- Email: [email protected]
NoGo task
Psychiatry Research: Neuroimaging 193 (2011) 177–181
Contents lists available at ScienceDirect
Psychiatry Research: Neuroimaging j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s n s
On the use of event-related potentials to auditory stimuli in the Go/NoGo task Janette Louise Smith ⁎, Kelly Maree Douglas School of Psychology, University of Newcastle, Brush Road, Ourimbah NSW 2258, Australia
a r t i c l e
i n f o
Article history: Received 20 December 2010 Received in revised form 20 January 2011 Accepted 9 March 2011 Keywords: Response inhibition Response conflict Modality-specific effect Perceptual overlap
a b s t r a c t The N2 and P3 components of the event-related potential (ERP) are putative markers of inhibition in the Go/ NoGo task. If this is the case, they should be unaffected by stimulus presentation modality. Theoretical researchers have suggested that the effect is smaller or absent with auditory stimuli, while others have shown that the effect depends on the perceptual similarity of the stimuli. Meanwhile, clinical researchers appear to be unaware of the debate. This study examined the N2 and P3 NoGo effects elicited by five sets of auditory stimuli varying in perceptual similarity. The N2 NoGo effect was significant for similar and different letters, and for similar tones, but not for different tones or novel sounds. As expected by the perceptual overlap hypothesis, the largest N2 NoGo effect was observed with the similar letters. In contrast, the P3 NoGo effect was significant and of a similar magnitude for all stimulus sets. The differential effect of the stimulus sets suggests separable underlying processes reflected in N2 and P3. Guidelines are provided for clinical researchers wishing to use auditory stimuli in the Go/NoGo task. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The Go/NoGo task is one behavioural paradigm that can be used to study inhibition, the suppression of an inappropriate response. In this task, participants are presented with two types of stimuli and asked to press a button to one type (Go trials) but not to the other type (NoGo trials). The task is simple for children and clinical groups to complete, and the need for inhibition following NoGo stimuli can be easily experimentally manipulated, for example, by requiring fast responses to Go stimuli (e.g., Jodo and Kayama, 1992), or by decreasing the probability of NoGo trials (e.g., Bruin and Wijers, 2002). One disadvantage of the Go/NoGo task is that the success of the inhibitory process produces little to measure behaviourally, when considered relative to other, more complex inhibitory tasks such as the stop-signal task (Logan and Cowan, 1984). Thus, event-related potential (ERP) methods, which represent the brain's average electrical response to a stimulus, can be used to provide information on this covert process. The ERP components N2 and P3 are robustly increased for NoGo compared to Go trials, termed the N2 NoGo and P3 NoGo effects, respectively. These potential markers of inhibitory processing have been studied in many versions of the Go/NoGo task in a range of disorders, including schizophrenia (e.g., Weisbrod et al., 2000; Roth et al., 2007), attention-deficit/hyperactivity disorder (e.g., Broyd et al., 2005; Kropotov et al., 2005), obsessive compulsive disorder (e.g., Ruchsow et al., 2007), depression (e.g., Kaiser et al., 2003), post-traumatic stress disorder (e.g., Shucard et al., 2008), foetal
⁎ Corresponding author. Tel.: + 61 2 4348 4149; fax: + 61 2 4348 4145. E-mail address: [email protected] (J.L. Smith). 0925-4927/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pscychresns.2011.03.002
alcohol syndrome (Burden et al., 2011) and substance dependence (e.g., Sokhadze et al., 2008). However, it appears that the clinical literature is not aware of a theoretical debate involving ERPs in the Go/NoGo task, namely, that the N2 NoGo effect may be difficult to obtain with auditory stimuli, and therefore may not represent a general inhibitory process. Falkenstein et al. (1999, 2002, 1995) observed a significant N2 NoGo effect only for the visual, but not the auditory, presentation of letter Go and NoGo stimuli (in 1995 and 1999, the letters ‘F’ and ‘J’, and in 2002, the letters ‘K’ and ‘T’, all with their German pronunciation). However, Nieuwenhuis et al. (2004) argued that the N2 NoGo effect depends on the perceptual similarity of stimuli: NoGo stimuli which are perceptually similar to Go stimuli will evoke more incorrect response activation, producing a greater need for inhibition and a greater N2 effect. Nieuwenhuis et al. further argued that Falkenstein et al.'s ‘F’ and ‘J’, and ‘K’ and ‘T’, were perhaps less perceptually similar when presented in the auditory than visual modality, and that this might explain Falkenstein et al.'s reduced effect for auditory stimuli. Consistent with this, they presented blocks of the letters ‘F’ and ‘S’ (which look different but sound similar in their English pronunciation) and the letters ‘F’ and ‘T’ (which look similar but sound different) in both the auditory and visual modalities. They observed a significant visual N2 NoGo effect regardless of stimulus similarity, but observed a significant effect for auditory stimuli only for the similar stimulus set. Thus, it appears that the selection of stimuli is a critical factor in auditory Go/NoGo task design, and some clinical papers aiming to study inhibitory differences in clinical groups may inadvertently reduce their statistical power if certain types of auditory stimuli are selected (e.g., Weisbrod et al., 2000; Kaiser et al., 2003; Broyd et al., 2005; Roth et al., 2007; Shucard et al., 2008).
178
J.L. Smith, K.M. Douglas / Psychiatry Research: Neuroimaging 193 (2011) 177–181
This study uses five sets of stimuli of the type popularly used in both theoretical and clinical research; that is, pure tone stimuli that are perceptually similar (1000/1100 Hz) and different (1000/ 2000 Hz), letters that are similar (F/S) and different (F/J), and novel complex sounds (bird whistle/frog croak). We examine whether each set elicits a significant NoGo N Go effect for N2 and P3, and also compare the results for the 1000/2000 Hz set, which has shown the most robust effect in previous studies (Schröger, 1993; Smith et al., 2004, 2006, 2007, 2008), to all other sets, to determine if a larger or smaller effect is gained with those sets. 2. Methods Participants were 21 students of the University of Newcastle (age range 18–52, mean age = 30.2; seven males, one left-handed) who participated for partial course credit or as volunteers. All subjects reported normal hearing and normal/corrected vision, and that they did not ingest any caffeine or use any nicotine products within the 2 h prior to the experiment. All participants gave informed consent and were made aware before the experiment began that they could withdraw at any time. The research protocol was approved by the University of Newcastle Human Research Ethics Committee before data collection began. Participants completed 10 experimental blocks (100 trials each) interspersed with 10 practice blocks (10 trials each) of an equiprobable Go/NoGo task. These were grouped into two blocks for each of the five sets of auditory stimuli. These were the spoken letters F and J, and F and S (English pronunciation), pure tones 1000 Hz and 2000 Hz, and 1000 and 1100 Hz, and novel complex sounds of a frog croak and bird whistle. Stimuli were 400 ms in duration (for tone stimuli, this included 50 ms rise and fall) and presented in stereo at 70 dB SPL over headphones with a stimulus onset asynchrony (SOA) that varied randomly between 1000 and 1400 ms (mean 1200 ms). A white fixation cross was continuously present on the participant screen throughout the experiment. The order of sets, as well as the Go/NoGo assignment for each stimulus in each set, was randomly determined by the stimulus computer at the start of the experiment, and the Go/ NoGo assignment switched between blocks. Participants were required to respond to Go stimuli by pressing a button with the right index finger within 600 ms. If participants did not make a button press on Go trials within 600 ms of stimulus onset the words “TOO SLOW” appeared on the screen 700 ms after the Go stimulus onset. Participants were familiarised with the testing procedure and laboratory before written informed consent was obtained, then filled out a brief questionnaire assessing for neurological disorders, drug use, etc. Once recording electrodes were fitted, subjects were seated in a quiet, dimly lit experimental room where testing took place. Instructions for the task appeared on a computer screen for the subject to read. Subjects were encouraged to keep as still as possible throughout the task and to keep eye movements to a minimum using a central fixation cross on the computer monitor. Continuous EEG was sampled at 2048 Hz from 64 scalp sites using an elasticised cap with sintered Ag/AgCl electrodes, with the BiosemiActiveTwo amplifier system. The reference and ground electrodes were embedded in the cap, with the active electrode: common mode sense (CMS), and passive electrode: driven right leg (DRL). Additional Ag/AgCl electrodes were placed on the left and right mastoids, 1 cm lateral to the outer canthus of each eye, and 1 cm above and below the left and right eyes. Behavioural data analysed included the percentage of commission errors (responding to a NoGo stimulus) and omission errors (failing to respond to Go stimuli). Mean reaction time to Go stimuli included trials where participants did not meet the 600 ms deadline. Eventrelated potentials were analysed using BESA version 5.3. All practice blocks and breaks between blocks were removed before recordings were corrected for eye blinks using BESA's built-in adaptive correction
technique (Ille et al., 2002) and all scalp electrodes were rereferenced to the mastoids. ERP epochs were time locked to 100 ms prior to stimulus onset until 700 ms post-stimulus. A high pass filter at .05 Hz (forward, down 6 dB/octave) and a low pass filter 30 Hz (zero phase, down 24 dB/octave) were applied. All trials were baseline corrected to the pre-stimulus period and trials with amplitudes exceeding ±100 μV were rejected. Average ERPs were created for Go and NoGo trials in each of the five stimulus sets (see Fig. 1, top). As is often the case with auditory stimuli, the N2 appeared in most conditions as a point of inflexion on the flank of the P3, appearing as a clear (and very delayed) peak only for the F/S set. Difference waveforms (NoGo–Go) were subsequently calculated for each participant for each stimulus set (see Fig. 1, bottom). N2 and P3 peaks were identified in the grand average difference waveforms at FCz (with reference to the raw ERPs, to waveforms at surrounding frontocentral sites, and to the typical latencies of N2 and P3) separately for each stimulus set, and mean amplitude was calculated in a 40 ms window centred on these peaks (see Table 1). For Go reaction time, a repeated measures ANOVA with the factor Stimulus Set (1000/2000 Hz, 1000/1100 Hz, F/J, F/S, Novels) was performed. Planned contrasts compared the mean for the control set (1000/2000 Hz) to all others separately. Because these contrasts were planned, and there were no more of them than the degrees of freedom for effect, no Bonferroni type adjustments to alpha were necessary (Tabachnick and Fidell, 1996). The square roots of Go and NoGo errors were calculated due to non-normal distributions of raw error scores, although the figures show raw errors for ease of understanding. Errors were analysed in a two-way repeated measures ANOVA with the factors Trial Type (Go/NoGo) and Stimulus Set, with the same contrasts as above. To determine whether a significant difference was observed for N2 and P3, 10 one-sample t-tests were performed comparing N2 and P3 for the five stimulus sets to a test value of zero (representing no significant difference between Go and NoGo waveforms). To determine whether different stimulus sets produced a significantly larger (or smaller) NoGo effect than the control set, repeated measures ANOVAs were performed with the Stimulus Set factor and contrasts outlined above, separately for N2 and P3. The degrees of freedom for all F contrasts were (1, 20), and for all t-tests were 20. All tests were compared to a P value of 0.05, two-tailed. 3. Results Fig. 2 shows the mean reaction time to Go stimuli (top left), errors to both types of stimuli (bottom left), and average N2 (top right) and P3 (bottom right) difference measurements. Reaction time to Go stimuli was significantly faster for the 1000/2000 Hz stimulus set compared to the other four stimulus sets (1000/1100 Hz: F = 13.7, P = 0.001; F/J: F = 48.3, P b 0.001; F/S: F = 372.1, P b 0.001; Novels: F = 22.1, P b 0.001). There were more errors for NoGo trials across all five stimulus sets compared to Go trials (F = 30.0, P b 0.001). The F/J and Novels stimulus sets did not significantly differ from 1000/2000 Hz (F/J: F b 1, P = 0.435; Novels: F = 2.8, P = 0.111), although there was a tendency for more errors in the 1000/1100 Hz set (F = 4.3, P = 0.053). However, there were significantly more errors in the F/S condition (F = 9.1, P = 0.007), and a significant Trial Type × Stimulus Set interaction (F = 9.0, P = 0.007), such that the F/S set produced significantly more errors on NoGo trials, while the rate of Go errors appears similar to the other stimulus sets. The N2 NoGo effect was in the expected direction and significantly different from zero (one-sample t-test) for the stimulus sets F/J (t = −2.3, P = 0.032), F/S (t = − 6.6, P b 0.001), and 1000/1100 Hz (t = −2.2, P = 0.039; see Fig. 3 for topographic maps of activity). For the 1000/2000 Hz set, there was an effect in the expected direction but this did not reach significance (t = −1.4, P = 0.191). For the Novel stimulus set, the effect was in the opposite direction (indicating
J.L. Smith, K.M. Douglas / Psychiatry Research: Neuroimaging 193 (2011) 177–181
179
Fig. 1. Grand mean waveforms at FCz to Go and NoGo trials for each stimulus set (top), and NoGo–Go difference waveforms for each set (bottom). Time zero represents the onset of the stimulus.
increased positivity for NoGo compared to Go waveforms) but did not reach significance (t = 1.2, P = 0.235). Relative to the control 1000/ 2000 Hz stimulus set, there was no significant difference in amplitude for the 1000/1100 Hz (F b 1, P = 0.726) and F/J stimulus sets (F b 1, P = 0.951). The F/S stimulus set produced a significantly larger N2 NoGo effect than the 1000/2000 Hz set (F = 21.9, P b 0.001). The Novels stimulus set produced a difference at N2 in the opposite direction compared to 1000/2000 Hz set (F = 4.9, P = 0.038). In contrast to the results for N2, a significant P3 NoGo effect was obtained for all five stimulus sets (1000/2000 Hz: t = 3.6, P = 0.002; 1000/1100 Hz: t = 4.0, P = 0.001; F/J: t = 2.9, P = 0.008; F/S: t = 2.8, P = 0.010; Novels: t = 3.9, P = 0.001). Relative to the P3 NoGo effect in the 1000/2000 Hz control set, there was no significant difference in the magnitude for the 1000/1100 Hz set (F b 1, P = 0.346), the F/S set (F = 2.1, P = 0.162), or the Novels set (F b 1, P = 0.944). However, there was a non-significant tendency for a smaller P3 NoGo effect for the F/J compared to the control set (F = 4.1, P = 0.058).
effect for similar tones and for similar and different letter stimuli, with the effect being the largest for similar letter stimuli (F/S). In contrast, the P3 NoGo effect was significant for all stimulus sets, and did not differ between them. Examination of difference waveforms was necessitated by the lack of a clear N2 peak in most raw waveforms (with the exception of the later N2/P3 complex for the F/S set). As is sometimes the case when using auditory stimuli in the Go/NoGo task (compare the waveforms of Smith et al., 2008), the NoGo N2 was visible for the other sets only as a point of inflexion on the flank of the P3 component. This is the case not only in the grand averages, but also for individual participants' raw waveforms, making peak detection difficult. One disadvantage of using difference waveforms is that they hide
4. Discussion The current study aimed to discover whether significant N2 and P3 NoGo effects could be obtained with any auditory stimuli, and if so, which produced the largest effect. We observed a significant N2 NoGo
Table 1 Latency windows for mean amplitude measurements at FCz for N2 and P3 for each stimulus set.
1000/2000 Hz 1000/1100 Hz F/J F/S Novels
N2 window (ms)
P3 window (ms)
155–195 170–210 170–210 360–400 180–220⁎
245–285 270–310 260–300 470–510 260–300
⁎ Note that the negative peak for Novels occurs too early (140 ms) to be considered an N2 effect; hence the latency window was set to a typical range for the N2 peak.
Fig. 2. Condition means and standard error for each stimulus set. Reaction time (top left), errors (bottom left), N2 difference amplitude at FCz (top right), and P3 difference amplitude at FCz (bottom right).
180
J.L. Smith, K.M. Douglas / Psychiatry Research: Neuroimaging 193 (2011) 177–181
Fig. 3. Isopotential maps of N2 and P3 difference amplitude across the scalp. Scale spacing between adjacent colours is 0.5 μV.
differences in peak latency but not amplitude in the raw waveforms. For example, examination of the 1000/1100 Hz raw waveforms reveals that the Go N2 peak occurs slightly later but is of similar amplitude to the NoGo N2 (point of inflexion). Thus, caution is required in interpreting results from that condition. However, examination of the other four conditions reveals that Go and NoGo peaks occur at approximately the same latency, and that results would not differ substantially if peak detection was performed on raw waveforms. Our failure to find a significant N2 NoGo effect for the dissimilar pure tone set (1000/2000 Hz) was unexpected, given that it is the stimulus set used most often, with a robust effect in most studies (Schröger, 1993; Smith et al., 2004, 2006, 2007, 2008). The reason for our small effect is not clear, but it is possible that temporal cueing is required to obtain a significant effect with these stimuli, as all of the above studies used a cue-target trial structure. However, it should be noted that the effect is at least in the correct direction. Similarly, the effect in the opposite direction for novel stimuli was also unexpected. A Go N NoGo effect is usually only observed under conditions of rare Go stimuli (Nieuwenhuis et al., 2003), or when Go stimuli are otherwise unexpected (e.g., following a NoGo cue: Randall and Smith, 2011; following a run of NoGo stimuli: Smith et al., 2010). However, it should be noted that this effect in itself was not significantly different from zero, and the apparent difference relative to 1000/2000 Hz set may simply be a consequence of two random (non-significant) deviations from zero in opposite directions. Despite this, a significant N2 NoGo effect was observed for the similar tones and the similar and different letter sets. Consequently, it appears that Falkenstein and associates (Falkenstein et al. 1995, 1999, 2002) are incorrect in its statement that the N2 NoGo effect can only be obtained with visual stimuli, and that it represents a modality specific process. Conversely, the results support Nieuwenhuis et al.'s (2004) suggestion that perceptually similar stimuli will evoke a N2 NoGo effect. In particular, the F/S condition produced the largest N2 NoGo effect, along with a significant increase in RT and NoGo error rate. The increase in RT, relative to the 1000/2000 Hz condition, is approximately equal to the period during which the F and S sounds are almost indistinguishable (i.e., the first 180 ms). Note that the endogenous (i.e., decision-related) N2/P3 components are also delayed by about the same period in the F/S condition. This difficulty in distinguishing the two stimuli elicited greater activation of the response on NoGo trials, resulting in increased likelihood of responding to NoGo targets, and an increased N2 NoGo effect. In contrast to the N2 results, the P3 NoGo effect was significant and of equal magnitude for all stimulus sets. It is beyond the scope of this paper to include a thorough discussion of the functional significance of the N2 and P3 components, including debates on the conflict vs. inhibition interpretations of N2 (e.g., Folstein and Van Petten, 2008). The simple experimental method of this study was aimed at determining whether a significant effect could be obtained with auditory
stimuli, as rather more complex designs are needed to provide discriminating data on the conflict/inhibition question (and indeed, the current results are predicted under both theories). However, the fact that the N2 effect was strongly dependent on the stimulus set used, while the P3 effect was significant regardless of stimulus set, adds weight to the notion that these components represent separate psychological processes. In summary, this study suggests that theoretical Go/NoGo research should not be limited to the use of visual stimuli. In addition, clinical researchers wishing to use auditory Go/NoGo tasks should select stimulus sets carefully, depending on the component of interest. If the researcher is interested in P3 differences, then the choice of stimuli does not appear to matter, but if interested in N2 differences, and particularly if one wishes to tax the inhibitory process in clinical groups, then similar letters will produce large ERP and behavioural effects. Acknowledgements The authors wish to thank Mr Tony Kemp for writing the stimulus presentation programme. References Broyd, S.J., Johnstone, S.J., Barry, R.J., Clarke, A.R., McCarthy, R., Selikowitz, M., Lawrence, C.A., 2005. The effect of methylphenidate on response inhibition and the eventrelated potential of children with Attention Deficit/Hyperactivity Disorder. International Journal of Psychophysiology 58, 47–58. Bruin, K.J., Wijers, A.A., 2002. Inhibition, response mode, and stimulus probability: a comparative event-related potential study. Clinical Neurophysiology 113, 1172–1182. Burden, M.J., Westerlund, A., Muckle, G., Dodge, N., Dewailly, E., Nelson, C.A., Jacobson, S.W., Jacobson, J.L., 2011. The effects of maternal binge drinking during pregnancy on neural correlates of response inhibition and memory in childhood. Alcoholism, Clinical and Experimental Research 35, 1–14. Falkenstein, M., Hoormann, J., Hohnsbein, J., 1999. ERP components in Go/Nogo tasks and their relation to inhibition. Acta Psychologica 101, 267–291. Falkenstein, M., Hoormann, J., Hohnsbein, J., 2002. Inhibition-related ERP components: variation with modality, age, and time-on-task. Journal of Psychophysiology 16, 167–175. Falkenstein, M., Koshlykova, N.A., Kiroj, V.N., Hoormann, J., Hohnsbein, J., 1995. Late ERP components in visual and auditory Go/Nogo tasks. Electroencephalography and Clinical Neurophysiology: Evoked Potentials 96, 36–43. Folstein, J.R., Van Petten, C., 2008. Influence of cognitive control and mismatch on the N2 component of the ERP: a review. Psychophysiology 45, 152–170. Ille, N., Berg, P., Scherg, M., 2002. Artifact correction of the ongoing EEG using spatial filters based on artifact and brain signal topographies. Journal of Clinical Neurophysiology 19, 113–124. Jodo, E., Kayama, Y., 1992. Relation of a negative ERP component to response inhibition in a Go/No-go task. Electroencephalography and Clinical Neurophysiology 82, 477–482. Kaiser, S., Unger, J., Kiefer, M., Markela, J., Mundt, C., Weisbrod, M., 2003. Executive control deficit in depression: event-related potentials in a Go/Nogo task. Neuroimaging 122, 169–184. Kropotov, J.D., Grin-Yatsenko, V.A., Ponomarev, V.A., Chutko, L.S., Yakovenko, E.A., Nikishena, I.S., 2005. ERPs correlates of EEG relative beta training in ADHD children. International Journal of Psychophysiology 55, 23–34. Logan, G.D., Cowan, W.B., 1984. On the ability to inhibit thought and action: a theory of an act of control. Psychological Review 91, 295–327.
J.L. Smith, K.M. Douglas / Psychiatry Research: Neuroimaging 193 (2011) 177–181 Nieuwenhuis, S., Yeung, N., Cohen, J.D., 2004. Stimulus modality, perceptual overlap, and the go/no-go N2. Psychophysiology 41, 157–160. Nieuwenhuis, S., Yeung, N., van den Wildenberg, W., Ridderinkhof, K.R., 2003. Electrophysiological correlates of anterior cingulate function in a go/no-go task: effects of response conflict and trial type frequency. Cognitive, Affective & Behavioral Neuroscience 3, 17–26. Randall, W.M., Smith, J.L., 2011. Conflict and inhibition in the cued-Go/NoGo task. Clinical Neurophysiology. doi:10.1016/j.clinph.2011.05.012. Roth, A., Roesch-Ely, D., Bender, S., Weisbrod, M., Kaiser, S., 2007. Increased eventrelated potential latency and amplitude variability in schizophrenia detected through wavelet-based single trial analysis. International Journal of Psychophysiology 66, 244–254. Ruchsow, M., Reuter, K., Hermle, L., Ebert, D., Kiefer, M., Falkenstein, M., 2007. Executive control in obsessive-compulsive disorder: event-related potentials in a Go/Nogo task. Journal of Neural Transmission 114, 1595–1601. Schröger, E., 1993. Event-related potentials to auditory stimuli following transient shifts of spatial attention in a Go/Nogo task. Biological Psychology 36, 183–207. Shucard, J.L., McCabe, D.C., Szymanski, H., 2008. An event-related potential study of attention deficits in posttraumatic stress disorder during auditory and visual Go/ NoGo continuous performance tasks. Biological Psychology 79, 223–233. Smith, J.L., Johnstone, S.J., Barry, R.J., 2004. Inhibitory processing during the Go/NoGo task: an ERP analysis of children with attention-deficit/hyperactivity disorder. Clinical Neurophysiology 115, 1320–1331.
181
Smith, J.L., Johnstone, S.J., Barry, R.J., 2006. Effects of pre-stimulus processing on subsequent events in a warned Go/NoGo paradigm: response preparation, execution and inhibition. International Journal of Psychophysiology 61, 121–133. Smith, J.L., Johnstone, S.J., Barry, R.J., 2007. Response priming in the Go/NoGo task: the N2 reflects neither inhibition nor conflict. Clinical Neurophysiology 118, 343–355. Smith, J.L., Johnstone, S.J., Barry, R.J., 2008. Movement-related potentials in the Go/NoGo task: the P3 reflects both cognitive and motor inhibition. Clinical Neurophysiology 119, 704–714. Smith, J.L., Smith, E.A., Provost, A.L., Heathcote, A., 2010. Sequence effects support the conflict theory of N2 and P3 in the Go/NoGo task. International Journal of Psychophysiology 75, 217–226. Sokhadze, E., Stewart, C., Hollifield, M., Tasman, A., 2008. Event-related potential study of executive dysfunctions in a speeded reaction task in cocaine addiction. Journal of Neurotherapy 12, 185–204. Tabachnick, B.G., Fidell, L.S., 1996. Using Multivariate Statistics, 3rd ed. HarperCollins, New York. Weisbrod, M., Kiefer, M., Marzinzik, F., Spitzer, M., 2000. Executive control is disturbed in schizophrenia: evidence from event-related potentials in a Go/NoGo task. Biological Psychiatry 47, 51–60.
Contents lists available at ScienceDirect
Psychiatry Research: Neuroimaging j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / p s yc h r e s n s
On the use of event-related potentials to auditory stimuli in the Go/NoGo task Janette Louise Smith ⁎, Kelly Maree Douglas School of Psychology, University of Newcastle, Brush Road, Ourimbah NSW 2258, Australia
a r t i c l e
i n f o
Article history: Received 20 December 2010 Received in revised form 20 January 2011 Accepted 9 March 2011 Keywords: Response inhibition Response conflict Modality-specific effect Perceptual overlap
a b s t r a c t The N2 and P3 components of the event-related potential (ERP) are putative markers of inhibition in the Go/ NoGo task. If this is the case, they should be unaffected by stimulus presentation modality. Theoretical researchers have suggested that the effect is smaller or absent with auditory stimuli, while others have shown that the effect depends on the perceptual similarity of the stimuli. Meanwhile, clinical researchers appear to be unaware of the debate. This study examined the N2 and P3 NoGo effects elicited by five sets of auditory stimuli varying in perceptual similarity. The N2 NoGo effect was significant for similar and different letters, and for similar tones, but not for different tones or novel sounds. As expected by the perceptual overlap hypothesis, the largest N2 NoGo effect was observed with the similar letters. In contrast, the P3 NoGo effect was significant and of a similar magnitude for all stimulus sets. The differential effect of the stimulus sets suggests separable underlying processes reflected in N2 and P3. Guidelines are provided for clinical researchers wishing to use auditory stimuli in the Go/NoGo task. © 2011 Elsevier Ireland Ltd. All rights reserved.
1. Introduction The Go/NoGo task is one behavioural paradigm that can be used to study inhibition, the suppression of an inappropriate response. In this task, participants are presented with two types of stimuli and asked to press a button to one type (Go trials) but not to the other type (NoGo trials). The task is simple for children and clinical groups to complete, and the need for inhibition following NoGo stimuli can be easily experimentally manipulated, for example, by requiring fast responses to Go stimuli (e.g., Jodo and Kayama, 1992), or by decreasing the probability of NoGo trials (e.g., Bruin and Wijers, 2002). One disadvantage of the Go/NoGo task is that the success of the inhibitory process produces little to measure behaviourally, when considered relative to other, more complex inhibitory tasks such as the stop-signal task (Logan and Cowan, 1984). Thus, event-related potential (ERP) methods, which represent the brain's average electrical response to a stimulus, can be used to provide information on this covert process. The ERP components N2 and P3 are robustly increased for NoGo compared to Go trials, termed the N2 NoGo and P3 NoGo effects, respectively. These potential markers of inhibitory processing have been studied in many versions of the Go/NoGo task in a range of disorders, including schizophrenia (e.g., Weisbrod et al., 2000; Roth et al., 2007), attention-deficit/hyperactivity disorder (e.g., Broyd et al., 2005; Kropotov et al., 2005), obsessive compulsive disorder (e.g., Ruchsow et al., 2007), depression (e.g., Kaiser et al., 2003), post-traumatic stress disorder (e.g., Shucard et al., 2008), foetal
⁎ Corresponding author. Tel.: + 61 2 4348 4149; fax: + 61 2 4348 4145. E-mail address: [email protected] (J.L. Smith). 0925-4927/$ – see front matter © 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.pscychresns.2011.03.002
alcohol syndrome (Burden et al., 2011) and substance dependence (e.g., Sokhadze et al., 2008). However, it appears that the clinical literature is not aware of a theoretical debate involving ERPs in the Go/NoGo task, namely, that the N2 NoGo effect may be difficult to obtain with auditory stimuli, and therefore may not represent a general inhibitory process. Falkenstein et al. (1999, 2002, 1995) observed a significant N2 NoGo effect only for the visual, but not the auditory, presentation of letter Go and NoGo stimuli (in 1995 and 1999, the letters ‘F’ and ‘J’, and in 2002, the letters ‘K’ and ‘T’, all with their German pronunciation). However, Nieuwenhuis et al. (2004) argued that the N2 NoGo effect depends on the perceptual similarity of stimuli: NoGo stimuli which are perceptually similar to Go stimuli will evoke more incorrect response activation, producing a greater need for inhibition and a greater N2 effect. Nieuwenhuis et al. further argued that Falkenstein et al.'s ‘F’ and ‘J’, and ‘K’ and ‘T’, were perhaps less perceptually similar when presented in the auditory than visual modality, and that this might explain Falkenstein et al.'s reduced effect for auditory stimuli. Consistent with this, they presented blocks of the letters ‘F’ and ‘S’ (which look different but sound similar in their English pronunciation) and the letters ‘F’ and ‘T’ (which look similar but sound different) in both the auditory and visual modalities. They observed a significant visual N2 NoGo effect regardless of stimulus similarity, but observed a significant effect for auditory stimuli only for the similar stimulus set. Thus, it appears that the selection of stimuli is a critical factor in auditory Go/NoGo task design, and some clinical papers aiming to study inhibitory differences in clinical groups may inadvertently reduce their statistical power if certain types of auditory stimuli are selected (e.g., Weisbrod et al., 2000; Kaiser et al., 2003; Broyd et al., 2005; Roth et al., 2007; Shucard et al., 2008).
178
J.L. Smith, K.M. Douglas / Psychiatry Research: Neuroimaging 193 (2011) 177–181
This study uses five sets of stimuli of the type popularly used in both theoretical and clinical research; that is, pure tone stimuli that are perceptually similar (1000/1100 Hz) and different (1000/ 2000 Hz), letters that are similar (F/S) and different (F/J), and novel complex sounds (bird whistle/frog croak). We examine whether each set elicits a significant NoGo N Go effect for N2 and P3, and also compare the results for the 1000/2000 Hz set, which has shown the most robust effect in previous studies (Schröger, 1993; Smith et al., 2004, 2006, 2007, 2008), to all other sets, to determine if a larger or smaller effect is gained with those sets. 2. Methods Participants were 21 students of the University of Newcastle (age range 18–52, mean age = 30.2; seven males, one left-handed) who participated for partial course credit or as volunteers. All subjects reported normal hearing and normal/corrected vision, and that they did not ingest any caffeine or use any nicotine products within the 2 h prior to the experiment. All participants gave informed consent and were made aware before the experiment began that they could withdraw at any time. The research protocol was approved by the University of Newcastle Human Research Ethics Committee before data collection began. Participants completed 10 experimental blocks (100 trials each) interspersed with 10 practice blocks (10 trials each) of an equiprobable Go/NoGo task. These were grouped into two blocks for each of the five sets of auditory stimuli. These were the spoken letters F and J, and F and S (English pronunciation), pure tones 1000 Hz and 2000 Hz, and 1000 and 1100 Hz, and novel complex sounds of a frog croak and bird whistle. Stimuli were 400 ms in duration (for tone stimuli, this included 50 ms rise and fall) and presented in stereo at 70 dB SPL over headphones with a stimulus onset asynchrony (SOA) that varied randomly between 1000 and 1400 ms (mean 1200 ms). A white fixation cross was continuously present on the participant screen throughout the experiment. The order of sets, as well as the Go/NoGo assignment for each stimulus in each set, was randomly determined by the stimulus computer at the start of the experiment, and the Go/ NoGo assignment switched between blocks. Participants were required to respond to Go stimuli by pressing a button with the right index finger within 600 ms. If participants did not make a button press on Go trials within 600 ms of stimulus onset the words “TOO SLOW” appeared on the screen 700 ms after the Go stimulus onset. Participants were familiarised with the testing procedure and laboratory before written informed consent was obtained, then filled out a brief questionnaire assessing for neurological disorders, drug use, etc. Once recording electrodes were fitted, subjects were seated in a quiet, dimly lit experimental room where testing took place. Instructions for the task appeared on a computer screen for the subject to read. Subjects were encouraged to keep as still as possible throughout the task and to keep eye movements to a minimum using a central fixation cross on the computer monitor. Continuous EEG was sampled at 2048 Hz from 64 scalp sites using an elasticised cap with sintered Ag/AgCl electrodes, with the BiosemiActiveTwo amplifier system. The reference and ground electrodes were embedded in the cap, with the active electrode: common mode sense (CMS), and passive electrode: driven right leg (DRL). Additional Ag/AgCl electrodes were placed on the left and right mastoids, 1 cm lateral to the outer canthus of each eye, and 1 cm above and below the left and right eyes. Behavioural data analysed included the percentage of commission errors (responding to a NoGo stimulus) and omission errors (failing to respond to Go stimuli). Mean reaction time to Go stimuli included trials where participants did not meet the 600 ms deadline. Eventrelated potentials were analysed using BESA version 5.3. All practice blocks and breaks between blocks were removed before recordings were corrected for eye blinks using BESA's built-in adaptive correction
technique (Ille et al., 2002) and all scalp electrodes were rereferenced to the mastoids. ERP epochs were time locked to 100 ms prior to stimulus onset until 700 ms post-stimulus. A high pass filter at .05 Hz (forward, down 6 dB/octave) and a low pass filter 30 Hz (zero phase, down 24 dB/octave) were applied. All trials were baseline corrected to the pre-stimulus period and trials with amplitudes exceeding ±100 μV were rejected. Average ERPs were created for Go and NoGo trials in each of the five stimulus sets (see Fig. 1, top). As is often the case with auditory stimuli, the N2 appeared in most conditions as a point of inflexion on the flank of the P3, appearing as a clear (and very delayed) peak only for the F/S set. Difference waveforms (NoGo–Go) were subsequently calculated for each participant for each stimulus set (see Fig. 1, bottom). N2 and P3 peaks were identified in the grand average difference waveforms at FCz (with reference to the raw ERPs, to waveforms at surrounding frontocentral sites, and to the typical latencies of N2 and P3) separately for each stimulus set, and mean amplitude was calculated in a 40 ms window centred on these peaks (see Table 1). For Go reaction time, a repeated measures ANOVA with the factor Stimulus Set (1000/2000 Hz, 1000/1100 Hz, F/J, F/S, Novels) was performed. Planned contrasts compared the mean for the control set (1000/2000 Hz) to all others separately. Because these contrasts were planned, and there were no more of them than the degrees of freedom for effect, no Bonferroni type adjustments to alpha were necessary (Tabachnick and Fidell, 1996). The square roots of Go and NoGo errors were calculated due to non-normal distributions of raw error scores, although the figures show raw errors for ease of understanding. Errors were analysed in a two-way repeated measures ANOVA with the factors Trial Type (Go/NoGo) and Stimulus Set, with the same contrasts as above. To determine whether a significant difference was observed for N2 and P3, 10 one-sample t-tests were performed comparing N2 and P3 for the five stimulus sets to a test value of zero (representing no significant difference between Go and NoGo waveforms). To determine whether different stimulus sets produced a significantly larger (or smaller) NoGo effect than the control set, repeated measures ANOVAs were performed with the Stimulus Set factor and contrasts outlined above, separately for N2 and P3. The degrees of freedom for all F contrasts were (1, 20), and for all t-tests were 20. All tests were compared to a P value of 0.05, two-tailed. 3. Results Fig. 2 shows the mean reaction time to Go stimuli (top left), errors to both types of stimuli (bottom left), and average N2 (top right) and P3 (bottom right) difference measurements. Reaction time to Go stimuli was significantly faster for the 1000/2000 Hz stimulus set compared to the other four stimulus sets (1000/1100 Hz: F = 13.7, P = 0.001; F/J: F = 48.3, P b 0.001; F/S: F = 372.1, P b 0.001; Novels: F = 22.1, P b 0.001). There were more errors for NoGo trials across all five stimulus sets compared to Go trials (F = 30.0, P b 0.001). The F/J and Novels stimulus sets did not significantly differ from 1000/2000 Hz (F/J: F b 1, P = 0.435; Novels: F = 2.8, P = 0.111), although there was a tendency for more errors in the 1000/1100 Hz set (F = 4.3, P = 0.053). However, there were significantly more errors in the F/S condition (F = 9.1, P = 0.007), and a significant Trial Type × Stimulus Set interaction (F = 9.0, P = 0.007), such that the F/S set produced significantly more errors on NoGo trials, while the rate of Go errors appears similar to the other stimulus sets. The N2 NoGo effect was in the expected direction and significantly different from zero (one-sample t-test) for the stimulus sets F/J (t = −2.3, P = 0.032), F/S (t = − 6.6, P b 0.001), and 1000/1100 Hz (t = −2.2, P = 0.039; see Fig. 3 for topographic maps of activity). For the 1000/2000 Hz set, there was an effect in the expected direction but this did not reach significance (t = −1.4, P = 0.191). For the Novel stimulus set, the effect was in the opposite direction (indicating
J.L. Smith, K.M. Douglas / Psychiatry Research: Neuroimaging 193 (2011) 177–181
179
Fig. 1. Grand mean waveforms at FCz to Go and NoGo trials for each stimulus set (top), and NoGo–Go difference waveforms for each set (bottom). Time zero represents the onset of the stimulus.
increased positivity for NoGo compared to Go waveforms) but did not reach significance (t = 1.2, P = 0.235). Relative to the control 1000/ 2000 Hz stimulus set, there was no significant difference in amplitude for the 1000/1100 Hz (F b 1, P = 0.726) and F/J stimulus sets (F b 1, P = 0.951). The F/S stimulus set produced a significantly larger N2 NoGo effect than the 1000/2000 Hz set (F = 21.9, P b 0.001). The Novels stimulus set produced a difference at N2 in the opposite direction compared to 1000/2000 Hz set (F = 4.9, P = 0.038). In contrast to the results for N2, a significant P3 NoGo effect was obtained for all five stimulus sets (1000/2000 Hz: t = 3.6, P = 0.002; 1000/1100 Hz: t = 4.0, P = 0.001; F/J: t = 2.9, P = 0.008; F/S: t = 2.8, P = 0.010; Novels: t = 3.9, P = 0.001). Relative to the P3 NoGo effect in the 1000/2000 Hz control set, there was no significant difference in the magnitude for the 1000/1100 Hz set (F b 1, P = 0.346), the F/S set (F = 2.1, P = 0.162), or the Novels set (F b 1, P = 0.944). However, there was a non-significant tendency for a smaller P3 NoGo effect for the F/J compared to the control set (F = 4.1, P = 0.058).
effect for similar tones and for similar and different letter stimuli, with the effect being the largest for similar letter stimuli (F/S). In contrast, the P3 NoGo effect was significant for all stimulus sets, and did not differ between them. Examination of difference waveforms was necessitated by the lack of a clear N2 peak in most raw waveforms (with the exception of the later N2/P3 complex for the F/S set). As is sometimes the case when using auditory stimuli in the Go/NoGo task (compare the waveforms of Smith et al., 2008), the NoGo N2 was visible for the other sets only as a point of inflexion on the flank of the P3 component. This is the case not only in the grand averages, but also for individual participants' raw waveforms, making peak detection difficult. One disadvantage of using difference waveforms is that they hide
4. Discussion The current study aimed to discover whether significant N2 and P3 NoGo effects could be obtained with any auditory stimuli, and if so, which produced the largest effect. We observed a significant N2 NoGo
Table 1 Latency windows for mean amplitude measurements at FCz for N2 and P3 for each stimulus set.
1000/2000 Hz 1000/1100 Hz F/J F/S Novels
N2 window (ms)
P3 window (ms)
155–195 170–210 170–210 360–400 180–220⁎
245–285 270–310 260–300 470–510 260–300
⁎ Note that the negative peak for Novels occurs too early (140 ms) to be considered an N2 effect; hence the latency window was set to a typical range for the N2 peak.
Fig. 2. Condition means and standard error for each stimulus set. Reaction time (top left), errors (bottom left), N2 difference amplitude at FCz (top right), and P3 difference amplitude at FCz (bottom right).
180
J.L. Smith, K.M. Douglas / Psychiatry Research: Neuroimaging 193 (2011) 177–181
Fig. 3. Isopotential maps of N2 and P3 difference amplitude across the scalp. Scale spacing between adjacent colours is 0.5 μV.
differences in peak latency but not amplitude in the raw waveforms. For example, examination of the 1000/1100 Hz raw waveforms reveals that the Go N2 peak occurs slightly later but is of similar amplitude to the NoGo N2 (point of inflexion). Thus, caution is required in interpreting results from that condition. However, examination of the other four conditions reveals that Go and NoGo peaks occur at approximately the same latency, and that results would not differ substantially if peak detection was performed on raw waveforms. Our failure to find a significant N2 NoGo effect for the dissimilar pure tone set (1000/2000 Hz) was unexpected, given that it is the stimulus set used most often, with a robust effect in most studies (Schröger, 1993; Smith et al., 2004, 2006, 2007, 2008). The reason for our small effect is not clear, but it is possible that temporal cueing is required to obtain a significant effect with these stimuli, as all of the above studies used a cue-target trial structure. However, it should be noted that the effect is at least in the correct direction. Similarly, the effect in the opposite direction for novel stimuli was also unexpected. A Go N NoGo effect is usually only observed under conditions of rare Go stimuli (Nieuwenhuis et al., 2003), or when Go stimuli are otherwise unexpected (e.g., following a NoGo cue: Randall and Smith, 2011; following a run of NoGo stimuli: Smith et al., 2010). However, it should be noted that this effect in itself was not significantly different from zero, and the apparent difference relative to 1000/2000 Hz set may simply be a consequence of two random (non-significant) deviations from zero in opposite directions. Despite this, a significant N2 NoGo effect was observed for the similar tones and the similar and different letter sets. Consequently, it appears that Falkenstein and associates (Falkenstein et al. 1995, 1999, 2002) are incorrect in its statement that the N2 NoGo effect can only be obtained with visual stimuli, and that it represents a modality specific process. Conversely, the results support Nieuwenhuis et al.'s (2004) suggestion that perceptually similar stimuli will evoke a N2 NoGo effect. In particular, the F/S condition produced the largest N2 NoGo effect, along with a significant increase in RT and NoGo error rate. The increase in RT, relative to the 1000/2000 Hz condition, is approximately equal to the period during which the F and S sounds are almost indistinguishable (i.e., the first 180 ms). Note that the endogenous (i.e., decision-related) N2/P3 components are also delayed by about the same period in the F/S condition. This difficulty in distinguishing the two stimuli elicited greater activation of the response on NoGo trials, resulting in increased likelihood of responding to NoGo targets, and an increased N2 NoGo effect. In contrast to the N2 results, the P3 NoGo effect was significant and of equal magnitude for all stimulus sets. It is beyond the scope of this paper to include a thorough discussion of the functional significance of the N2 and P3 components, including debates on the conflict vs. inhibition interpretations of N2 (e.g., Folstein and Van Petten, 2008). The simple experimental method of this study was aimed at determining whether a significant effect could be obtained with auditory
stimuli, as rather more complex designs are needed to provide discriminating data on the conflict/inhibition question (and indeed, the current results are predicted under both theories). However, the fact that the N2 effect was strongly dependent on the stimulus set used, while the P3 effect was significant regardless of stimulus set, adds weight to the notion that these components represent separate psychological processes. In summary, this study suggests that theoretical Go/NoGo research should not be limited to the use of visual stimuli. In addition, clinical researchers wishing to use auditory Go/NoGo tasks should select stimulus sets carefully, depending on the component of interest. If the researcher is interested in P3 differences, then the choice of stimuli does not appear to matter, but if interested in N2 differences, and particularly if one wishes to tax the inhibitory process in clinical groups, then similar letters will produce large ERP and behavioural effects. Acknowledgements The authors wish to thank Mr Tony Kemp for writing the stimulus presentation programme. References Broyd, S.J., Johnstone, S.J., Barry, R.J., Clarke, A.R., McCarthy, R., Selikowitz, M., Lawrence, C.A., 2005. The effect of methylphenidate on response inhibition and the eventrelated potential of children with Attention Deficit/Hyperactivity Disorder. International Journal of Psychophysiology 58, 47–58. Bruin, K.J., Wijers, A.A., 2002. Inhibition, response mode, and stimulus probability: a comparative event-related potential study. Clinical Neurophysiology 113, 1172–1182. Burden, M.J., Westerlund, A., Muckle, G., Dodge, N., Dewailly, E., Nelson, C.A., Jacobson, S.W., Jacobson, J.L., 2011. The effects of maternal binge drinking during pregnancy on neural correlates of response inhibition and memory in childhood. Alcoholism, Clinical and Experimental Research 35, 1–14. Falkenstein, M., Hoormann, J., Hohnsbein, J., 1999. ERP components in Go/Nogo tasks and their relation to inhibition. Acta Psychologica 101, 267–291. Falkenstein, M., Hoormann, J., Hohnsbein, J., 2002. Inhibition-related ERP components: variation with modality, age, and time-on-task. Journal of Psychophysiology 16, 167–175. Falkenstein, M., Koshlykova, N.A., Kiroj, V.N., Hoormann, J., Hohnsbein, J., 1995. Late ERP components in visual and auditory Go/Nogo tasks. Electroencephalography and Clinical Neurophysiology: Evoked Potentials 96, 36–43. Folstein, J.R., Van Petten, C., 2008. Influence of cognitive control and mismatch on the N2 component of the ERP: a review. Psychophysiology 45, 152–170. Ille, N., Berg, P., Scherg, M., 2002. Artifact correction of the ongoing EEG using spatial filters based on artifact and brain signal topographies. Journal of Clinical Neurophysiology 19, 113–124. Jodo, E., Kayama, Y., 1992. Relation of a negative ERP component to response inhibition in a Go/No-go task. Electroencephalography and Clinical Neurophysiology 82, 477–482. Kaiser, S., Unger, J., Kiefer, M., Markela, J., Mundt, C., Weisbrod, M., 2003. Executive control deficit in depression: event-related potentials in a Go/Nogo task. Neuroimaging 122, 169–184. Kropotov, J.D., Grin-Yatsenko, V.A., Ponomarev, V.A., Chutko, L.S., Yakovenko, E.A., Nikishena, I.S., 2005. ERPs correlates of EEG relative beta training in ADHD children. International Journal of Psychophysiology 55, 23–34. Logan, G.D., Cowan, W.B., 1984. On the ability to inhibit thought and action: a theory of an act of control. Psychological Review 91, 295–327.
J.L. Smith, K.M. Douglas / Psychiatry Research: Neuroimaging 193 (2011) 177–181 Nieuwenhuis, S., Yeung, N., Cohen, J.D., 2004. Stimulus modality, perceptual overlap, and the go/no-go N2. Psychophysiology 41, 157–160. Nieuwenhuis, S., Yeung, N., van den Wildenberg, W., Ridderinkhof, K.R., 2003. Electrophysiological correlates of anterior cingulate function in a go/no-go task: effects of response conflict and trial type frequency. Cognitive, Affective & Behavioral Neuroscience 3, 17–26. Randall, W.M., Smith, J.L., 2011. Conflict and inhibition in the cued-Go/NoGo task. Clinical Neurophysiology. doi:10.1016/j.clinph.2011.05.012. Roth, A., Roesch-Ely, D., Bender, S., Weisbrod, M., Kaiser, S., 2007. Increased eventrelated potential latency and amplitude variability in schizophrenia detected through wavelet-based single trial analysis. International Journal of Psychophysiology 66, 244–254. Ruchsow, M., Reuter, K., Hermle, L., Ebert, D., Kiefer, M., Falkenstein, M., 2007. Executive control in obsessive-compulsive disorder: event-related potentials in a Go/Nogo task. Journal of Neural Transmission 114, 1595–1601. Schröger, E., 1993. Event-related potentials to auditory stimuli following transient shifts of spatial attention in a Go/Nogo task. Biological Psychology 36, 183–207. Shucard, J.L., McCabe, D.C., Szymanski, H., 2008. An event-related potential study of attention deficits in posttraumatic stress disorder during auditory and visual Go/ NoGo continuous performance tasks. Biological Psychology 79, 223–233. Smith, J.L., Johnstone, S.J., Barry, R.J., 2004. Inhibitory processing during the Go/NoGo task: an ERP analysis of children with attention-deficit/hyperactivity disorder. Clinical Neurophysiology 115, 1320–1331.
181
Smith, J.L., Johnstone, S.J., Barry, R.J., 2006. Effects of pre-stimulus processing on subsequent events in a warned Go/NoGo paradigm: response preparation, execution and inhibition. International Journal of Psychophysiology 61, 121–133. Smith, J.L., Johnstone, S.J., Barry, R.J., 2007. Response priming in the Go/NoGo task: the N2 reflects neither inhibition nor conflict. Clinical Neurophysiology 118, 343–355. Smith, J.L., Johnstone, S.J., Barry, R.J., 2008. Movement-related potentials in the Go/NoGo task: the P3 reflects both cognitive and motor inhibition. Clinical Neurophysiology 119, 704–714. Smith, J.L., Smith, E.A., Provost, A.L., Heathcote, A., 2010. Sequence effects support the conflict theory of N2 and P3 in the Go/NoGo task. International Journal of Psychophysiology 75, 217–226. Sokhadze, E., Stewart, C., Hollifield, M., Tasman, A., 2008. Event-related potential study of executive dysfunctions in a speeded reaction task in cocaine addiction. Journal of Neurotherapy 12, 185–204. Tabachnick, B.G., Fidell, L.S., 1996. Using Multivariate Statistics, 3rd ed. HarperCollins, New York. Weisbrod, M., Kiefer, M., Marzinzik, F., Spitzer, M., 2000. Executive control is disturbed in schizophrenia: evidence from event-related potentials in a Go/NoGo task. Biological Psychiatry 47, 51–60.