical analysis. 2.6. Event-related potentials Epochs were baseline-corrected using the 100 ms prestimulus period, and were separately averaged according to conditions (iC, cC, cI, and iI) for the previous and present correct trials. We computed global field power (GFP) as the spatial standard deviation of all voltage of 29 recording site at each time point (Lehmann & Skrandies, 1980). N1 components were identified as GFP peaks with posterior negative scalp distribution between 100 and 200 ms after stimulus onset. The peak GFP amplitude of N1 was used for statistical analysis. In addition, we statistically analyzed ERP amplitudes of the negative peak from 100 to 200 ms at O1 and O2, i.e., an alternative measure of N1. 2.7. Standardized Low Resolution Tomography Analysis (sLORETA) We used sLORETA software (Pascual-Marqui, 2002) to estimate current source densities corresponding to each frequency band and N1. The sLORETA method solves the inverse problem under the assumption that neighboring neuronal sources should have similar electrical activity. Current source densities were calculated for each of the 6239 voxels at 5 mm spatial resolution. The voxels were arranged in the gray matter and the hippocampus according to the Montreal Neurological Institute coordinates corrected to the Talairach coordinates. To clarify difference activity corresponding to each frequency band and N1 between conditions, paired t-tests were used to compare current source density in each voxel
The first five trials in each block and trials following incorrect trials were excluded from analysis, because they were thought to affect the RT. The means of correct RTs and incorrect response rates are shown in Table 1. As expected, the mean correct RT was longer in cI than in iI trials, whereas it was shortened by cC as compared with iC trials. This observation was confirmed with a repeatedmeasures analysis of variance (ANOVA) including the present and previous trial compatibilities. Degrees of freedom were adjusted using Greenhouse–Geisser correction where appropriate. There were significant main effects of the present trial compatibility, F (1, 20) = 79.57, p < .001, e = 0.412, and a significant interaction between present and previous trial compatibilities, F(1, 20) = 9.22, p < .01, e = 0.412. A simple effect analysis showed significant main effects of the present trial compatibility at trials preceded by both compatible and incompatible stimuli, F(1, 20) = 54.73, p < .001; F (1, 20) = 102.23, p < .001, and the main effects of the previous trial compatibility at both present trial compatibilities’ conditions, F (1, 20) = 12.07, p < .01; F(1, 20) = 5.92, p < .05. The results showed that the mean correct RTs was significantly longer for iC than cC trials, and it was significantly shorter for iI than cI trials. Regarding incorrect response rates, a repeated measures ANOVA was conducted for compatibilities between present and previous trials. There were significant main effects of present and previous trial compatibilities, F(1, 20) = 90.24, p < .001, e = 0.54; F (1, 20) = 21.22, p < .001, e = 0.54, and significant interaction between them, F(1, 20) = 44.28, p < .001, e = 0.54. A simple effect analysis showed that the incorrect response rate was significantly higher for cI than for iI trials, F(1, 20) = 56.55, p < .001, and it was significantly higher for an incompatible stimulus than for a compatible stimulus in the present trial preceded by both a compatible and an incompatible stimulus, F(1, 20) = 116.37, p < .001; F(1, 20) = 27.59, p < .001. 3.2. Physiological data Five participants, who were not modulated by our experimental setting (i.e., an ineffective group), had positive value of subtraction scores of correct RTs on the ‘‘compatibility repetitive” trials (cC and iI) from the ‘‘compatibility switching” trials (iC and cI). Their physiological data were analyzed apart from other 16 participants, because their results seemed to be different from the others (Wang et al., 2015). 3.2.1. Prestimulus oscillations Fig. 1a shows power spectrums on the previous correct cC, iC, cI, and iI trials following correct trials at Fz, O1, and O2. We observed that alpha activity was enhanced in trials preceded by incompatible stimuli compared with those preceded by compatible stimuli. The topographical map showed that the difference of alpha activity
11
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
Present trial compatibility
(a)
Incompatible
5
O2 (μV)
-10
5
O1 (μV)
-10
0
GFP (μV)
4
Compatible
−100
0
100
200
300
(b) 8 (μV)
iC
cC
400
−100
0
100
200
300
400
Previous trial compatibility Compatible Incompatible cI iI
0
-8
(c)
Fig. 2. Results of N1: (a) Grand average GFP and ERP (O1, O2) waveforms for each condition, (b) topographies of the N1 peak for each condition, and (c) sLORETA images focusing on the largest difference between current densities of N1 for incompatible stimulus preceded by compatible stimulus and those preceded by incompatible stimulus.
between previous trial compatibilities were larger at frontal and occipital area (Fig. 1b). The observation was confirmed by a repeated measures ANOVA with the present and previous trials compatibilities on log-transformed powers of alpha1 and alpha2 at Fz, O1, and O2. We found significant main effects of the previous trial compatibilities on log-transformed powers of alpha1, alpha2 at Fz (F(1, 15) = 4.57, p < .05, e = 0.57; F(1, 15) = 9.93, p < .01,
e = 0.75), O1 (F(1, 15) = 16.35, p < .01, e = 0.79; F(1, 15) = 7.60, p < .05, e = 0.67), and O2 (F(1, 15) = 19.59, p < .001, e = 0.91; F (1, 15) = 11.10, p < .01, e = 0.65). We conducted sLORETA statistical comparisons between previous trial compatibilities for each frequency band. This analysis revealed that significant increases in alpha2 current densities in the superior/medial frontal gyrus (BA8, two-tailed p < .01; see
12
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
previous trial compatibilities, F(1, 15) = 12.28, p < .05, e = 0.30. There was a significant main effect of electrodes, meaning that ERP amplitudes were larger at O1 than at O2. Statistical comparisons using sLORETA were performed on N1 components (130–150 ms) between previous trial compatibilities in each present trial compatibility condition. Contrast iI vs cI trials showed a significant decrease in current density of the right inferior frontal gyrus (BA9, two-tailed p < .05; see Fig. 2c and Table 3). We could not find any significant increases or decreases in current densities for the contrast iC vs CC trials (one-tailed p > .1).
Table 3 Significant differences between N1 current densities for incompatible stimulus preceded by compatible stimulus and those preceded by incompatible stimulus in sLORETA (compatible > incompatible, two-tailed p < .05). Region
X(MNI)
Y(MNI)
Z(MNI)
BA
Inferior frontal gyrus Inferior frontal gyrus
50 55
10 10
35 40
9 9
MNI: Montreal Neurological Institute, BA: Broadmann area.
Fig. 1c and Table 2) in trials preceded by incompatible stimuli as compared with those preceded by compatible stimuli. There were no significant increases or decreases in current densities of alpha1 (one-tailed p > .1).
3.2.3. The ineffective group Fig. 3 represents power spectrums and GFP and ERP waveforms on cC, iC, cI and iI trials in the ineffective group. A repeated measures ANOVAs with groups and the present and previous trial compatibilities were conducted on log-transformed powers of each frequency band (Fz, O1, O2), and GFP amplitudes of N1. We found a significant interaction between groups and the previous trial compatibility on GFP amplitudes of N1, F(1, 19) = 5.13, p < .05, e = 0.76. There was no significant main effect of previous trial compatibility in the ineffective group, F(1, 4) = 1.15, p = .34, F(1, 4) = 0.42, p = .65. There were no significant interactions between groups and previous trial compatibility in alpha activities.
3.2.2. Event-related potentials Fig. 2a shows grand average GFP waveforms and ERP waveforms at O1 and O2 on the previous correct cC, iC, cI, and iI trials following correct trials. A larger GFP amplitude around 140 ms with posterior negative topography (Fig. 2b) was observed in trials preceded by compatible stimuli more than in those preceded by incompatible stimuli. The effect on GFP amplitudes of N1 were evaluated by ANOVA with present and previous trial compatibilities. We found a significant main effect of the previous trial compatibilities, F(1, 15) = 6.51, p < .05, e = 0.81. Consistent with the results of GFP, ANOVA with present and previous trial compatibilities and electrodes (O1, O2) showed a significant main effects of
Present trial compatibility
(a)
Incompatible
10
Compatible
50 40 0
O2 Power ( μV^2 )
0
O1 Power (μV^2 )
0
Fz Power (μV^2 )
3.2.4. The relationship between oscillatory activity and N1 The Pearson’s correlation test (n = 21) revealed that the index, i.e., (iC + iI)/(cC + cI) of alpha2 activity at Fz was negatively
0
5
10
15
20
25
30
0
5
10
15
20
25
30
300
400
Frequency (Hz)
Frequency (Hz)
0
GFP (μV)
4
(b)
−100
0
100
200
time (ms)
300
400
−100
0
100
200
time (ms) Previous trial compatibility Compatible Incompatible
Fig. 3. Results of the ineffective group (n = 5): (a) Grand averaged power spectrums for each condition at Fz, O1, O2, and (b) grand average GFP waveforms for each condition.
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
correlated with the GFP index (r = .52, p < .05), showing that participants with a larger increase in alpha2 activity at Fz tended to reduce GFP amplitude of N1 in trials preceded by incompatible stimuli. There were no significant correlations of other frequency indices with GFP index.
4. Discussion We found that posterior alpha activity was increased for trials preceded by incompatible stimuli more than for those preceded by compatible stimuli. Previous studies have shown that a high degree of conflict is associated with an increase in alpha activities (Min & Park, 2010), whereas a decrease in posterior alpha activities has been found in trials preceded by compatible stimuli in the Stroop task (Compton et al., 2012). These findings suggest that proactive control for the visual area is reflected in the increase in posterior alpha activities. A difference of current densities was evident for alpha2 band in the superior/medial frontal cortex (BA8) between previous trial compatibilities. These regions have been associated with the regulation of posterior alpha activity and the visual spatial attention (Capotosto et al., 2009; Sauseng et al., 2011). Theoretical work suggests that an incompatible stimulus enhances the inhibition of irrelevant sensory input as compared to a compatible stimulus (Botvinick et al., 2001). A previous study showed that high activity in these areas is associated with a high degree of alpha lateralization by the attendant visual field (Liu et al., 2014). Therefore, we consider that alpha2 current densities in the superior/medial frontal cortex are associated with proactive attentional control triggered by an incompatible stimulus in a previous trial. Moreover, we found a decrease in the N1 amplitude in trials preceded by an incompatible stimulus more often than in those preceded by a compatible one. A similar N1 modulation was observed in a figure of a previous study (Scerif, Worden, Davidson, Seiger, & Casey, 2006), though we could not confirm the accuracy of the result because N1 was not analyzed. It is assumed that proactive control has enhanced information processing on the target, leading to an increase in N1, while it has reduced information processing on flankers, leading to a decrease in N1. Therefore, we conclude that N1 is reduced in trials preceded by an incompatible stimulus since the physical property of stimulus (i.e., size) was larger in flankers (four arrows) than in targets (one arrow). The right inferior frontal cortex is considered to play a role in various types of inhibition such as regulating interference and task switching (Aron, Robbins, & Poldrack, 2004). Previous fMRI studies have also shown higher activities of the right inferior frontal cortex for incompatible stimuli than for compatible stimuli in the flanker task (Hazeltine, Bunge, Scanlon, & Gabrieli, 2003; Hazeltine, Poldrack, & Gabrieli, 2000). Consistent with this view, current densities of N1 were significantly larger in cI than in iI trials in the right inferior frontal cortex, whereas there is no significant difference of current densities in the region between iC and cC, in which it is unnecessary to inhibit information processing on flankers. We suggest that the right inferior frontal cortex is required to be transiently activated after stimulus in a cI trial in comparison to iI trials, as information processing about flankers is enforced in trials preceded by compatible stimuli, which is associated with the high demands of inhibitory control. A difference of alpha activities between previous trial compatibilities was also found at the frontal area (Fz). We suggested that frontal alpha activity was derived from frontal brain areas such as superior/medial frontal cortex. Thus, frontal alpha activity might be more directly associated with the proactive attentional control than posterior alpha activities. Consistent with this speculation,
13
in this study, participants with the larger increase in alpha2 activity at Fz tended to have a larger decrease in N1 in trials preceded by incompatible stimuli, whereas there were no significant correlations between indices of alpha and N1 at posterior electrodes. The previous study showed a shift between the modes of proactive control and reactive control (Braver et al., 2009). These findings suggested an inverse relationship between proactive control and reactive control. Our results indicate two cognitive control systems resolving the interference of incompatible stimulus in the flanker task. We suggest that the right inferior frontal cortex plays the role of the reactive control to transiently inhibit the information processing about flankers after stimulus presentation. Furthermore, after responding to an incompatible stimulus, the superior/medial frontal cortex is activated to preliminarily attend on the focal area of the visual field based on previous trial compatibility, which alleviates the demands of the right inferior frontal cortex activity. Wang et al. (2015) also reported that high connectivity of the posterior parietal cortex to the superior frontal cortex is observed in participants with a high conflict adaptation effect, whereas a low conflict adaptation effect is associated with high connectivity of the anterior cingulate cortex to the insula and the occipital cortex. The result implies that proactive control is dominant and reactive control is inferior when participants adapt to conflict of stimulus. Therefore, we suggest that conflict adaptation is associated with the transition of two temporarily distinct cognitive control systems. However, in terms of alpha activity, an inconsistent result has been reported in a study of the Stroop task, i.e., a decrease in posterior alpha activities has been observed in trials preceded by incompatible stimuli (Compton, Arnstein, Freedman, Dainer-Best, & Liss, 2011). A previous study showed that a decrease in posterior alpha activities predicts good performance on visual perceptual discrimination (Babiloni et al., 2006; Ergenoglu et al., 2004; Hanslmayr et al., 2007). Thus, it is thought that, in the Stroop task, the presentation of an incompatible stimulus leads to a decrease in alpha activities in order to precisely perceive a stimulus because word and color are presented as an identical object, while it leads to an increase in alpha activities to reduce sensory input in the irrelevant visual field since the target and flankers are presented separately on the visual field in the flanker task. We were not able to find a conflict adaptation effect on behavioral performance in five of the participants (i.e., the inefficient group). In the ineffective group, N1 is not modulated by the previous trial compatibility. Thus, we suggest that their strategies were different from the strategies of other participants, e.g., a sustained attention focusing on the target regardless of compatible stimulus. However, there is no significant interaction between the previous trial compatibility and groups for alpha2 bands. Although this result indicates that a modulation of alpha2 activity was not related to conflict adaptation, we were not able to resolve these issues, since the ineffective group had only five members, and it was possible that strategies differed within the ineffective group. Atypical susceptibility to previous trial compatibility is reported in various neurological and psychiatric disorders, e.g., autism spectrum disorders (Larson, South, Clayson, & Clawson, 2012) and obsessive–compulsive disorders (Liu, Gehring, Weissman, Taylor, & Fitzgerald, 2012). We hope that further studies with such disorder groups will clarify the relationship between prestimulus oscillatory activity and conflict adaptation. In this study, the same type of compatibility was repeated at least thrice. Thus, it is possible that conflict adaptation is associated with a prediction based on a repeating number. Although the modulation of alpha activity was reduced in trials after repeating the same type of compatibility thrice (i.e., cI and iC) if the prediction contributed to conflict adaptation, differences between trials of the same previous compatibility (i.e., iI vs iC and cC vs
14
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
cI) were not found in this study. Hence, we concluded that the change in alpha activity was triggered by previous trial compatibility rather than by prediction. 5. Conclusion We examined neural activities involving conflict adaptation. Before stimulus presentation, frontal and posterior alpha activities were enhanced in trials preceded by incompatible stimuli, in comparison with those preceded by compatible stimuli. The source of alpha2 activities was estimated to be located in the superior/medial frontal cortex, which is suggested to be associated with proactive attentional control to reduce the sensory input on flankers. Following the stimulus presentation, N1 was reduced in iI more than in cI trials, where the difference of current densities was found in the right inferior frontal cortex. We consider that the small N1 amplitude reflects a decrease in demands of transient inhibition by the proactive attention control. Moreover, we showed the inverse relationship between the changes of the frontal alpha2 activity and N1 by previous trial compatibilities. These findings suggest that conflict adaptation is associated with the transition from reactive control to proactive control. References Aron, A. R., Robbins, T. W., & Poldrack, R. A. (2004). Inhibition and the right inferior frontal cortex. Trends in Cognitive Sciences, 8(4), 170–177. Babiloni, C., Vecchio, F., Bultrini, A., Luca Romani, G., & Rossini, P. M. (2006). Pre- and poststimulus alpha rhythms are related to conscious visual perception: A highresolution EEG study. Cerebral Cortex, 16(12), 1690–1700. Basar, E., & Stampfer, H. G. (1985). Important associations among EEG-dynamics, event-related potentials, short-term memory and learning. The International Journal of Neuroscience, 26(3–4), 161–180. Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001). Conflict monitoring and cognitive control. Psychological Review, 108(3), 624–652. Braver, T. S. (2012). The variable nature of cognitive control: A dual mechanisms framework. Trends in Cognitive Sciences, 16(2), 106–113. Braver, T. S., Paxton, J. L., Locke, H. S., & Barch, D. M. (2009). Flexible neural mechanisms of cognitive control within human prefrontal cortex. Proceedings of the National Academy of Sciences, 106(18), 7351–7356. Capotosto, P., Babiloni, C., Romani, G. L., & Corbetta, M. (2009). Frontoparietal cortex controls spatial attention through modulation of anticipatory alpha rhythms. The Journal of Neuroscience, 29(18), 5863–5872. Compton, R. J., Arnstein, D., Freedman, G., Dainer-Best, J., & Liss, A. (2011). Cognitive control in the intertrial interval: Evidence from EEG alpha power. Psychophysiology, 48(5), 583–590. Compton, R. J., Huber, E., Levinson, A. R., & Zheutlin, A. (2012). Is ‘‘conflict adaptation” driven by conflict? Behavioral and EEG evidence for the underappreciated role of congruent trials. Psychophysiology, 49(5), 583–589. Correa, Á., Rao, A., & Nobre, A. C. (2009). Anticipating conflict facilitates controlled stimulus-response selection. Journal of Cognitive Neuroscience, 21(8), 1461–1472. Delorme, A., & Makeig, S. (2004). EEGLAB: An open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. Journal of Neuroscience Methods, 134(1), 9–21. Ergenoglu, T., Demiralp, T., Bayraktaroglu, Z., Ergen, M., Beydagi, H., & Uresin, Y. (2004). Alpha rhythm of the EEG modulates visual detection performance in humans. Cognitive Brain Research, 20(3), 376–383. Eriksen, C. W., & Eriksen, B. A. (1979). Target redundancy in visual search: Do repetitions of the target within the display impair processing? Perception & Psychophysics, 26(3), 195–205. Fabiani, M., & Gratton, G. (2007). Event-related brain potentials: Methods, theory, and applications. In J. T. Cacioppo, L. G. Tassinary, & G. G. Berntson (Eds.), Handbook of psychophysiology (pp. 53–84). New York: Cambridge University Press. Freunberger, R., Holler, Y., Griesmayr, B., Gruber, W., Sauseng, P., & Klimesch, W. (2008). Functional similarities between the P1 component and alpha oscillations. European Journal of Neuroscience, 27(9), 2330–2340. Gratton, G., Coles, M. G., & Donchin, E. (1992). Optimizing the use of information: Strategic control of activation of responses. Journal of Experimental Psychology. General, 121(4), 480–506.
Hanslmayr, S., Aslan, A., Staudigl, T., Klimesch, W., Herrmann, C. S., & Bauml, K. H. (2007). Prestimulus oscillations predict visual perception performance between and within subjects. NeuroImage, 37(4), 1465–1473. Hazeltine, E., Bunge, S. A., Scanlon, M. D., & Gabrieli, J. D. (2003). Materialdependent and material-independent selection processes in the frontal and parietal lobes: An event-related fMRI investigation of response competition. Neuropsychologia, 41(9), 1208–1217. Hazeltine, E., Poldrack, R., & Gabrieli, J. D. (2000). Neural activation during response competition. Journal Cognitive Neuroscience, 12(Suppl 2), 118–129. Hillyard, S. A., Teder-Salejarvi, W. A., & Munte, T. F. (1998). Temporal dynamics of early perceptual processing. Current Opinion in Neurobiology, 8(2), 202–210. Johannes, S., Munte, T. F., Heinze, H. J., & Mangun, G. R. (1995). Luminance and spatial attention effects on early visual processing. Cognitive Brain Research, 2 (3), 189–205. Kerns, J. G., Cohen, J. D., MacDonald, A. W., 3rd, Cho, R. Y., Stenger, V. A., & Carter, C. S. (2004). Anterior cingulate conflict monitoring and adjustments in control. Science, 303(5660), 1023–1026. Klimesch, W., Sauseng, P., & Hanslmayr, S. (2007). EEG alpha oscillations: The inhibition-timing hypothesis. Brain Research Reviews, 53(1), 63–88. Larson, M. J., South, M., Clayson, P. E., & Clawson, A. (2012). Cognitive control and conflict adaptation in youth with high-functioning autism. Jounal of Child Psychology and Psychiatry, 53(4), 440–448. Lehmann, D., & Skrandies, W. (1980). Reference-free identification of components of checkerboard-evoked multichannel potential fields. Electroencephalography and Clinical Neurophysiology, 48(6), 609–621. Liu, Y., Bengson, J., Huang, H., Mangun, G. R., & Ding, M. (2014). Top-down modulation of neural activity in anticipatory visual attention: Control mechanisms revealed by simultaneous EEG-fMRI. Cerebral Cortex. Liu, Y., Gehring, W. J., Weissman, D. H., Taylor, S. F., & Fitzgerald, K. D. (2012). Trialby-trial adjustments of cognitive control following errors and response conflict are altered in pediatric obsessive compulsive disorder. Frontiers in Psychiatry, 3, 41. Min, B. K., & Park, H. J. (2010). Task-related modulation of anterior theta and posterior alpha EEG reflects top-down preparation. BMC Neuroscience, 11, 79. Pascual-Marqui, R. D. (2002). Standardized low-resolution brain electromagnetic tomography (sLORETA): Technical details. Methods and Findings in Experimental and Clinical Pharmacology, 24(Suppl D), 5–12. Rahn, E., & Basar, E. (1993). Enhancement of visual evoked potentials by stimulation during low prestimulus EEG stages. International Journal of Neuroscience, 72(1– 2), 123–136. Rana, K. D., & Vaina, L. M. (2014). Functional roles of 10 Hz alpha-band power modulating engagement and disengagement of cortical networks in a complex visual motion task. PLoS ONE, 9(10), e107715. Rihs, T. A., Michel, C. M., & Thut, G. (2007). Mechanisms of selective inhibition in visual spatial attention are indexed by alpha-band EEG synchronization. European Journal of Neuroscience, 25(2), 603–610. Rihs, T. A., Michel, C. M., & Thut, G. (2009). A bias for posterior alpha-band power suppression versus enhancement during shifting versus maintenance of spatial attention. NeuroImage, 44(1), 190–199. Sauseng, P., Feldheim, J. F., Freunberger, R., & Hummel, F. C. (2011). Right prefrontal TMS disrupts interregional anticipatory EEG alpha activity during shifting of visuospatial attention. Frontiers in Psychology, 2, 241. Sauseng, P., Klimesch, W., Stadler, W., Schabus, M., Doppelmayr, M., Hanslmayr, S., et al. (2005). A shift of visual spatial attention is selectively associated with human EEG alpha activity. European Journal of Neuroscience, 22(11), 2917–2926. Scerif, G., Worden, M. S., Davidson, M., Seiger, L., & Casey, B. J. (2006). Context modulates early stimulus processing when resolving stimulus-response conflict. Journal of Cognitive Neuroscience, 18(5), 781–792. Simon, J. R. (1969). Reactions toward the source of stimulation. Journal of Experimental Psychology, 81(1), 174–176. Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18(6), 643–662. Thut, G., Nietzel, A., Brandt, S. A., & Pascual-Leone, A. (2006). Alpha-band electroencephalographic activity over occipital cortex indexes visuospatial attention bias and predicts visual target detection. The Journal of Neuroscience, 26(37), 9494–9502. Wang, X., Wang, T., Chen, Z., Hitchman, G., Liu, Y., & Chen, A. (2015). Functional connectivity patterns reflect individual differences in conflict adaptation. Neuropsychologia, 70, 177–184. Worden, M. S., Foxe, J. J., Wang, N., & Simpson, G. V. (2000). Anticipatory biasing of visuospatial attention indexed by retinotopically specific alpha-band electroencephalography increases over occipital cortex. The Journal of Neuroscience, 20(6), RC63. Yeung, N., Botvinick, M. M., & Cohen, J. D. (2004). The neural basis of error detection: Conflict monitoring and the error-related negativity. Psychological Review, 111 (4), 931–959.
Contents lists available at ScienceDirect
Brain and Cognition journal homepage: www.elsevier.com/locate/b&c
Transition from reactive control to proactive control across conflict adaptation: An sLORETA study Kota Suzuki a,⇑, Haruo Shinoda b a Department of Developmental Disorders, National Institute of Mental Health, National Center of Neurology and Psychiatry (NCNP), 4-1-1, Ogawahigashi, Kodaira, Tokyo 187-8553, Japan b Department of Clinical Psychology, Faculty of Psychology, Rissho University, Osaki 4-2-16, Shinagawa, Tokyo 141-8602, Japan
a r t i c l e
i n f o
Article history: Received 23 June 2015 Revised 28 August 2015 Accepted 10 September 2015
Keywords: Conflict adaptation Standardized Low Resolution Brain Electromagnetic Tomography Alpha oscillation N1
a b s t r a c t In a flanker task, behavioral performance is modulated by previous trial compatibility (i.e., conflict adaptation); a longer response time (RT) is found for a compatible stimulus preceded by an incompatible stimulus than by a compatible stimulus, whereas a shorter RT is found for an incompatible stimulus preceded by an incompatible stimulus than by a compatible stimulus. We examined the temporal characteristics of cognitive control across conflict adaptation using prestimulus electroencephalogram oscillatory activity and an event-related potential component, N1. Prestimulus frontal (Fz) and posterior (O1 and O2) alpha1 (7–9 Hz) and alpha2 (10–13 Hz) activities were enhanced in trials preceded by incompatible stimuli more than those preceded by compatible stimuli. Furthermore, there were significant differences of alpha2 current densities between previous trial compatibilities in the superior/medial frontal cortex. We suggested that the modulation of alpha activity by previous trial compatibility was associated with proactive attentional control. N1 amplitude was decreased in trials preceded by incompatible stimuli more than in those preceded by compatible stimuli. N1 current densities in the right inferior frontal cortex were smaller for an incompatible stimulus preceded by an incompatible stimulus than those preceded by a compatible stimulus, suggesting that demands of transient cognitive control induced by an incompatible stimulus were decreased by the proactive control. Moreover, correlational analysis showed that participants with a larger increase in alpha2 activity tended to have a larger decrease in N1 in trials preceded by incompatible stimulus. These findings revealed that the manner of cognitive control for the incompatible stimulus was transited from reactive control to proactive control across conflict adaptation. Ó 2015 Elsevier Inc. All rights reserved.
1. Introduction It is important for adaption to the environment to be able to flexibly change the manner of cognitive control according to the context. Types of cognitive control are temporally distinguished into at least two cognitive controls (Braver, 2012; Braver, Paxton, Locke, & Barch, 2009),1 i.e., ‘‘proactive” control and ‘‘reactive” control. Proactive control represents anticipatory and preparatory attentional control for the interference of the upcoming stimulus. Reactive control is postulated as transient recruitment of cognitive control after the stimulus presentation. ⇑ Corresponding author. E-mail addresses: [email protected] (K. Suzuki), [email protected] (H. Shinoda). In the context of studies regarding conflict adaptation (e.g., Correa, Rao, & Nobre, 2009), proactive control refers to cognitive control triggered by a cue informing the upcoming stimulus compatibility, whereas reactive control refers to cognitive control triggered by the previous trial compatibility. In this study, terms ‘‘proactive” and ‘‘reactive” control were used in the broader definition, in which we assumed that proactive and reactive control was triggered by previous trial compatibilities. 1
http://dx.doi.org/10.1016/j.bandc.2015.09.001 0278-2626/Ó 2015 Elsevier Inc. All rights reserved.
Cognitive control has been well-studied using interference tasks such as the flanker task (Eriksen & Eriksen, 1979), the Stroop task (Stroop, 1935), and the Simon task (Simon, 1969). For example, in the flanker task, participants are required to respond to targets presented in the central visual field, while flankers appear that aid (i.e., a compatible stimulus) or interfere with (i.e., an incompatible stimulus) participants’ accomplishment of the task. Gratton, Coles, and Donchin (1992) reported that shorter response times (RTs) and higher accuracies were observed for a compatible stimulus preceded by a compatible stimulus (cC) than by an incompatible stimulus (iC), whereas longer RTs and lower accuracies were observed for an incompatible stimulus preceded by a compatible stimulus (cI) than by an incompatible stimulus (iI). This phenomenon is called conflict adaptation (or the Gratton effect). Conflict has been used to refer to the competition between target information and irrelevant information induced by an incompatible stimulus (Botvinick, Braver, Barch, Carter, & Cohen, 2001). Researchers have interpreted conflict adaptation as a cognitive control mechanism driven by the degree of conflict
8
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
Table 1 Means (standard deviation) of correct RTs and incorrect response rates classified by present and previous trial compatibility. Present trial compatibility Previous trial compatibility
Compatible
Correct RT (ms)
363.41 (32.57) 4.38 (7.87)
Incorrect response rate (%)
Incompatible
Compatible Incompatible Compatible Incompatible 371.92 (40.96) 5.69 (12.80)
428.14 (65.94) 25.93 (8.25)
417.17 (48.18) 14.73 (7.05)
(Botvinick et al., 2001). Then, it is hypothesized that a high degree of conflict (i.e., induced by an incompatible stimulus) causes the attention to be focused on the central visual field in order to inhibit the interference of flankers on the next trial, whereas a low degree of conflict (i.e., induced by a compatible stimulus) makes the attentional area broaden in order to utilize information about flankers on the next trial (Botvinick et al., 2001; Kerns et al., 2004). In accordance with this hypothesis, cognitive control is transiently required to resolve a high degree of conflict when an incompatible stimulus is presented after a compatible stimulus (Wang et al., 2015). On the other hand, when responding to an incompatible stimulus preceded by an incompatible stimulus, cognitive control is preliminarily executed before the presentation of the stimulus. Thus, we postulate that conflict adaptation is associated with a transition from reactive control to proactive control. An oscillation of the electroencephalogram (EEG) or magnetoencephalogram (MEG) of approximately 10 Hz (i.e., alpha) is sensitive to the proactive attentional state before the stimulus presentation. Typically, enhanced alpha activity in a sensory area is interpreted as functional inhibition (Klimesch, Sauseng, & Hanslmayr, 2007). Consistent with this view, enhanced posterior alpha activity predicted a decrease in the visual perceptual performance (Babiloni, Vecchio, Bultrini, Luca Romani, & Rossini, 2006; Ergenoglu et al., 2004; Hanslmayr et al., 2007). Moreover, the change in alpha activity has been considered as reflecting topdown attentional control, which enhances relevant sensory inputs and inhibits irrelevant ones. Several previous studies have shown that posterior alpha activity is enhanced on the ipsilateral side of the visual field attended to, while it is attenuated at the contralateral side (Freunberger et al., 2008; Rihs, Michel, & Thut, 2007; Rihs, Michel, & Thut, 2009; Sauseng et al., 2005; Thut, Nietzel, Brandt, & Pascual-Leone, 2006; Worden, Foxe, Wang, & Simpson, 2000). In terms of interference tasks, Compton, Huber, Levinson, and Zheutlin (2012) reported that alpha activity is reduced in trials preceded by compatible stimuli in the Stroop task, as compared with neutral stimuli, suggesting that reduced alpha activity is related to the enhancement of information processing about the irrelevant aspect. In addition, Min and Park (2010) examined prestimulus alpha activity in color and shape discrimination tasks, in which participants discriminated colors or shapes of objects, ignoring other aspects. They reported that task performance was worse and posterior alpha activity was enhanced more for the shapediscrimination task than the color task, suggesting that alpha activity was enhanced for inhibition of irrelevant sensory processing. These findings suggest that proactive control for the inhibition of flankers is associated with an increase in posterior alpha activity during the prestimulus interval. Posterior alpha activity has been reported to be regulated by frontoparietal areas. Previous studies have shown that the reduced excitability in the frontal eye fields and intraparietal sulcus cause the loss of the modulation of posterior alpha activities by the attending visual field (Capotosto, Babiloni, Romani, & Corbetta, 2009; Sauseng, Feldheim, Freunberger, & Hummel, 2011). A simultaneous EEG and functional magnetic resonance imaging (fMRI)
study has shown that activity of the superior/medial frontal cortex is positively correlated with alpha lateralization by selective attention (Liu, Bengson, Huang, Mangun, & Ding, 2014). Rana and Vaina (2014) reported that alpha activity of frontoparietal areas was increased when controlling spatial attention. Thus, it is predicted that the alpha activity derived from these regions is also modulated by previous trial compatibility. We consider that N1 reflects the demands of the recruitment of cognitive control after stimulus presentation. N1 is an eventrelated potential (ERP) component observed as a negative peak from 100 to 200 ms at the occipital scalp electrode (Hillyard, Teder-Salejarvi, & Munte, 1998), preceding ERP components relevant to detection of conflict (i.e., N2 and ERN; Yeung, Botvinick, & Cohen, 2004). Typically, N1 is called a ‘‘mesogenous” component (Fabiani & Gratton, 2007). That is, both physical stimulus properties (e.g., size and luminance) and internal processing (e.g., attention) influence N1 (Johannes, Munte, Heinze, & Mangun, 1995). Some previous studies have reported the inverse relationship between N1 and the prestimulus alpha activity (Basar & Stampfer, 1985; Rahn & Basar, 1993), suggesting that N1 is reduced by the functional inhibition of sensory processing concerning flankers associated with alpha activity. Thus, it is predictable that N1 would be enhanced in trials preceded by compatible stimuli as compared with those preceded by incompatible stimuli, as the visual area of flankers is not inhibited before stimulus presentation. The main purpose of this study is to elucidate the change in cognitive control across conflict adaptation. Thus, we examine the difference of alpha activity and N1 between previous trial compatibilities. Furthermore, we estimate the source of alpha activity and N1 associated with conflict adaptation using standardized Low Resolution Brain Electromagnetic Tomography (sLORETA) (Pascual-Marqui, 2002). 2. Materials and methods 2.1. Participants Twenty-one graduate and undergraduate students (seven female; mean age of 24.16 ± 2.54 years, age range: 20–29 years) voluntarily participated in the experiment. All participants gave written informed consent. All except four were right-handed, and all had normal or corrected-to-normal vision. They did not report any neurological and/or psychiatric illness. This study was approved by the ethics committee of the graduate school of psychology at Rissho University. 2.2. Stimulus and procedure A modified arrow version of the flanker task was employed using STIM2 software (NeuroScan, Inc.). In the task, participants were required to press the right or left button with their thumb corresponding to the direction of a target arrow presented in the center of the screen. The target arrow was flanked by four arrows. The stimulus types were classified into compatible stimuli, where the directions of the central arrow and flanker arrows were the same (i.e., <<<<< and >>>>>), and incompatible stimuli, where they were different (i.e., >><>> and <<><<). The arrows were approximately 1° tall and 0.9° wide (total = 5.85°), and were presented for 100 ms. Stimulus onset asynchrony (SOA) varied randomly between 2500 and 3000 ms (in 100 ms increments). The same compatibility was repeated 3–5 times. Participants performed 24 blocks, after two practice blocks. The compatible stimulus was always presented in the beginning trial in each block. Response directions (left, right), and the number of times the same
9
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
(a)
Present trial compatibility
Incompatible
0
O2 Power (μ V^2 )
40
0
O1 Power (μ V^2 )
50
0
Fz Power (μ V^2 )
10
Compatible
0
5
10
15
20
25
30
Frequency (Hz)
0
5
10
iC - cC
iC - cC
iI - cI -0.5
20
25
30
Alpha2
Alpha1
(b)
15
Frequency (Hz) Previous trial compatibility Compatible Incompatible
0
0.5
(c)
iI - cI -2.5
0
2.5
Power (μ V^2 )
Fig. 1. Results of frequency analysis: (a) Grand averaged power spectrums for each condition at Fz, O1, and O2, (b) Grand averaged topographies of difference between power of each frequency band in trials preceded by the compatible stimulus and those preceded by the incompatible stimulus for both stimuli, and (c) sLORETA images focusing on the largest difference between current densities of alpha2 (10–13 Hz) in trials preceded by the compatible stimulus and those preceded by incompatible stimulus.
compatibility was repeated (3–5) were equally and randomly arranged in a block (72 trials). Participants were instructed to respond as fast as possible, while maintaining a low error rate. 2.3. EEG recordings Electroencephalogram (EEG) and electrooculograms (EOGs) were recorded using easycaps (EASYCAP Inc.) and NUAMPS
(NeuroScan Inc.). Data were collected from 29 scalp electrodes (Fz, FCz, Cz, CPz, Pz, Fp1, Fp2, F7, F3, F4, F8, FC5, FC1, FC2, FC6, T7, C3, C4, T8, CP5, CP1, CP2, CP6, P7, P3, P4, P8, O1, and O2), and electrodes placed on the left and right earlobes, above and below the left eye, and at the outer canthi of both eyes. The recordings were referenced to the nose tip, grounded at AFz, and were sampled at 500 Hz (0.1–80 Hz band-pass). Electrode impedance was less than 5 KX.
10
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
Table 2 Significant differences between alpha2 (10–13 Hz) current densities in trials preceded by compatible stimulus and those preceded by incompatible stimulus in sLORETA (compatible < incompatible, two-tailed p < .01). Region
X(MNI)
Superior frontal gyrus Superior frontal gyrus Medial frontal gyrus Superior frontal gyrus Medial frontal gyrus
15 10 10 20 15
Y(MNI)
Z(MNI)
BA
35 40 30 40 30
55 50 45 50 45
8 8 8 8 8
MNI: Montreal Neurological Institute, BA: Broadmann area.
between two conditions. For each comparison, sLORETA software computed the log of the ratio of averages with 5000 random permutations, and significance levels were corrected. 2.8. Correlational analysis To examine the relationship between oscillatory activity and N1, we calculated the indices of the change of activity between previous trial compatibilities as the sum of iC and iI divided by the sum of cC and cI of powers of each frequency band and GFP amplitudes of N1.
2.4. Data analysis
3. Results
The analyses were conducted with MATLAB2010Ra (Mathworks Inc.) and EEGLAB (Delorme & Makeig, 2004). Offline EEG was lowpass filtered at 50 Hz. Epochs were extracted from 1000 ms before to 1000 ms after stimulus onset. Baseline corrections were separately conducted for prestimulus oscillatory and ERP analysis (see below). We rejected epochs that included artifacts such as eye blinking and eye movement by visual inspection after epochs containing ±50 lV were automatically rejected. The remaining epochs were re-referenced to a common average reference.
3.1. Behavioral results
2.5. Prestimulus oscillatory analysis Epochs of prestimulus interval (1000 ms) were used for power spectrum analysis. A Hanning window with a 10% taper was applied to the interval. Then, a Fast Fourier Transform (FFT) was used to compute the power spectrum. The spectra were separately averaged according to present and previous trial compatibilities (iC, cC, cI, and iI) after epochs in the previous and present incorrect trials were excluded. The spectrum powers were integrated into two frequency bands, i.e., alpha1 (7–9 Hz), alpha2 (10–13 Hz). The mean spectrum powers of each band were then logtransformed (base 10). We used the log-transformed powers of each band at Fz, O1, and O2 for the statistical analysis. 2.6. Event-related potentials Epochs were baseline-corrected using the 100 ms prestimulus period, and were separately averaged according to conditions (iC, cC, cI, and iI) for the previous and present correct trials. We computed global field power (GFP) as the spatial standard deviation of all voltage of 29 recording site at each time point (Lehmann & Skrandies, 1980). N1 components were identified as GFP peaks with posterior negative scalp distribution between 100 and 200 ms after stimulus onset. The peak GFP amplitude of N1 was used for statistical analysis. In addition, we statistically analyzed ERP amplitudes of the negative peak from 100 to 200 ms at O1 and O2, i.e., an alternative measure of N1. 2.7. Standardized Low Resolution Tomography Analysis (sLORETA) We used sLORETA software (Pascual-Marqui, 2002) to estimate current source densities corresponding to each frequency band and N1. The sLORETA method solves the inverse problem under the assumption that neighboring neuronal sources should have similar electrical activity. Current source densities were calculated for each of the 6239 voxels at 5 mm spatial resolution. The voxels were arranged in the gray matter and the hippocampus according to the Montreal Neurological Institute coordinates corrected to the Talairach coordinates. To clarify difference activity corresponding to each frequency band and N1 between conditions, paired t-tests were used to compare current source density in each voxel
The first five trials in each block and trials following incorrect trials were excluded from analysis, because they were thought to affect the RT. The means of correct RTs and incorrect response rates are shown in Table 1. As expected, the mean correct RT was longer in cI than in iI trials, whereas it was shortened by cC as compared with iC trials. This observation was confirmed with a repeatedmeasures analysis of variance (ANOVA) including the present and previous trial compatibilities. Degrees of freedom were adjusted using Greenhouse–Geisser correction where appropriate. There were significant main effects of the present trial compatibility, F (1, 20) = 79.57, p < .001, e = 0.412, and a significant interaction between present and previous trial compatibilities, F(1, 20) = 9.22, p < .01, e = 0.412. A simple effect analysis showed significant main effects of the present trial compatibility at trials preceded by both compatible and incompatible stimuli, F(1, 20) = 54.73, p < .001; F (1, 20) = 102.23, p < .001, and the main effects of the previous trial compatibility at both present trial compatibilities’ conditions, F (1, 20) = 12.07, p < .01; F(1, 20) = 5.92, p < .05. The results showed that the mean correct RTs was significantly longer for iC than cC trials, and it was significantly shorter for iI than cI trials. Regarding incorrect response rates, a repeated measures ANOVA was conducted for compatibilities between present and previous trials. There were significant main effects of present and previous trial compatibilities, F(1, 20) = 90.24, p < .001, e = 0.54; F (1, 20) = 21.22, p < .001, e = 0.54, and significant interaction between them, F(1, 20) = 44.28, p < .001, e = 0.54. A simple effect analysis showed that the incorrect response rate was significantly higher for cI than for iI trials, F(1, 20) = 56.55, p < .001, and it was significantly higher for an incompatible stimulus than for a compatible stimulus in the present trial preceded by both a compatible and an incompatible stimulus, F(1, 20) = 116.37, p < .001; F(1, 20) = 27.59, p < .001. 3.2. Physiological data Five participants, who were not modulated by our experimental setting (i.e., an ineffective group), had positive value of subtraction scores of correct RTs on the ‘‘compatibility repetitive” trials (cC and iI) from the ‘‘compatibility switching” trials (iC and cI). Their physiological data were analyzed apart from other 16 participants, because their results seemed to be different from the others (Wang et al., 2015). 3.2.1. Prestimulus oscillations Fig. 1a shows power spectrums on the previous correct cC, iC, cI, and iI trials following correct trials at Fz, O1, and O2. We observed that alpha activity was enhanced in trials preceded by incompatible stimuli compared with those preceded by compatible stimuli. The topographical map showed that the difference of alpha activity
11
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
Present trial compatibility
(a)
Incompatible
5
O2 (μV)
-10
5
O1 (μV)
-10
0
GFP (μV)
4
Compatible
−100
0
100
200
300
(b) 8 (μV)
iC
cC
400
−100
0
100
200
300
400
Previous trial compatibility Compatible Incompatible cI iI
0
-8
(c)
Fig. 2. Results of N1: (a) Grand average GFP and ERP (O1, O2) waveforms for each condition, (b) topographies of the N1 peak for each condition, and (c) sLORETA images focusing on the largest difference between current densities of N1 for incompatible stimulus preceded by compatible stimulus and those preceded by incompatible stimulus.
between previous trial compatibilities were larger at frontal and occipital area (Fig. 1b). The observation was confirmed by a repeated measures ANOVA with the present and previous trials compatibilities on log-transformed powers of alpha1 and alpha2 at Fz, O1, and O2. We found significant main effects of the previous trial compatibilities on log-transformed powers of alpha1, alpha2 at Fz (F(1, 15) = 4.57, p < .05, e = 0.57; F(1, 15) = 9.93, p < .01,
e = 0.75), O1 (F(1, 15) = 16.35, p < .01, e = 0.79; F(1, 15) = 7.60, p < .05, e = 0.67), and O2 (F(1, 15) = 19.59, p < .001, e = 0.91; F (1, 15) = 11.10, p < .01, e = 0.65). We conducted sLORETA statistical comparisons between previous trial compatibilities for each frequency band. This analysis revealed that significant increases in alpha2 current densities in the superior/medial frontal gyrus (BA8, two-tailed p < .01; see
12
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
previous trial compatibilities, F(1, 15) = 12.28, p < .05, e = 0.30. There was a significant main effect of electrodes, meaning that ERP amplitudes were larger at O1 than at O2. Statistical comparisons using sLORETA were performed on N1 components (130–150 ms) between previous trial compatibilities in each present trial compatibility condition. Contrast iI vs cI trials showed a significant decrease in current density of the right inferior frontal gyrus (BA9, two-tailed p < .05; see Fig. 2c and Table 3). We could not find any significant increases or decreases in current densities for the contrast iC vs CC trials (one-tailed p > .1).
Table 3 Significant differences between N1 current densities for incompatible stimulus preceded by compatible stimulus and those preceded by incompatible stimulus in sLORETA (compatible > incompatible, two-tailed p < .05). Region
X(MNI)
Y(MNI)
Z(MNI)
BA
Inferior frontal gyrus Inferior frontal gyrus
50 55
10 10
35 40
9 9
MNI: Montreal Neurological Institute, BA: Broadmann area.
Fig. 1c and Table 2) in trials preceded by incompatible stimuli as compared with those preceded by compatible stimuli. There were no significant increases or decreases in current densities of alpha1 (one-tailed p > .1).
3.2.3. The ineffective group Fig. 3 represents power spectrums and GFP and ERP waveforms on cC, iC, cI and iI trials in the ineffective group. A repeated measures ANOVAs with groups and the present and previous trial compatibilities were conducted on log-transformed powers of each frequency band (Fz, O1, O2), and GFP amplitudes of N1. We found a significant interaction between groups and the previous trial compatibility on GFP amplitudes of N1, F(1, 19) = 5.13, p < .05, e = 0.76. There was no significant main effect of previous trial compatibility in the ineffective group, F(1, 4) = 1.15, p = .34, F(1, 4) = 0.42, p = .65. There were no significant interactions between groups and previous trial compatibility in alpha activities.
3.2.2. Event-related potentials Fig. 2a shows grand average GFP waveforms and ERP waveforms at O1 and O2 on the previous correct cC, iC, cI, and iI trials following correct trials. A larger GFP amplitude around 140 ms with posterior negative topography (Fig. 2b) was observed in trials preceded by compatible stimuli more than in those preceded by incompatible stimuli. The effect on GFP amplitudes of N1 were evaluated by ANOVA with present and previous trial compatibilities. We found a significant main effect of the previous trial compatibilities, F(1, 15) = 6.51, p < .05, e = 0.81. Consistent with the results of GFP, ANOVA with present and previous trial compatibilities and electrodes (O1, O2) showed a significant main effects of
Present trial compatibility
(a)
Incompatible
10
Compatible
50 40 0
O2 Power ( μV^2 )
0
O1 Power (μV^2 )
0
Fz Power (μV^2 )
3.2.4. The relationship between oscillatory activity and N1 The Pearson’s correlation test (n = 21) revealed that the index, i.e., (iC + iI)/(cC + cI) of alpha2 activity at Fz was negatively
0
5
10
15
20
25
30
0
5
10
15
20
25
30
300
400
Frequency (Hz)
Frequency (Hz)
0
GFP (μV)
4
(b)
−100
0
100
200
time (ms)
300
400
−100
0
100
200
time (ms) Previous trial compatibility Compatible Incompatible
Fig. 3. Results of the ineffective group (n = 5): (a) Grand averaged power spectrums for each condition at Fz, O1, O2, and (b) grand average GFP waveforms for each condition.
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
correlated with the GFP index (r = .52, p < .05), showing that participants with a larger increase in alpha2 activity at Fz tended to reduce GFP amplitude of N1 in trials preceded by incompatible stimuli. There were no significant correlations of other frequency indices with GFP index.
4. Discussion We found that posterior alpha activity was increased for trials preceded by incompatible stimuli more than for those preceded by compatible stimuli. Previous studies have shown that a high degree of conflict is associated with an increase in alpha activities (Min & Park, 2010), whereas a decrease in posterior alpha activities has been found in trials preceded by compatible stimuli in the Stroop task (Compton et al., 2012). These findings suggest that proactive control for the visual area is reflected in the increase in posterior alpha activities. A difference of current densities was evident for alpha2 band in the superior/medial frontal cortex (BA8) between previous trial compatibilities. These regions have been associated with the regulation of posterior alpha activity and the visual spatial attention (Capotosto et al., 2009; Sauseng et al., 2011). Theoretical work suggests that an incompatible stimulus enhances the inhibition of irrelevant sensory input as compared to a compatible stimulus (Botvinick et al., 2001). A previous study showed that high activity in these areas is associated with a high degree of alpha lateralization by the attendant visual field (Liu et al., 2014). Therefore, we consider that alpha2 current densities in the superior/medial frontal cortex are associated with proactive attentional control triggered by an incompatible stimulus in a previous trial. Moreover, we found a decrease in the N1 amplitude in trials preceded by an incompatible stimulus more often than in those preceded by a compatible one. A similar N1 modulation was observed in a figure of a previous study (Scerif, Worden, Davidson, Seiger, & Casey, 2006), though we could not confirm the accuracy of the result because N1 was not analyzed. It is assumed that proactive control has enhanced information processing on the target, leading to an increase in N1, while it has reduced information processing on flankers, leading to a decrease in N1. Therefore, we conclude that N1 is reduced in trials preceded by an incompatible stimulus since the physical property of stimulus (i.e., size) was larger in flankers (four arrows) than in targets (one arrow). The right inferior frontal cortex is considered to play a role in various types of inhibition such as regulating interference and task switching (Aron, Robbins, & Poldrack, 2004). Previous fMRI studies have also shown higher activities of the right inferior frontal cortex for incompatible stimuli than for compatible stimuli in the flanker task (Hazeltine, Bunge, Scanlon, & Gabrieli, 2003; Hazeltine, Poldrack, & Gabrieli, 2000). Consistent with this view, current densities of N1 were significantly larger in cI than in iI trials in the right inferior frontal cortex, whereas there is no significant difference of current densities in the region between iC and cC, in which it is unnecessary to inhibit information processing on flankers. We suggest that the right inferior frontal cortex is required to be transiently activated after stimulus in a cI trial in comparison to iI trials, as information processing about flankers is enforced in trials preceded by compatible stimuli, which is associated with the high demands of inhibitory control. A difference of alpha activities between previous trial compatibilities was also found at the frontal area (Fz). We suggested that frontal alpha activity was derived from frontal brain areas such as superior/medial frontal cortex. Thus, frontal alpha activity might be more directly associated with the proactive attentional control than posterior alpha activities. Consistent with this speculation,
13
in this study, participants with the larger increase in alpha2 activity at Fz tended to have a larger decrease in N1 in trials preceded by incompatible stimuli, whereas there were no significant correlations between indices of alpha and N1 at posterior electrodes. The previous study showed a shift between the modes of proactive control and reactive control (Braver et al., 2009). These findings suggested an inverse relationship between proactive control and reactive control. Our results indicate two cognitive control systems resolving the interference of incompatible stimulus in the flanker task. We suggest that the right inferior frontal cortex plays the role of the reactive control to transiently inhibit the information processing about flankers after stimulus presentation. Furthermore, after responding to an incompatible stimulus, the superior/medial frontal cortex is activated to preliminarily attend on the focal area of the visual field based on previous trial compatibility, which alleviates the demands of the right inferior frontal cortex activity. Wang et al. (2015) also reported that high connectivity of the posterior parietal cortex to the superior frontal cortex is observed in participants with a high conflict adaptation effect, whereas a low conflict adaptation effect is associated with high connectivity of the anterior cingulate cortex to the insula and the occipital cortex. The result implies that proactive control is dominant and reactive control is inferior when participants adapt to conflict of stimulus. Therefore, we suggest that conflict adaptation is associated with the transition of two temporarily distinct cognitive control systems. However, in terms of alpha activity, an inconsistent result has been reported in a study of the Stroop task, i.e., a decrease in posterior alpha activities has been observed in trials preceded by incompatible stimuli (Compton, Arnstein, Freedman, Dainer-Best, & Liss, 2011). A previous study showed that a decrease in posterior alpha activities predicts good performance on visual perceptual discrimination (Babiloni et al., 2006; Ergenoglu et al., 2004; Hanslmayr et al., 2007). Thus, it is thought that, in the Stroop task, the presentation of an incompatible stimulus leads to a decrease in alpha activities in order to precisely perceive a stimulus because word and color are presented as an identical object, while it leads to an increase in alpha activities to reduce sensory input in the irrelevant visual field since the target and flankers are presented separately on the visual field in the flanker task. We were not able to find a conflict adaptation effect on behavioral performance in five of the participants (i.e., the inefficient group). In the ineffective group, N1 is not modulated by the previous trial compatibility. Thus, we suggest that their strategies were different from the strategies of other participants, e.g., a sustained attention focusing on the target regardless of compatible stimulus. However, there is no significant interaction between the previous trial compatibility and groups for alpha2 bands. Although this result indicates that a modulation of alpha2 activity was not related to conflict adaptation, we were not able to resolve these issues, since the ineffective group had only five members, and it was possible that strategies differed within the ineffective group. Atypical susceptibility to previous trial compatibility is reported in various neurological and psychiatric disorders, e.g., autism spectrum disorders (Larson, South, Clayson, & Clawson, 2012) and obsessive–compulsive disorders (Liu, Gehring, Weissman, Taylor, & Fitzgerald, 2012). We hope that further studies with such disorder groups will clarify the relationship between prestimulus oscillatory activity and conflict adaptation. In this study, the same type of compatibility was repeated at least thrice. Thus, it is possible that conflict adaptation is associated with a prediction based on a repeating number. Although the modulation of alpha activity was reduced in trials after repeating the same type of compatibility thrice (i.e., cI and iC) if the prediction contributed to conflict adaptation, differences between trials of the same previous compatibility (i.e., iI vs iC and cC vs
14
K. Suzuki, H. Shinoda / Brain and Cognition 100 (2015) 7–14
cI) were not found in this study. Hence, we concluded that the change in alpha activity was triggered by previous trial compatibility rather than by prediction. 5. Conclusion We examined neural activities involving conflict adaptation. Before stimulus presentation, frontal and posterior alpha activities were enhanced in trials preceded by incompatible stimuli, in comparison with those preceded by compatible stimuli. The source of alpha2 activities was estimated to be located in the superior/medial frontal cortex, which is suggested to be associated with proactive attentional control to reduce the sensory input on flankers. Following the stimulus presentation, N1 was reduced in iI more than in cI trials, where the difference of current densities was found in the right inferior frontal cortex. We consider that the small N1 amplitude reflects a decrease in demands of transient inhibition by the proactive attention control. Moreover, we showed the inverse relationship between the changes of the frontal alpha2 activity and N1 by previous trial compatibilities. These findings suggest that conflict adaptation is associated with the transition from reactive control to proactive control. References Aron, A. R., Robbins, T. W., & Poldrack, R. A. (2004). Inhibition and the right inferior frontal cortex. Trends in Cognitive Sciences, 8(4), 170–177. Babiloni, C., Vecchio, F., Bultrini, A., Luca Romani, G., & Rossini, P. M. (2006). Pre- and poststimulus alpha rhythms are related to conscious visual perception: A highresolution EEG study. Cerebral Cortex, 16(12), 1690–1700. Basar, E., & Stampfer, H. G. (1985). Important associations among EEG-dynamics, event-related potentials, short-term memory and learning. The International Journal of Neuroscience, 26(3–4), 161–180. Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001). Conflict monitoring and cognitive control. Psychological Review, 108(3), 624–652. Braver, T. S. (2012). The variable nature of cognitive control: A dual mechanisms framework. Trends in Cognitive Sciences, 16(2), 106–113. Braver, T. S., Paxton, J. L., Locke, H. S., & Barch, D. M. (2009). Flexible neural mechanisms of cognitive control within human prefrontal cortex. Proceedings of the National Academy of Sciences, 106(18), 7351–7356. Capotosto, P., Babiloni, C., Romani, G. L., & Corbetta, M. (2009). Frontoparietal cortex controls spatial attention through modulation of anticipatory alpha rhythms. The Journal of Neuroscience, 29(18), 5863–5872. Compton, R. J., Arnstein, D., Freedman, G., Dainer-Best, J., & Liss, A. (2011). Cognitive control in the intertrial interval: Evidence from EEG alpha power. Psychophysiology, 48(5), 583–590. Compton, R. J., Huber, E., Levinson, A. R., & Zheutlin, A. (2012). Is ‘‘conflict adaptation” driven by conflict? Behavioral and EEG evidence for the underappreciated role of congruent trials. Psychophysiology, 49(5), 583–589. Correa, Á., Rao, A., & Nobre, A. C. (2009). Anticipating conflict facilitates controlled stimulus-response selection. Journal of Cognitive Neuroscience, 21(8), 1461–1472. Delorme, A., & Makeig, S. (2004). EEGLAB: An open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. Journal of Neuroscience Methods, 134(1), 9–21. Ergenoglu, T., Demiralp, T., Bayraktaroglu, Z., Ergen, M., Beydagi, H., & Uresin, Y. (2004). Alpha rhythm of the EEG modulates visual detection performance in humans. Cognitive Brain Research, 20(3), 376–383. Eriksen, C. W., & Eriksen, B. A. (1979). Target redundancy in visual search: Do repetitions of the target within the display impair processing? Perception & Psychophysics, 26(3), 195–205. Fabiani, M., & Gratton, G. (2007). Event-related brain potentials: Methods, theory, and applications. In J. T. Cacioppo, L. G. Tassinary, & G. G. Berntson (Eds.), Handbook of psychophysiology (pp. 53–84). New York: Cambridge University Press. Freunberger, R., Holler, Y., Griesmayr, B., Gruber, W., Sauseng, P., & Klimesch, W. (2008). Functional similarities between the P1 component and alpha oscillations. European Journal of Neuroscience, 27(9), 2330–2340. Gratton, G., Coles, M. G., & Donchin, E. (1992). Optimizing the use of information: Strategic control of activation of responses. Journal of Experimental Psychology. General, 121(4), 480–506.
Hanslmayr, S., Aslan, A., Staudigl, T., Klimesch, W., Herrmann, C. S., & Bauml, K. H. (2007). Prestimulus oscillations predict visual perception performance between and within subjects. NeuroImage, 37(4), 1465–1473. Hazeltine, E., Bunge, S. A., Scanlon, M. D., & Gabrieli, J. D. (2003). Materialdependent and material-independent selection processes in the frontal and parietal lobes: An event-related fMRI investigation of response competition. Neuropsychologia, 41(9), 1208–1217. Hazeltine, E., Poldrack, R., & Gabrieli, J. D. (2000). Neural activation during response competition. Journal Cognitive Neuroscience, 12(Suppl 2), 118–129. Hillyard, S. A., Teder-Salejarvi, W. A., & Munte, T. F. (1998). Temporal dynamics of early perceptual processing. Current Opinion in Neurobiology, 8(2), 202–210. Johannes, S., Munte, T. F., Heinze, H. J., & Mangun, G. R. (1995). Luminance and spatial attention effects on early visual processing. Cognitive Brain Research, 2 (3), 189–205. Kerns, J. G., Cohen, J. D., MacDonald, A. W., 3rd, Cho, R. Y., Stenger, V. A., & Carter, C. S. (2004). Anterior cingulate conflict monitoring and adjustments in control. Science, 303(5660), 1023–1026. Klimesch, W., Sauseng, P., & Hanslmayr, S. (2007). EEG alpha oscillations: The inhibition-timing hypothesis. Brain Research Reviews, 53(1), 63–88. Larson, M. J., South, M., Clayson, P. E., & Clawson, A. (2012). Cognitive control and conflict adaptation in youth with high-functioning autism. Jounal of Child Psychology and Psychiatry, 53(4), 440–448. Lehmann, D., & Skrandies, W. (1980). Reference-free identification of components of checkerboard-evoked multichannel potential fields. Electroencephalography and Clinical Neurophysiology, 48(6), 609–621. Liu, Y., Bengson, J., Huang, H., Mangun, G. R., & Ding, M. (2014). Top-down modulation of neural activity in anticipatory visual attention: Control mechanisms revealed by simultaneous EEG-fMRI. Cerebral Cortex. Liu, Y., Gehring, W. J., Weissman, D. H., Taylor, S. F., & Fitzgerald, K. D. (2012). Trialby-trial adjustments of cognitive control following errors and response conflict are altered in pediatric obsessive compulsive disorder. Frontiers in Psychiatry, 3, 41. Min, B. K., & Park, H. J. (2010). Task-related modulation of anterior theta and posterior alpha EEG reflects top-down preparation. BMC Neuroscience, 11, 79. Pascual-Marqui, R. D. (2002). Standardized low-resolution brain electromagnetic tomography (sLORETA): Technical details. Methods and Findings in Experimental and Clinical Pharmacology, 24(Suppl D), 5–12. Rahn, E., & Basar, E. (1993). Enhancement of visual evoked potentials by stimulation during low prestimulus EEG stages. International Journal of Neuroscience, 72(1– 2), 123–136. Rana, K. D., & Vaina, L. M. (2014). Functional roles of 10 Hz alpha-band power modulating engagement and disengagement of cortical networks in a complex visual motion task. PLoS ONE, 9(10), e107715. Rihs, T. A., Michel, C. M., & Thut, G. (2007). Mechanisms of selective inhibition in visual spatial attention are indexed by alpha-band EEG synchronization. European Journal of Neuroscience, 25(2), 603–610. Rihs, T. A., Michel, C. M., & Thut, G. (2009). A bias for posterior alpha-band power suppression versus enhancement during shifting versus maintenance of spatial attention. NeuroImage, 44(1), 190–199. Sauseng, P., Feldheim, J. F., Freunberger, R., & Hummel, F. C. (2011). Right prefrontal TMS disrupts interregional anticipatory EEG alpha activity during shifting of visuospatial attention. Frontiers in Psychology, 2, 241. Sauseng, P., Klimesch, W., Stadler, W., Schabus, M., Doppelmayr, M., Hanslmayr, S., et al. (2005). A shift of visual spatial attention is selectively associated with human EEG alpha activity. European Journal of Neuroscience, 22(11), 2917–2926. Scerif, G., Worden, M. S., Davidson, M., Seiger, L., & Casey, B. J. (2006). Context modulates early stimulus processing when resolving stimulus-response conflict. Journal of Cognitive Neuroscience, 18(5), 781–792. Simon, J. R. (1969). Reactions toward the source of stimulation. Journal of Experimental Psychology, 81(1), 174–176. Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental Psychology, 18(6), 643–662. Thut, G., Nietzel, A., Brandt, S. A., & Pascual-Leone, A. (2006). Alpha-band electroencephalographic activity over occipital cortex indexes visuospatial attention bias and predicts visual target detection. The Journal of Neuroscience, 26(37), 9494–9502. Wang, X., Wang, T., Chen, Z., Hitchman, G., Liu, Y., & Chen, A. (2015). Functional connectivity patterns reflect individual differences in conflict adaptation. Neuropsychologia, 70, 177–184. Worden, M. S., Foxe, J. J., Wang, N., & Simpson, G. V. (2000). Anticipatory biasing of visuospatial attention indexed by retinotopically specific alpha-band electroencephalography increases over occipital cortex. The Journal of Neuroscience, 20(6), RC63. Yeung, N., Botvinick, M. M., & Cohen, J. D. (2004). The neural basis of error detection: Conflict monitoring and the error-related negativity. Psychological Review, 111 (4), 931–959.