Neuropsychologia 47 (2009) 2866–2875
Contents lists available at ScienceDirect
Neuropsychologia
journal homepage: www.elsevier.com/locate/neuropsychologia
Lesion neuroanatomy of the Sustained Attention to Response task
Pascal Molenberghs a , Céline R. Gillebert a , Hanne Schoofs b , Patrick Dupont a ,
Ronald Peeters c , Rik Vandenberghe a,b,∗
a
Cognitive Neurology Laboratory, Experimental Neurology Section, K.U. Leuven, Belgium
Neurology Department, University Hospitals Leuven, Herestraat 49, 3000 Leuven, Belgium
c
Radiology Department, University Hospitals Leuven, Herestraat 49, 3000 Leuven, Belgium
b
a r t i c l e
i n f o
Article history:
Received 19 February 2009
Received in revised form 20 May 2009
Accepted 14 June 2009
Available online 21 June 2009
Keywords:
Inferior frontal
Stroke
Cognitive control
Go no-go
a b s t r a c t
The Sustained Attention to Response task is a classical neuropsychological test that has been used by many
centres to characterize the attentional deficits in traumatic brain injury, ADHD, autism and other disorders.
During the SART a random series of digits 1–9 is presented repeatedly and subjects have to respond to
each digit (go trial) except the digit ‘3’ (no-go trial). Using voxel-based lesion symptom mapping (VLSM) in
a consecutive series of 44 ischemic unifocal non-lacunar hemispheric stroke patients we determined the
neuroanatomy of 4 SART parameters: commission and omission error rate, reaction time variability and
post-error slowing. Lesions of the right inferior frontal gyrus significantly increased commission error
rate. Lesions of the middle third of the right inferior frontal sulcus (IFS) reduced post-error slowing, a
measure of how well subjects can utilize errors to adjust cognitive resource allocation. Omissions and
reaction time variability had less localising value in our sample. To conclude, commission errors and
post-error slowing in the SART mainly probe right inferior frontal integrity.
© 2009 Elsevier Ltd. All rights reserved.
1. Introduction
The Sustained Attention to Response task (SART) is a conventional neuropsychological test for sustained attention and cognitive
control (Robertson, Manly, Andrade, Baddeley, & Yiend, 1997). In
the original version sequences of digits 1–9 are visually presented
in random order up to 25 times and subjects have to respond to
each digit except for the digit ‘3’. The SART differs from typical vigilance tasks in being relatively short and in requiring very frequent
responses and rare withholding of responses. It differs from typical go no-go task by the disproportion of go trials over no-go trials.
Over the past 15 years, the SART (Robertson, Manly, Andrade, et al.,
1997) has been intensively used in children and adults by a number of centres to analyze the attentional deficits arising in frequent
disorders such as traumatic brain injury (Chan, 2001; Robertson,
Manly, Andrade, et al., 1997; Whyte, Grieb-Neff, Gantz, & Polanksy,
2006), ADHD (Belgrove, Hawi, Gill, & Robertson, 2006; Johnson et
al., 2007; Shallice, Marzocchi, Coser, DelSavio, Meuter, & Rumiati,
2002), and autism (Johnson et al., 2007). The SART has been used
as an outcome measure for therapy (O’Connell, Bellgrove, Dockree,
Lau, Fitzgerald, & Robertson, 2008) and also to define an endophenotype for studying the genetics of attention (Greene, Bellgrove,
∗ Corresponding author at: Neurology Department, University Hospitals Leuven,
Herestraat 49, 3000 Leuven, Belgium. Tel.: +32 16 344280; fax: +32 16 3444285.
E-mail address: rik.vandenberghe@uz.kuleuven.ac.be (R. Vandenberghe).
0028-3932/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.neuropsychologia.2009.06.012
Gill, & Robertson, 2009). The neuroanatomy and electrophysiology has been studied in cognitively intact volunteers by means
of functional imaging (Fassbender et al., 2004; Manly et al., 2003)
and evoked potentials (Zordan, Sarlo, & Stablum, 2008). One element however that is lacking at the moment is a validation in
terms of brain lesion neuroanatomy. We prospectively recruited a
consecutive series of non-lacunar hemispheric stroke patients independently of lesion site to determine the cognitive neuroanatomy
of the SART. In a voxel-based manner we aimed to define which
brain regions are critically implicated in the SART using a similar
approach as we have applied in previous studies of spatial attentional deficits (Molenberghs, Gillebert, Peeters, & Vandenberghe,
2008).
Different behavioral measures that capture different aspects of
sustained attention and cognitive control can be derived from the
SART. SART omission errors and reaction time variability likely
relate to the ability to sustain attention across trials (Stuss, Shallice,
Alexander, & Picton, 1995; Wilkins, Shallice, & McCarthy, 1987).
Compared to more typical vigilance tasks which have a low target
frequency, the SART can pick up lapses of attention in a sensitive
way thanks to the high target frequency allowing to monitor the
attentional level nearly continuously during the entire task course.
These fluctuations will be manifest as increased reaction time variability or, if they are more severe, response omissions. Commission
errors in the SART may also reflect a sustained attention problem: when subjects fail to maintain the task goal on-line (Braver,
Reynolds, & Donaldson, 2003; Duncan, Emslie, & Williams, 1996),
P. Molenberghs et al. / Neuropsychologia 47 (2009) 2866–2875
an automatic response mode driven by the highly frequent go trials
may start to dominate behavior and subjects may neglect the task
requirement of perceptual identification of the digit prior to a go
response decision. Alternatively, commission errors may arise from
impaired inhibitory control (Drewe, 1975): subjects may be aware
of the task goal but the time needed to identify the stimulus may
exceed the period of time during which the prepotent go response
can be stopped (Aron, Robbins, & Poldrack, 2004; Logan & Cowan,
1984). Behaviorally, it is challenging to discriminate between these
two accounts. Manly et al. (2003) introduced a variant of the SART
in which digits are presented in a fixed series of 1–9 rather than in a
random sequence (Manly et al., 2003), increasing predictability. The
fixed version is more sensitive to frontal damage than the random
version (Manly et al., 2003). During the fixed version a commission
error is more typically preceded by electrophysiological markers of
a failure to maintain the task goal on-line (Braver et al., 2003), such
as increased alpha synchronization or a diminished late positivity potential (O’Connell et al., 2008). In contrast, commission errors
during the random version of the SART are associated with a diminished inhibitory event-related potential (ERP) complex (O’Connell
et al., 2008), indicative of a problem at the level of inhibitory control. A fourth measure of interest, post-error slowing (Fellows &
Farah, 2005; Picton et al., 2007; Rabbitt, 1966), reflects the ability to adjust cognitive resources on-line when a mismatch occurs
between actual outcome and task goal. Changes in this feedback
loop may arise at the level of error evaluation or error utilization
(Konow & Pribram, 1970; Scheffers, Coles, Bernstein, Gehring, &
Donchin, 1996; Stuss et al., 1995). Alternatively, post-error slowing
may also reflect a ‘hangover’ from erroneous processes engaged in
the previous trial.
From paradigms that resemble the SART we derived a couple of a
priori hypotheses. On the go no-go paradigm frontal patients make
more commission errors than non-frontal patients (Alexander,
Stuss, Picton, Shallice, & Gillingham, 2007; Drewe, 1975; Picton
et al., 2007) and show increased response time variability (Stuss,
Murphy, Binns, & Alexander, 2003). A related paradigm, the stopsignal reaction time (SSRT) (Logan & Cowan, 1984) paradigm,
measures the ability to interrupt a planned motor response. SSRT
is increased in patients with lesions of the right frontal operculum (Aron, Fletcher, Bullmore, Sahakian, & Robbins, 2003; Aron et
al., 2004). Regions outside prefrontal cortex may also contribute:
neglect patients are impaired on tasks that require the detection of
unfrequent targets in an auditory stream of distracters (Hjaltason,
Tegner, Tham, Levander, & Ericson, 1996; Husain & Rorden, 2003;
Robertson, Manly, Beschin, et al., 1997; Samuelsson, Hjelmquist,
Jensen, Ekholm, & Blomstrand, 1998), similarly to prefrontal lesion
patients (Richer & Lepage, 1996; Wilkins et al., 1987). The attentional dwell time, a measure of cognitive resource allocation, is also
increased in neglect patients (Husain, Shapiro, Martin, & Kennard,
1997). Since neglect is most often associated with right inferior
parietal lesions, this suggested to us that the right inferior parietal lobule contributes to SART performance in addition to its
well-established role in spatial attention (Husain & Rorden, 2003;
Mennemeier et al., 1994; Robertson, Manly, Beschin, et al., 1997;
Robertson, Mattingley, Rorden, & Driver, 1998; Vandenberghe,
Gitelman, Parrish, & Mesulam, 2001).
Functional magnetic resonance imaging (fMRI) experiments of
cognitive control in healthy volunteers provide an anatomically
more refined and more complicated picture than the lesion studies. fMRI studies implicate the junction between the inferior frontal
sulcus (IFS) and the inferior precentral sulcus (IFJ) (Derrfuss, Brass,
Neumann, & von Cramon, 2005; Hon, Epstein, Owen, & Duncan,
2006; Nakahara, Hayashi, Konishi, & Miyashita, 2002), the middle
third of the IFS (Derrfuss et al., 2005; Duncan & Owen, 2000), the
frontal operculum extending into the anterior insula (Duncan &
Owen, 2000), the dorsolateral prefrontal cortex (McDonald, Cohen,
2867
Stenger, & Carter, 2000; Sohn, Ursu, Anderson, Stenger, & Carter,
2000) and the anterior cingulate (Botvinick, Nystrom, Fissell, Carter,
& Cohen, 1999; Carter et al., 1998; Duncan & Owen, 2000; Kerns et
al., 2004; McDonald et al., 2000) in different cognitive control processes. fMRI studies of response inhibition have often found the
right inferior frontal gyrus (IFG) to be part of an extensive network
that includes IFS, anterior cingulate (Konishi et al., 1999; Konishi,
Nakajima, Uchida, Sekihara, & Miyashita, 1998) and also the inferior
parietal lobule (Bunge, Dudukovic, Thomason, Vaidya, & Gabrieli,
2002; Fassbender et al., 2004; Garavan, Ross, & Stein, 1999), particularly in case of go no-go compared to SSRT tasks (Garavan et al.,
1999; Rubia et al., 2001).
In the majority of previous lesion studies of cognitive control,
lesions were classified using a region-based template (Alexander
et al., 2007; Alexander, Stuss, Shallice, Picton, & Gillingham, 2005;
Aron et al., 2003, 2004; Drewe, 1975; Fellows & Farah, 2005; Picton
et al., 2007; Wilkins et al., 1987). The functional demarcations
however within prefrontal cortex cannot be directly derived from
structural anatomical boundaries (Stuss et al., 1995). Voxel-based
lesion-symptom mapping (VLSM) allows one to quantify a deficit
on a continuous scale and relate the degree of impairment to lesion
site in a voxelwise manner (Bates et al., 2003; Rorden, Karnath,
& Bonilha, 2007) without need for an a priori region-based template. VLSM provides coordinates of critical lesion sites that can be
directly compared to those obtained using fMRI in healthy volunteers (Molenberghs et al., 2008).
We examined which SART parameters are linked to specific
cortical lesion sites (Husain & Rorden, 2003; Robertson, Manly,
Andrade, et al., 1997; Robertson et al., 1998). We predicted that
apart from prefrontal cortex, the inferior parietal lobule would also
be involved (Garavan et al., 1999; Hjaltason et al., 1996; Robertson,
Manly, Andrade, et al., 1997; Rubia et al., 2001; Samuelsson et al.,
1998).
2. Methods
2.1. Subjects
All participants gave written informed consent in accordance with the Declaration of Helsinki. The ethical commission, University Hospitals Leuven, approved the
experimental protocol.
We screened a consecutive series of 616 ischemic stroke patients during their
hospitalization at the stroke unit or on occasion of their first follow-up visit to the
outpatient clinic. Fifty-six of these patients fulfilled the study criteria, 44 of whom
consented to the study (Table 1). All participants had suffered a recent non-lacunar
unifocal ischemic hemispheric stroke, confirmed on clinical Fluid Attenuation Inversion Recovery (FLAIR) or Diffusion-Weighted Imaging (DWI) magnetic resonance
imaging (MRI). Patients were recruited via the acute stroke unit (n = 29) or via their
first follow-up visit to the outpatient stroke clinic (n = 15). Twenty of the patients
also participated in a study of spatial attention reported previously (Molenberghs
et al., 2008). Exclusion criteria were age above 85 years, pre-existing periventricular or subcortical white matter lesions or a pre-existing stroke on MRI, insufficient
balance to sit autonomously and general inability to understand and carry out a
computerized task. The anatomical distribution of the ischemic lesions is shown in
Fig. 1. Visual fields were intact except in case 10 (left hemianopia), 14 (right lower
quadrantanopia), 15 (left upper quadrantanopia) and 32 (left lower quadrantanopia).
A control group of 19 healthy subjects (10 women), aged between 51 and 73
years (mean 61.2 years, S.D. 7.1) underwent the same protocol.
2.2. Stimuli and tasks
Subjects were seated at 114 cm distance from a 19 in. computer screen. Using
Superlab for PC version 2.0 (Cedrus, Phoenix, AR, USA) each of nine digits was foveally
presented 25 times in pseudorandom order over a total period of 4.3 min (Fig. 2). In
the version used in the current study the digits were randomized across the entire
run rather than per sequences of 1–9. Each digit (luminance = 105 cd/m2 ) was presented for 250 ms on a black background, followed by a 900 ms white mask, after
which the next trial started immediately. This resulted in a total task duration of
4 min 19 s. The mask consisted of a ring (diameter = 5.51◦ ; width = 105 cd/m2 ) with
a diagonal cross in the middle. The digits were presented in one of five randomly
allocated font sizes (48, 72, 94, 100 or 120 points symbol font size; height ranging between 2.28 and 5.51 visual degrees). Subjects were instructed to respond to
each digit with a key press (response box from Cedrus, Phoenix, AR, USA) except to
2868
P. Molenberghs et al. / Neuropsychologia 47 (2009) 2866–2875
Table 1
Behavioral parameters
Case
Age
Side
Size (cm3 )
Time (days)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
43
82
44
69
53
88
72
65
80
74
73
79
79
47
52
68
64
79
75
74
84
61
37
76
79
65
62
37
42
54
42
64
77
34
66
55
64
61
62
35
60
44
71
80
R
R
R
L
L
R
L
R
R
R
L
L
L
L
R
R
R
R
R
R
L
L
L
L
R
R
R
R
R
R
L
R
L
L
L
R
R
L
L
L
R
R
L
R
26.9
20.2
302.7
19.0
108.0
84.1
46.8
17.0
29.8
173.0
16.4
4.8
2.1
13.9
14.3
11.0
216.0
191.0
15.4
117.0
12.5
1.0
11.2
4.0
40.8
49.5
89.7
84.8
43.4
30.2
13.8
197.0
17.2
64.9
95.1
2.6
107.0
18.5
17.0
64.4
29.6
161.0
25.8
64.6
4
5
4
6
4
7
3
5
6
6
4
3
6
5
147
154
5
4
3
7
6
217
21
5
14
10
4
14
6
5
133
196
126
168
126
140
196
7
133
63
168
91
14
126
Bells
Bisection (%)
L/M/R
0/0/1
1/0/0
1/0/1
0/0/0
3/0/3
2/2/4
2/4/0
2/0/0
0/0/0
14/0/1
0/0/0
2/1/1
0/0/1
0/1/0
2/1/0
0/0/0
15/4/2
15/4/0
2/1/0
0/0/1
0/0/0
0/0/0
0/0/0
1/1/2
2/1/1
1/1/1
2/0/0
2/0/1
4/3/1
2/0/0
2/1/1
2/0/0
0/0/1
0/1/0
1/1/2
1/0/0
3/0/1
0/0/1
0/0/0
0/0/0
1/1/0
0/0/0
1/0/1
3/1/0
+5.8
+3.7
−5.3
+0.7
+4.1
+8.1
−1.7
+5.8
+1.2
+20.5
−0.3
+1.9
+1.7
+3.8
−5.9
−3.0
+18.4
+33.4
+1.2
−1.7
+9.6
+2.0
+0.4
+4.6
−3.8
+4.0
+5.3
+0.9
+18.7
+6.6
+2.1
−5.3
−5.9
3.9
+0.5
−3.6
+4.1
+5.2
+0.4
+0.1
−1.8
+5.0
+2.1
+6.3
Sustained Attention to Response task
comm./25
omiss./200
post (ms)
mean RT (ms)
S.D./mean (ms)
2
5
8
7
12
2
1
2
3
3
6
9
7
3
7
2
8
6
8
1
5
5
7
5
2
0
1
15
8
7
8
17
6
9
6
7
18
12
11
11
2
18
2
1
6
12
2
10
4
3
8
1
17
2
15
5
5
3
20
23
2
8
12
3
10
2
5
2
4
1
5
9
8
7
6
4
1
2
5
5
3
3
1
2
1
8
4
13
105
288
−20
276
−45
31
114
7
94
117
8
62
−37
124
51
79
38
−19
−16
81
−47
21
47
101
94
no errors
10
−54
104
102
−8
38
6
11
318
260
64
103
74
−46
234
1
62
−31
408
526
485
473
580
655
704
414
508
363
625
471
638
647
415
486
539
465
308
581
400
423
346
558
625
501
483
392
661
421
404
491
502
360
389
460
477
422
414
332
533
490
613
521
0.23
0.34
0.17
0.38
0.15
0.15
0.27
0.35
0.34
0.25
0.32
0.21
0.21
0.19
0.25
0.31
0.17
0.20
0.15
0.15
0.20
0.20
0.16
0.16
0.13
0.30
0.14
0.43
0.42
0.31
0.23
0.18
0.23
0.21
0.25
0.36
0.17
0.26
0.35
0.21
0.19
0.46
0.17
0.28
Bold: Significantly different from normals at P < 0.05 (Crawford & Howell, 1998). Legend: side: hemispheric side of the lesion. L: Left. R: Right. M: Middle. Size: Lesion size.
time = time to stroke onset. post = post error slowing. comm.: total number of commission errors; omiss.: total number of omission errors.
the digit 3. In case of a “3” they had to withhold from responding. Throughout the
experiment subjects used their preferred hand. Subjects were asked to give equal
importance to accuracy and speed in doing the task. Each subject started with a
practice run consisting of 16 go and 2 no-go trials.
Our 2 accuracy measures (Table 1) were:
1. Total number of commission errors: Key press responses to a digit “3”.
2. Total number of omission errors: Missed responses to a digit other than 3.
Our 2 reaction time (RT) measures (Stuss et al., 2003; Picton et al., 2007) were:
1. Post-error slowing: RT during the trial following a commission error minus that
preceding the commission error, divided by the mean RT in that subject.
2. RT variability: Standard deviation of the RTs on correct response trials, divided
by the mean RT in that subject
Subjects also received a more extensive conventional neuropsychological test
battery consisting, among other tests, of target cancellation (Bells test) (Gauthier,
Dehaut, & Joanette, 1989), line bisection (Schenkenberg, Bradford, & Ajax, 1980),
the number location test from the Visual Object Space Perception Battery (VOSP)
(Warrington & James, 1991), Coloured Progressive Matrices (Raven, Court, & Raven,
1995), visual extinction (Bender, 1952), BORB (Riddoch & Humphreys, 1993) and
digit span (Wechsler, 1998).
2.3. Image acquisition and preprocessing
A 3 T Philips Intera system (Best, Netherlands) equipped with a head volume
coil provided T1 images (TR = 1975 ms, TE = 30 ms, in-plane resolution 1 mm) as
well as Fluid Attenuation Inversion Recovery (FLAIR) 3D images (TR = 10741 ms,
TE = 150 ms) in each patient. Using SPM2 (http://www.fil.ion.ucl.ac.uk, Welcome
Trust Centre for Neuroimaging, London, UK) the T1 and FLAIR images were coregistered. The T1 scan was normalized to the Montreal Neurological Institute
(MNI) T1 template in Talairach space (Friston et al., 1995; Talairach & Tournoux,
1988). The spatial normalization involved both linear (12 affine transformations) and nonlinear (7 × 9 × 7 basis functions, 16 reiterations) transformations
(Ashburner & Friston, 1999). High regularization was used to constrain the nonlinear part of the algorithm and penalize unlikely deformations associated with
the presence of lesions (Ashburner & Friston, 1999; Tyler, Marslen-Wilson, &
Stamatakis, 2005). The same normalization matrix was applied to the FLAIR
images. The match between each patient’s normalized brain and the brain template was carefully evaluated through visual inspection and use of a cross-hair
yoked between the template image and the normalized image. After verification of the normalization, lesions were semi-automatically delineated using
MRIcro version 1.37 (http://www.sph.sc.edu/comd/rorden/mricro.html) and manually set intensity thresholds (Rorden et al., 2007). Subsequently the lesion
volumes were imported into the MRIcron lesion-symptom mapping software
(http://www.sph.sc.edu/comd/rorden/mricron). A voxel was included in the analysis
only if it was lesioned in at least 4 of the subjects.
P. Molenberghs et al. / Neuropsychologia 47 (2009) 2866–2875
2869
Fig. 1. Total group. Lesion distribution volume. The color code indicates in how many individuals of our patient sample (n = 44) a given voxel was lesioned (ranging from 1 to 13).
(A) Transverse sections. (B) Projection on the Population-Average, Landmark and Surface based (PALS) (van Essen, 2005) atlas using Computerized Anatomical Reconstruction
and Editing Toolkit (sideview and ventral view) (CARET, Washington University School of Medicine, Department of Anatomy and Neurobiology, http://brainmap.wustl.edu(van
Essen et al., 2001)).
2.4. Statistical analysis
3. Results
2.4.1. Factor analysis of neuropsychological test scores
The neuropsychological data were subjected to a principal components factor analysis with orthogonal rotation (Statistical Package for the Social Sciences,
http://www.spss.com/spss/). We included the 4 SART parameters (commission
error rate, omission error rate, post-error slowing and within-subject RT variability) as well as contra- versus ipsilesional omissions on the Bells test,
mean percentage deviation on the line bisection (ipsilesional deviation being
coded positively), and the score on the number location test of the VOSP
(Table 1).
3.1. Factor analysis
2.4.2. Voxel-based lesion symptom mapping of SART
Each of the 4 SART parameters was entered separately into a VLSM analysis
(Rorden et al., 2007). We examined which voxels when lesioned were associated with
significantly worse scores compared to patients in whom these voxels were intact
(Brunner and Munzel t test; Brunner & Munzel, 2000). Our significance threshold
was set at P < 0.05, with a Bonferroni correction for the brain search volume (FamilyWise Error correction; Rorden et al., 2007). When VLSM yielded a significant effect
in a particular region, we determined the origin of the finding in further detail: we
evaluated the average scores in the patient group who had a lesion in that region
and the group who did not have a lesion in that region. Furthermore, we evaluated
how many of the individuals with a lesion in that region had a significant deficit at
the individual level compared to cognitively intact controls using a modified t test
(Crawford & Howell, 1998).
2.4.3. Linear regression analyses
We conducted a multiple linear regression with age, lesion extent and time to
stroke onset as independent variables and the different SART parameters as dependent variables.
We also conducted two simple linear regression analyses with reaction times
as independent variable and omission error rate and commission error rate, respectively, as dependent variables.
In the total study population, a factor analysis of the neuropsychological data revealed 3 factors with an Eigenvalue higher than 1:
the first factor explained 31.0% of the total variance (Eigenvalue =
2.17) and clustered the target cancellation (r = 0.94), line bisection
(r = 0.86), and the number location score (r = 0.72). The second
factor (Eigenvalue = 1.36) explained 18.7% of the total variance
and clustered RT variability (r = 0.84) and omission error rate
(r = 0.63). A third factor (Eigenvalue = 1.16) explained 13.9% of the
variance and clustered the commission error rate (r = 0.85) and
post-error slowing (r = −0.61).
3.2. VLSM of SART
When we used commission error rate as input for the VLSM analysis, lesions of the right IFG (centre of mass x = 43, y = 25, z = 14,
ext. 16,586 mm3 , P < 0.05) were associated with significantly more
commission errors than was seen in patients in whom this region
was intact (Fig. 3A). Six patients (cases 3, 17, 28, 32, 37, 42) had
a lesion that overlapped with this right IFG region. Their average
commission error rate was 14.0 (S.D. 4.77) whereas the average
commission error rate in the remainder of the patients was 5.29
(S.D. 3.36), similarly to what we saw in controls (mean 7.6, S.D.
2.53). At the individual level, 4 out of 6 patients with a right IFG
lesion (cases 28, 32, 37, 42) had a significantly higher commis-
2870
P. Molenberghs et al. / Neuropsychologia 47 (2009) 2866–2875
Fig. 2. Experimental paradigm. In the main experiment subjects were instructed to
respond to each digit with a key press except to the digit 3 (no-go trial).
sion error rate than controls (P < 0.05; Crawford & Howell, 1998)
(Fig. 4).
In order to examine the apparent right-sided laterality we
flipped the right IFG VLSM result and examined whether the flipped
region belonged to the VLSM lesion distribution volume and how
subjects behaved who had a left-sided lesion that overlapped with
the flipped region. Four left-hemispheric patients (cases 5, 7, 35, 40)
had a lesion that overlapped with the flipped region. The average
commission error rate in these 4 patients was 7.5 (S.D. 5.06) closely
similar to the normal values. At the individual level, none of the
patients with a left IFG lesion showed a pathological commission
error rate (Crawford & Howell, 1998).
Overall, left-hemispheric lesions (29.3 voxels, S.D. 32.5) were
significantly smaller than right-hemispheric lesions (84.7 voxels,
S.D. 79.6) (P < 0.01) (Figs. 1 and 5). This could cause a spurious
laterality effect in favor of the right hemisphere. For that reason
we conducted a VLSM analysis with the left-hemispheric patients
only. This did not yield any significant effect in the left IFG, even if
we lowered the significance threshold to an uncorrected P < 0.05.
A similar analysis with the right-sided lesion patients confirmed
the results from the main analysis (corrected P < 0.05). Within the
right-hemispheric group, commission error rate tended to correlate with lesion size (Pearson r = 0.376, P = 0.06) but this could
not account for the right IFG result since patients with similar-sized
lesions sparing the right IFG did not show an increased commission
error rate (Fig. 5).
When we used post-error slowing as input for the VLSM analysis, lesions of the middle third of the right IFS (centre of mass
x = 35 mm, y = 33 mm, z = 27 mm, ext. 718 mm3 , P < 0.05) were
associated with a significant decrease in post-error slowing compared to what was seen in patients in whom this region was intact
(Fig. 3B). Six patients (cases 3, 17, 18, 28, 32 and 42) had a lesion
overlapping with this right IFS lesion. They responded as fast after
a commission error (479 ms, S.D. 84) as before (481 ms, S.D. 58)
while patients who did not have an IFS lesion showed post-error
slowing (post-commission error: 539 ms (S.D. 125), prior to commission error: 463 ms (S.D. 116)) similar in degree to what we saw
in the healthy controls (post-error: 477 ms, S.D. 206; prior to error
360 ms, S.D. 47) (Fig. 6). At the individual level, the reduction in
post-error slowing did not reach significance in any of the 6 patients
compared to controls (Crawford & Howell, 1998).
The functional dissociation between the right IFG and the right
middle IFS was confirmed when we lowered the threshold of the
VLSM analysis to an uncorrected P < 0.001: even at this lower
threshold, commission error rate and post-error slowing localised
to two nearby but anatomically distinct areas.
If we used the difference in reaction times before and after a
commission error as VLSM input without dividing by the individual’s mean RT, results were entirely comparable (corrected P <
0.05). At a slightly lower significance threshold, this was also true
if the reaction times following a correctly withheld response rather
than those preceding a commission error were subtracted from the
reaction times following a commission error.
VLSM analysis of omission error rate and RT variability did not
yield any significant results at the pre-set significance threshold.
In order to evaluate whether the absence of any parietal involvement in the SART was due to lack of sensitivity for parietal
involvement in the specific patient sample examined, we tested
whether a VLSM with the spatial bias measure of the target
cancellation task as input yielded the predicted parietal effects
(Molenberghs et al., 2008; Mort et al., 2003). Lesions of right inferior
parietal cortex extending into the lower bank of IPS were associated
with a significantly higher rightward bias than what was seen in the
absence of inferior parietal lesions.
Our a priori measures of sustained attention did not capture
performance changes over time. In order to evaluate whether this
explained the absence of parietal involvement, we calculated for
each subject the difference between commission error rate in the
first 12 no-go trials and in the last 12 no-go trials. Likewise, for
the omission error rate and the reaction time variability we calculated the difference between the first 100 go trials and the last
100 go trials. Each of these measures were used as input for VLSM
but none of these analyses yielded any inferior parietal effect,
even if we lowered the threshold to an uncorrected P < 0.05. Furthermore we performed a linear regression analysis with reaction
time as dependent variable and trial position as the independent
variable and used the regression coefficient as input for VLSM.
No parietal effect was seen, even when the threshold was lowered at an uncorrected P < 0.05. In a final effort to relate any
changes in task performance over time to inferior parietal cortex,
we divided the trial series into 5 epochs (one epoch per 45 trials) and conducted a linear regression analysis with epoch order
as independent variable (1–5) and commission or omission error
rates per 45-trials epoch as dependent variable. These coefficients
were used as input for VLSM. No significant effects were obtained in
IPL.
3.3. Linear regression analyses
A multiple linear regression analysis with commission error rate
as outcome and with age, lesion size and time-to-stroke onset
as independent variables showed a significant effect (F(3, 40) =
P. Molenberghs et al. / Neuropsychologia 47 (2009) 2866–2875
2871
Fig. 3. Total group. Lesion-symptom maps thresholded at Bonferroni corr. P < 0.05 (Brunner-Munzel test). (A) VLSM with post-error slowing as input. (B) VLSM with
commission error rate as input.
5.39, P < 0.005). Surprisingly, commission error rate correlated
significantly and inversely with age (r = −0.33, P = 0.021). The correlation with lesion size (r = 0.27, P = 0.056) and time-to-stroke
onset (r = 0.25, P = 0.079) did not reach significance.
Post-error slowing, reaction time variability and omission error
rate did not correlate with age, lesion size or time-to-stroke onset
(F(3, 39) = 0.90, P = 0.45, F(3, 40) = 1.11, P = 0.36 and F(3, 40) =
1.08, P = 0.37, respectively).
According to a linear regression analysis with reaction times as
independent variable and commission error rate as dependent variable, commission errors tended to be more frequent with faster
reaction times (Pearson correlation coefficient −0.30, P = 0.051).
Fig. 4. (A–D) Lesions of the four subjects with a pathological commission error rate on the SART.
2872
P. Molenberghs et al. / Neuropsychologia 47 (2009) 2866–2875
of go trials. When however no-go trials had a low probability, as in
the original SART version, they failed to withhold their responses.
Fig. 5. Regression plot. X-axis: Lesion size. Y-axis: Commission error rate. Legend:
Empty squares: left-hemispheric patients. Full circles: Right-hemispheric patients.
The datapoints from subjects with a lesion that overlapped with the right IFG result
are labelled with their case number. Scores that fall above the dashed line are pathological compared to healthy controls.
A similar analysis with omission error rate as dependent variable
remained far below significance.
3.4. Ancillary experiments
3.4.1. Commission errors
We investigated whether the pathological commission error rate
in case of a right-sided IFG lesion was due to the low probability of the ‘no-go’ events, the ‘no-go’ character of those events, or
a combination of both. We contacted each of the 7 cases who had
a right IFG lesion. Cases 3 and 42 were available for re-testing, 4
and 2 years after the first evaluation, respectively. The tasks were
the SART using a randomization scheme per sequence of 1–9 and
a variant of the SART in which subjects had to respond to a digit
‘3’ but not to any of the other digits. Subjects first performed the
original SART version, subsequently an unrelated visuoperceptual
task (45 min) and the SART variant thereafter.
On the original SART version cases 3 and 42 made 18 and 15 commission errors, respectively, out of 25 no-go trials, confirming the
results obtained in the main study. On the SART variant with proportionally more no-go trials, case 3 made only one commission error
out of 200 no-go trials and one omission error out of 25 go trials.
Case 42 made one commission error and no omission errors. These
subjects therefore were able to correctly withhold responses to nogo trials when the no-go trials were highly expected and were also
able to correctly respond to low-probability trials if these consisted
Fig. 6. Mean reaction times before and after a commission error in controls, patients
with and patients without an IFS lesion. Light gray: Mean reaction time in trials
preceding a commission error (and S.D.). Dark gray: Mean reaction time in trials
following a commission error (and S.D.).
3.4.2. Post-error slowing
We conducted two further variants of the SART to further elucidate the source of the reduced post-error slowing. Cases 3 and
42 each participated in one of these variants, which was performed after another unrelated visuoperceptual task (duration
5 min). Stimuli and task were identical to those of the original SART
(i.e. high-probability go trials and low-probability no-go trials) with
one exception: we instructed case 3 to pronounce the word ‘hit’
when he had made a commission error to a no-go trial. Case 3 made
22 commission errors and pronounced the word ‘hit’ in 20 of those
22 trials. He also pronounced ‘hit’ to the 3 no-go trials where he
correctly withheld responses and never did in go trials. The subject
therefore gave proof of knowledge of a distinction between the digit
‘3’ trials and the other ones and was able to use this knowledge to
positively respond to the ‘3’ trials in a discriminate way even when
he was not able to withhold his response to this type of trial.
Finally, in case 42 we provided auditory feedback with a high
tone for correct responses and a low tone for errors to examine
whether increased error awareness was able to restore post-error
slowing. When case 42 was tested on this version, he made 18 commission errors but did not show post-error slowing even when
negative feedback was provided explicitly (mean RT pre-error:
545 ms, S.D. 99; mean RT post-error: 493 ms, S.D. 99).
4. Discussion
Accurate performance of the SART in the current patient sample
relied principally on the integrity of right inferior frontal cortex, in
agreement with our first a priori hypothesis. Our data did not provide evidence for our second hypothesis regarding the contribution
of the right inferior parietal lobule. The two outcome measures with
most localising value were the commission error rate and post-error
slowing. Our two other outcome measures, omission error rate and
reaction time variability, were not linked to any particular lesion
site.
The reduction in post-error slowing with IFS lesions cannot
be accounted for by a ‘hang-over’ from the erroneous processes
engaged in the previous trial. If that were the case, one would expect
an increase in post-error slowing in patients compared to controls
rather than a decrease.
The right-sided laterality must be interpreted with caution. The
average lesion size was smaller in left than in right hemisphere
patients due to exclusion of patients with significant language
comprehension deficits. This may bias the sensitivity of the VLSM
analysis in favor of the right hemisphere. When we conducted a
VLSM analysis separately in the left and in the right hemisphere
patients, this confirmed the right-sided effect in the absence of a
significant left IFG effect. Previous studies have also suggested that
right IFG lesions may cause more response inhibition problems than
left IFG lesions (Aron et al., 2004).
Our consecutive series did not include cases of anterior cerebral
artery ischemia. Our data therefore are neutral to the possible contribution of anterior cingulate or other medial frontal regions to
cognitive control (Alexander et al., 2005, 2007; Fellows & Farah,
2005; Picton et al., 2007). Medial frontal lesions may cause an
increase in both omission (Picton et al., 2007) and commission
errors during a go-no go task (Drewe, 1975) as well as increased
response variability (Picton et al., 2007). According to single neuron recording studies (Schall, Stuphorn, & Brown, 2002), patient
lesion studies (Swick & Turken, 2002) and fMRI of the intact brain
(Garavan, Ross, Kaufman, & Stein, 2003; Lütcke & Frahm, 2008) the
anterior cingulate plays a role in monitoring of conflict (Botvinick
P. Molenberghs et al. / Neuropsychologia 47 (2009) 2866–2875
et al., 1999; Carter et al., 1998; McDonald et al., 2000) and error. In a
case study of an anterior cingulate lesion, error-related negativity,
an event-related potential thought to reflect the activity of a neural
system responsible for error detection (Scheffers et al., 1996), was
abolished (Swick & Turken, 2002).
Inferior parietal lesions were not associated with changes in
any of the SART parameters compared to other lesion sites, contrary to our a priori hypothesis (Hjaltason et al., 1996; Robertson,
Manly, Beschin, et al., 1997; Samuelsson et al., 1998). In previous studies (Hjaltason et al., 1996; Robertson, Manly, Beschin, et
al., 1997; Samuelsson et al., 1998), patients were selected on the
basis of the presence of a clinical neglect syndrome, regardless of
lesion extent or site. The lesion in individual neglect patients however often extends beyond the parietal lobe. Possibly, the sustained
attention deficits seen in neglect patients are due to collateral rightsided inferior frontal damage. Alternatively, if response variability
or other measures of sustained attention rely on a brain circuit
that includes other areas besides IPL, patients with IPL lesions may
not be significantly different on this parameter from patients with
lesions of this circuit outside IPL. Conceivably, variations of the current paradigm could modify the exact sustained attention demands
increasing the sensitivity of this task for parietal damage but this
remains to be tested empirically. For instance, had we reversed the
go no-go rule, the need to inhibit automatic responses would have
been much reduced and the relative contribution of inferior frontal
versus inferior parietal cortex might have been reversed. The SART
requires subjects to attend to the identity of the digit rather than its
location (Malhotra, Coulthard, & Husain, 2009). Had we placed the
digits at peripheral locations and requested the subjects to monitor stimulus location rather digit identity, we would predict that
right inferior parietal involvement would have been much stronger:
According to a recent lesion study in neglect patients (Malhotra et
al., 2009), right IPL lesions impair sustained attention mainly when
attention is directed to locations compared to object identity. This
is in agreement with a previous fMRI study implicating the right
angular gyrus in spatial short-term memory (Vandenberghe et al.,
2001). Had we presented the digits at varying peripheral locations
and instructed subjects to maintain attention to a selection of these
locations, right inferior parietal involvement could have been much
more prominent (Malhotra et al., 2009).
We did not obtain any significant lesion correlate for intrasubject
response variability. This differs from an earlier region-based lesion
study implicating the lateral convexity and medial wall of prefrontal
cortex in increased individual performance variability (Stuss et al.,
2003) during forced-choice simple target detection tasks and even
more so during more complex tasks, such as detection of a target
defined by feature conjunction (Stuss et al., 2003). In that study only
orbitofrontal lesions did not affect RT variability. The effect of prefrontal lesions on RT variability in previous studies was principally
based on comparisons with normal control populations (Picton et
al., 2007; Stuss et al., 2003). If RT variability can be affected by a
relatively wide range of prefrontal regions, it may not be detected
by VLSM since VLSM compares the measure of interest between
patients who have a lesion in a given voxel to those in whom the
voxel is spared: the comparison group, i.e. the patients in whom a
given voxel is intact, may also show a deficit of this parameter.
Our findings confirm the critical role of the right IFG in inhibitory
control (Bunge et al., 2002; Fassbender et al., 2004; Garavan et
al., 1999; Konishi et al., 1998, 1999). In the SSRT paradigm, subjects with right inferior opercular lesions are able to abort their
responses only when the delay interval between go- and stop-signal
is relatively brief compared to controls or patients with lesions elsewhere in prefrontal cortex. The extent of the right inferior frontal
lesion correlates positively with SSRT (Aron et al., 2003, 2004).
In healthy controls, transcranial magnetic stimulation of the right
frontal operculum shortens the delay between go- and stop-signal
2873
that is required to interrupt the go-response, in the absence of any
effect upon go trials (Chambers et al., 2006). Because no-go trials
are relatively rare in the SART and the intertrial interval is fixed,
subjects will have a propensity to respond to each trial. Identification of a “3” may be analogous to a stop signal. In patients with
right inferior frontal lesions the time required to identify a stimulus
may exceed that needed to interrupt the motor plan, resulting in a
commission error.
Post-error slowing relies on a reaction time contrast between
two temporally juxtaposed trials within a same subject. Post-error
slowing can be calculated in a variety of ways, e.g. by subtracting
either the reaction time prior to the false-positive trial or by subtracting the reaction time following a correctly withheld response.
One can normalize the post-error slowing by mean response time
or not. Whichever formula is used, the VLSM results were closely
similar. Post-error slowing was significantly reduced in patients
with lesions confined to the middle third of the right IFS (Fig. 3),
superiorly and anteriorly to the relatively large IFG region implicated in commission errors. In a previous region-based study,
post-error slowing was reversed in patients with left dorsolateral prefrontal lesions (more rapid responses following errors) and
reduced in patients with right dorsolateral prefrontal lesions (Stuss
et al., 2003). Several of the lesions in that study included the right
IFS (Stuss et al., 2003). Reduced post-error slowing results from
alterations of a feedback loop that encompasses the evaluation of
mismatches between actual outcome and task goal and the adaptation of behavioral schemata (Shallice, 1988) on the basis of the
mismatch. In a classical case study, a frontal lesion patient was
able to verbally report errors he or others made but failed to use
these errors to modify his behavior (Konow & Pribram, 1970). A set
of follow-up experiments we carried out provides evidence that
subjects were aware of the distinction between go and no-go trials. Even when they received feedback, they did not slow their
responses, compatible with a deficiency in error utilization (Konow
& Pribram, 1970) rather than error evaluation.
According to a recent and influential model of the neuroanatomy
of spatial and nonspatial attention (Corbetta & Shulman, 2002), a
ventral network consisting of the right temporoparietal junction,
inferior and middle frontal gyrus, is implicated in stimulus-driven
reorienting. This ventral network has been proposed as an alerting
system that detects novel stimuli (Corbetta & Shulman, 2002). Our
data call for some modification of this model. The contribution of
IFS in our study was, strictly speaking, not exogenously driven: it
was driven by the error made in response to the no-go stimulus.
The reduction of post-error slowing following lesions of the middle
third of the right IFS provides strong evidence that this region is
involved in on-line readjustment of attentional resources and that
the role of this region is not limited to exogenous, stimulus-driven
reorienting conditions.
To conclude, commission error rate and post-error slowing in the
random version of the SART rely in particular on the integrity of the
right inferior frontal cortex. We did not find any positive arguments
in favor of a contribution of the inferior parietal lobule.
Acknowledgements
Supported by FWO grants G.0076.02 and G0668.07 (EuroCores)
(R.V.), KU Leuven Research grants OT/04/41 and EF/05/014 (R.V.),
and Inter-University Attraction Pole P6/29. R.V. is a Clinical Investigator of the Fund for Scientific Research (FW0), Flanders (Belgium),
and CRG an FWO research fellow.
References
Alexander, M., Stuss, D., Picton, T., Shallice, T., & Gillingham, S. (2007). Regional
frontal injuries cause distinct impairments in cognitive control. Neurology, 68,
1515–1523.
2874
P. Molenberghs et al. / Neuropsychologia 47 (2009) 2866–2875
Alexander, M., Stuss, D., Shallice, T., Picton, T., & Gillingham, S. (2005). Impaired concentration due to frontal lobe damage from two distinct lesion sites. Neurology,
65, 572–579.
Aron, A., Fletcher, P., Bullmore, E., Sahakian, B., & Robbins, T. (2003). Stop-signal
inhibition disrupted by damage to right inferior frontal gyrus in humans. Nature
Neuroscience, 6, 115–116.
Aron, A., Robbins, T., & Poldrack, R. (2004). Inhibition and the right inferior frontal
cortex. Trends of Cognitive Neuroscience, 8, 170–177.
Ashburner, J., & Friston, K. (1999). Nonlinear spatial normalization using basis functions. Human Brain Mapping, 7, 254–266.
Bates, E., Wilson, S. M., Saygin, A. P., Dick, F., Sereno, M. I., Knight, R. T. ,., et al. (2003).
Voxel-based lesion-symptom mapping. Nature Neuroscience, 6, 448–450.
Belgrove, M., Hawi, Z., Gill, M., & Robertson, I. (2006). The cognitive genetics of
attention deficit hyperactivity disorder: Sustained attention as a candidate phenotype. Cortex, 42, 838–845.
Bender, M. (1952). Disorders in perception (with particular reference to the phenomena
of extinction and displacement). Springfield, IL: C. C. Thomas.
Botvinick, M., Nystrom, L., Fissell, K., Carter, C., & Cohen, J. (1999). Conflict monitoring
versus selection-for-action in anterior cingulate cortex. Nature, 402, 179–181.
Braver, T., Reynolds, J., & Donaldson, D. (2003). Neural mechanisms of transient and
sustained cognitive control during task switching. Neuron, 39, 713–726.
Brunner, E., & Munzel, U. (2000). The nonparametric Behrens-Fisher problem:
Asymptotic theory and a small-sample approximation. Biometric Journal, 1,
17–25.
Bunge, S., Dudukovic, N., Thomason, M., Vaidya, C., & Gabrieli, J. (2002). Immature
frontal lobe contributions to cognitive control in children: Evidence from fMRI.
Neuron, 33, 301–311.
Carter, C., Braver, T., Barch, D., Botvinick, M., Noll, D., & Cohen, J. (1998). Anterior
cingulate cortex, error detection, and the online monitoring of performance.
Science, 280, 747–749.
Chambers, C., Bellgrove, M., Stokes, M., Henderson, T., Garavan, H., Robertson, I. ,.,
et al. (2006). Executive “brake failure” following deactivation of human frontal
lobe. Journal of Cognitive Neuroscience, 18, 444–455.
Chan, R. (2001). A further study on the sustained attention to response task (SART):
The effect of age, gender and education. Brain Injury, 15, 819–829.
Corbetta, M., & Shulman, G. (2002). Control of goal-directed and stimulus-driven
attention in the brain. Nature Review Neuroscience, 3, 201–215.
Crawford, J., & Howell, D. (1998). Comparing an individual’s test score against norms
derived from small samples. Clinical Neuropsychology, 12, 482–486.
Derrfuss, J., Brass, M., Neumann, J., & von Cramon, D. (2005). Involvement of the
inferior frontal junction in cognitive control: Meta-analyses of switching and
Stroop studies. Human Brain Mapping, 25, 22–34.
Drewe, E. (1975). An experimental investigation of Luria’s theory on the effects of
frontal lobe lesions in man. Neuropsychologia, 13, 421–429.
Duncan, J., Emslie, H., & Williams, P. (1996). Intelligence and the frontal lobe: The
organization of goal-directed behavior. Cognitive Psychology, 30, 257–303.
Duncan, J., & Owen, A. M. (2000). Common regions of the human frontal lobe
recruited by diverse cognitive demands. Trends in Neurosciences, 23, 475–483.
Fassbender, C., Murphy, K., Foxe, J. J., Wylie, G., Javitt, D. C., & Robertson, I. (2004). A
topography of executive functions and their interactions revealed by functional
magnetic resonance imaging. Cognitive Brain Research, 20(2), 132–143.
Fellows, L., & Farah, M. (2005). Is anterior cingulate cortex necessary for cognitive
control? Brain, 128, 788–796.
Friston, K., Holmes, A., Worsley, K., Poline, J., Frith, C., Heather, J. ,., et al. (1995).
Statistical parametric maps in functional imaging: A general approach. Human
Brain Mapping, 2, 189–210.
Garavan, H., Ross, T., Kaufman, J., & Stein, E. (2003). A midline dissociation between
error-processing and response-conflict monitoring. Neuroimage, 20, 1132–1139.
Garavan, H., Ross, T., & Stein, E. (1999). Right hemispheric dominance of inhibitory
control: An event-related functional MRI study. Proceedings of National Academy
of Sciences of United States of America, 96, 8301–8306.
Gauthier, L., Dehaut, F., & Joanette, Y. (1989). The bells test: A quantitative and qualitative test for visual neglect. International Journal of Clinical Neuropsychology, 11,
49–54.
Greene, C., Bellgrove, M., Gill, M., & Robertson, I. (2009). Noradrenergic genotype
predicts lapses in sustained attention. Neuropsychologia, 47, 591–594.
Hjaltason, H., Tegner, R., Tham, K., Levander, M., & Ericson, K. (1996). Sustained attention and awareness of disability in chronic neglect. Neuropsychologia, 34(12),
1229–1233.
Hon, N., Epstein, R., Owen, A., & Duncan, J. (2006). Frontoparietal activity with minimal decision and control. Journal of Neuroscience, 26, 9805–9809.
Husain, M., & Rorden, C. (2003). Non-spatially lateralized mechanisms in hemispatial
neglect. Nature Review Neuroscience, 4, 26–36.
Husain, M., Shapiro, K., Martin, J., & Kennard, C. (1997). Abnormal temporal dynamics
of visual attention in spatial neglect patients. Nature, 385, 154–156.
Johnson, K. A., Robertson, I. H., Kelly, S. P., Silk, T. J., Barry, E., Dáibhis, A. ,., et al. (2007).
Dissociation in performance of children with adhd and high-functioning autism
on a task of sustained attention. Neuropsychologia, 45, 2234–2245.
Kerns, J., Cohen, J., MacDonald, A., Cho, R., Stenger, V., & Carter, C. (2004). Anterior cingulate conflict monitoring and adjustments in control. Science, 303, 1023–1026.
Konishi, S., Nakajima, K., Uchida, I., Kikyo, H., Kameyama, M., & Miyashita, Y. (1999).
Common inhibitory mechanism in human inferior prefrontal cortex revealed by
event-related functional mri. Brain, 122, 981–991.
Konishi, S., Nakajima, K., Uchida, I., Sekihara, K., & Miyashita, Y. (1998). No-go dominant brain activity in human inferior prefrontal cortex revealed by functional
magnetic resonance imaging. European Journal of Neuroscience, 10, 1209–1213.
Konow, A., & Pribram, K. (1970). Error recognition and utilization produced by injury
to the frontal cortex in man. Neuropsychologia, 8, 489–491.
Logan, G., & Cowan, W. (1984). On the ability to inhibit thought and action: A theory
of an act of control. Psychological Review, 91, 295–327.
Lütcke, H., & Frahm, J. (2008). Lateralized anterior cingulate function during error
processing and conflict monitoring as revealed by high-resolution fMRI. Cerebral
Cortex, 18, 508–515.
Malhotra, P., Coulthard, E., & Husain, M. (2009). Role of right posterior parietal cortex
in maintaining attention to spatial locations over time. Brain, 132, 645–660.
Manly, T., Owen, A. M., McAvinue, L., Datta, A., Lewis, G. H., Scott, S. K. ,., et al.
(2003). Enhancing the sensitivity of a sustained attention task to frontal damage:
Convergent clinical and functional imaging evidence. Neurocase, 9(4), 340–349.
McDonald, A., Cohen, J., Stenger, V., & Carter, C. (2000). Dissociating the role of the
dorsolateral prefrontal and anterior cingulate cortex in cognitive control. Science,
288, 1835–1838.
Mennemeier, M., Chatterjee, A., Watson, R., Wertman, E., Carter, L., & Heilman, K.
(1994). Contributions of the parietal and frontal lobes to sustained attention
and habituation. Neuropsychologia, 32, 703–716.
Molenberghs, P., Gillebert, C., Peeters, R., & Vandenberghe, R. (2008). Convergence
between lesion-symptom mapping and fmri of spatially selective attention in
the intact brain. Journal of Neuroscience, 28, 3359–3373.
Mort, D., Malhotra, P., Mannan, S., Rorden, C., Pambakian, A., Kennard, C. ,., et al.
(2003). The anatomy of visual neglect. Brain, 126, 1986–1997.
Nakahara, K., Hayashi, T., Konishi, S., & Miyashita, Y. (2002). Functional MRI
of Macaque monkeys performing a cognitive set-shifting task. Science, 295,
1532–1536.
O’Connell, R., Bellgrove, M., Dockree, P., Lau, A., Fitzgerald, M., & Robertson, I. (2008).
Self-alert training: Volitional modulation of autonomic arousal improves sustained attention. Neuropsychologia, 46, 1379–1390.
Picton, T., Stuss, D., Alexander, M., Shallice, T., Binns, M., & Gillingham, S. (2007).
Effects of focal frontal lesions on response inhibition. Cerebral Cortex, 17,
826–838.
Rabbitt, P. (1966). Errors and error-correction in choice-response tasks. Journal of
Experimental Psychology, 71, 264–272.
Raven, J., Court, J., & Raven, J. (1995). Coloured progressive matrices. Oxford: Oxford
Psychologists Press.
Richer, F., & Lepage, M. (1996). Frontal lesions increase post-target interference in
rapid stimulus streams. Neuropsychologia, 34, 509–514.
Riddoch, M., & Humphreys, G. (1993). Birmingham object recognition battery. Hove,
UK: Lawrence Erlbaum Associates Ltd.
Robertson, I., Manly, T., Andrade, J., Baddeley, B., & Yiend, J. (1997). ‘oops!’: Performance correlates of everyday attentional failures in traumatic brain injured and
normal subjects. Neuropsychologia, 35(6), 747–758.
Robertson, I., Manly, T., Beschin, N., Daini, R., aeske Dewick, H., Homberg, V. ,., et
al. (1997). Auditory sustained attention is a marker of unilateral spatial neglect.
Neuropsychologia, 35(12), 1527–1532.
Robertson, I., Mattingley, J., Rorden, C., & Driver, J. (1998). Phasic alerting of
neglect patients overcomes their spatial deficit in visual awareness. Nature, 395,
169–172.
Rorden, C., Karnath, H., & Bonilha, L. (2007). Improving lesion-symptom mapping.
Journal of Cognitive Neuroscience, 19, 1081–1088.
Rubia, K., Russell, T., Overmeyer, S., Brammer, M., Bullmore, E., Sharma, T. ,., et al.
(2001). Mapping motor inhibition: Conjunctive brain activations across different
versions of go/no-go and stop tasks. NeuroImage, 13, 250–261.
Samuelsson, H., Hjelmquist, E., Jensen, C., Ekholm, S., & Blomstrand, C. (1998).
Nonlateralized attentional deficits: An important component behind persisting visuospatial neglect? Journal of Clinical and Experimental Neuropsychology,
20, 73–88.
Schall, J., Stuphorn, V., & Brown, J. (2002). Monitoring and control of action by the
frontal lobes. Neuron, 36, 309–322.
Scheffers, M., Coles, M., Bernstein, P., Gehring, W., & Donchin, E. (1996). Event-related
brain potentials and error-related processing: An analysis of incorrect responses
to go and no-go stimuli. Psychophysiology, 33, 42–53.
Schenkenberg, T., Bradford, D., & Ajax, E. (1980). Line bisection and unilateral visual
neglect in patients with neurologic impairment. Neurology, 30, 509–517.
Shallice, T. (1988). From neuropsychology to mental structure. Cambridge: Cambridge
University Press.
Shallice, T., Marzocchi, G., Coser, S., DelSavio, M., Meuter, R., & Rumiati, R. (2002).
Executive function profile of children with attention deficit hyperactivity disorder. Developmental Neuropsychology, 21, 43–71.
Sohn, M., Ursu, S., Anderson, J., Stenger, V., & Carter, C. (2000). The role of prefrontal
cortex and posterior parietal cortex in task switching. Proceedings of National
Academy of Sciences of United States of America, 97, 13448–13453.
Stuss, D., Murphy, K., Binns, M., & Alexander, M. (2003). Staying on the job: The
frontal lobes control individual performance variability. Brain, 126, 2363–2380.
Stuss, D., Shallice, T., Alexander, M., & Picton, T. (1995). A multidisciplinary approach
to anterior attentional functions. Annals of New York Academy of Sciences, 769,
191–212.
Swick, D., & Turken, A. (2002). Dissociation between conflict detection and error
monitoring in the human anterior cingulate cortex. Proceedings of National
Academy of Sciences of United States of America, 99, 16354–16359.
Talairach, J., & Tournoux, P. (1988). Co-planar stereotaxic atlas of the human brain. New
York: Thieme Medical Publishers, Inc.
Tyler, L., Marslen-Wilson, W., & Stamatakis, E. (2005). Dissociating neuro-cognitive
component processes: Voxel-based correlational methodology. Neuropsychologia, 43, 771–778.
P. Molenberghs et al. / Neuropsychologia 47 (2009) 2866–2875
van Essen, D. (2005). A population-average, landmark- and surface-based (PALS)
atlas of human cerebral cortex. Neuroimage, 28, 635–662.
van Essen, D., Drury, H., Dickson, J., Harwell, J., Hanlon, D., & Anderson, C. (2001). An
integrated software suite for surface-based analyses of cerebral cortex. Journal
of American Medical Informatics Association, 8, 443–459.
Vandenberghe, R., Gitelman, D., Parrish, T., & Mesulam, M. (2001). Location- or
feature-based targeting of peripheral attention. NeuroImage, 14, 34–47.
Warrington, E., & James, M. (1991). Visual object and space perception battery. Thamys
Valley Test Company Ltd.
2875
Wechsler, D. (1998). Wechsler Memory Scale-III. San Antonio, TX: The Psychological
Corporation.
Whyte, J., Grieb-Neff, P., Gantz, C., & Polanksy, M. (2006). Measuring sustained attention after traumatic brain injury: Differences in key findings from the sustained
attention to response task. Neuropsychologia, 44, 2007–2014.
Wilkins, A., Shallice, T., & McCarthy, R. (1987). Frontal lesions and sustained attention.
Neuropsychologia, 25, 359–365.
Zordan, L., Sarlo, M., & Stablum, F. (2008). ERP components activate by the ‘GO’ and
‘withhold’ conflict in the random SART. Brain Cognition, 66, 57–64.