Attentional and

neuromotor deficits in
ADHD
J Steger PhD, Laboratory for Biomechanics, Eidgenssiche
Technische Hochschule (ETH) Zurich;
K Imhof PhD;
E Coutts lic phil;
R Gundelfinger MD;
H-Ch Steinhausen MD PhD;
D Brandeis*PhD, Department of Child and Adolescent
Psychiatry, University of Zurich, Switzerland.
*Correspondence to last author at Department of Child and
Adolescent Psychiatry, University of Zurich,
Neumnsterallee 9/ Postfach, CH-8032 Zrich, Switzerland.
E-mail: brandeis@kjpd.unizh.ch

In order to classify attention-deficithyperactivity disorder

(ADHD) in 11-year-old children, the role of specific
attentional and motor deficits was examined. Participants
comprised 22 children with ADHD (19 male, 3 female;
median age 11 years, range 8.8 to 13.5 years) and 20 control
children (17 male, 3 female; median age 10.6 years, range 8.2
to 12.6 years). Neuromotor assessment indicated that while
both groups needed more time to complete finger compared to
hand movements, this increase was more pronounced in
children with ADHD. Reaction-time testing with continuousforce recording identified both motor and attentional deficits
in children with ADHD. Longer intervals between force onset
and force peak, and higher rate of responses with multiple
force peaks (particularly in the bilateral condition) revealed
specific deficits in the speed and quality of their motor output.
Increase in errors and variability of force onsets indicated
attentional deficits. Prediction analysis indicated that forceonset variability contributed significantly to group
classification which was 85.7% correct. Neither neuromotor
assessment nor specific motor deficits contributed
significantly to classification, indicating that pure motorspeed measures play a minor role in characterizing ADHD in
this age range.

172

Developmental Medicine & Child Neurology 2001, 43: 172179

Attention-deficithyperactivity disorder (ADHD) is characterized by developmentally inappropriate attention, overactivity, and impulsivity, but several lines of evidence suggest
that children with ADHD also have motor deficits. Some
research suggests that the overlap between ADHD and motor
deficits is close to 50% and therefore high enough to warrant a
separate diagnostic category (deficit in attention, motor control, and perception DAMP; Gillberg 1995) while other
research suggests that motor processing is typically unaffected
in ADHD (Leung and Connolly 1998). Such large discrepancies are likely to result from addressing different populations,
and from differences in the assessment and conceptualization
of motor deficits, ADHD, and their combination.
Motor deficits in ADHD have been assessed in a wide
range of tests. In the neuromotor assessment battery of
Denckla and Rudel (1978), the speed of pure motor functions is strongly represented. The time needed for repetitive
simple movements or movement sequences in the different
tasks, is measured for finger, hand, and foot. Denckla and
Rudel (1978), Carte and colleagues (1996), and Schuerholz
and coworkers (1997) all used these tasks to examine children with ADHD with a mean age below 10 years. Denckla
and Rudel (1978) reported that slowing of gross movements
(i.e. heeltoe taps) differentiated the ADHD and control
groups most accurately while Carte and colleagues (1996)
found that hand and foot movements were significantly
slower in children with ADHD. Both Denckla and Rudel
(1978) and Carte and colleagues (1996) showed that neuromotor parameters contributed significantly to group discrimination between children with ADHD and control
children. Schuerholz and colleagues (1997) reported that
even though children with ADHD showed no overall slowing of finger movements, they had a larger time discrepancy
between simple and patterned movements, and performed
fewer tasks within the age-appropriate speed limits than
children with Tourette syndrome. More subtle motor
deficits which affect movement quality rather than speed are
also found in ADHD. These include motor overflow, i.e.
involuntary associated movements such as mirror movements
(Denckla and Rudel 1978). Motor overflow also appears to correlate with inattentiveness in normally developing school-age
children (Waber et al. 1985, Lazarus and Todor 1991).
More complex tests of fine motor skills based on handwriting, manual dexterity, ball skills, and balance have also
indicated deficits in children with ADHD. However these
tests tend to measure both pure motor as well as visuomotor
or higher cognitive functions, similar to neuropsychological
frontal-lobe tests such as copying a complex design or maze
tracing. Compared to control children, children with ADHD
were found to have impaired handwriting skills (McMahon
and Greenberg 1977, Barkley 1990), poorer fine visuomotor
ability (Whitmont and Clark 1996), and poorer performance
in visuomotor frontal-lobe tests (review in Barkley et al.
1992). In addition, different types of motor deficits appear
to characterize different ADHD subgroups (Piek et al. 1999).
Children of the inattentive subtype of ADHD had significantly
poorer manual dexterity skills, whereas children of the combined subtype experienced greater difficulty with balance.
Direct experimental manipulation of premotor processing
via low or irregular event rate (Chee et al. 1989, van der Meere
et al. 1992, Fldnyi et al. 2000) or via stimulusresponse
incompatibility (van der Meere et al. 1989) have also revealed

measures to the prediction of an ADHD diagnosis.

some specific premotor deficits in ADHD. However, these

premotor deficits are derived from manipulations affecting
higher-order regulation, and are less consistent than the
attentional and inhibitory deficits (Oosterlaan and Sergeant
1998) revealed by increases of response-time variability and
errors of omission and commission, i.e. misses and false alarms
(Fldnyi et al. 2000). The fact that neither choice reaction
time nor movement time are increased in children with ADHD
(Leung and Connolly 1998) suggests that there is little evidence for specific motor-speed deficits in simple motor tests.
In summary, the majority of studies with neuromotor
standard assessments indicate that the quality and speed of
simple movement sequences discriminate between those
with ADHD and control children. However this conclusion
receives little support from a study using simple motor laboratory tests that measure motor sequencing speed on a trialby-trial basis (Leung and Connolly 1998). Continuous-force
recordings rather than discrete button presses would be
required to assess the quality of motor output in such laboratory tests. Other types of motor coordination have not been
examined on a trial-by-trial basis in ADHD children.
Bilateral responding (i.e. synchronization of unilateral
responses) reflects a basic form of motor coordination that
results in slower reaction times (around 5%) than unilateral
response (Jeeves and Dixon 1970, Ohtsuki 1981, Anson and
Bird 1993). This bilateral reaction-time deficit is present even if
separate blocks of uni- and bilateral trials permit optimal attentional strategies and response preparation (Kawabe 1989).
This may be due to additional motor coordination rather than
attentional or cognitive coordination. In addition, the reduction of this bilateral timing deficit in patients with split-brains
(Gazzaniga and Sperry 1966, Wyke 1971) indicates that it is
mediated by the corpus callosum. As some studies indicate
that certain callosal regions are reduced in children with
ADHD (Semrud-Clikeman et al. 1994, Castellanos et al. 1996,
but see Overmeyer et al. 2000), impaired bilateral coordination may be expected in this group.
The aims of the present study were: (1) to determine specific motor processing deficits in children with ADHD using
neuromotor assessment and continuous-force recordings
in a reaction-time test involving bilateral coordination, and
(2) to clarify the contribution of attentional and motor

Method
PARTICIPANTS

Forty-two right-handed children (IQ>80; age range 8.2 to

13.5 years) were selected for this study. The 22 children with
ADHD (mean age 10.9 years, range 8.8 to 13.5 years) and 20
sex- and age-matched control children (mean age 10.6 years
range 8.2 to 12.6 years) were grouped on the basis of DSMIII-R (American Psychiatric Association 1987) diagnoses
derived from a structured interview (Schaffer et al. 1993) carried out by trained undergraduate psychology students. The
children with ADHD were recruited from our patients (n=11),
from parents organizations (n=5), from an epidemiological
field study (n=4; Steinhausen et al. 1998), and from nearby
schools through teachers (n=1), or school psychologists
(n=1). They all met the ADHD diagnosis and had no comorbid disorders except for oppositional defiant disorder (11
children) and/or conduct disorder (two children), and/or
one isolated phobia (five children). Control children were
recruited from regular schools through teachers or laboratory
staff members or from the same field study. No DSM-III-R diagnoses were allowed except for a single isolated phobia (two
children). Control childrens T-scores (mean 50, SD 10) on the
Attention Problem Scale of the Child Behavior Check List
(CBCL; Achenbach 1991a) had to be lower than 55, i.e. less
than half a standard deviation above the normative value. The
sample description (for both the full group and subgroup
selected for additional analyses) is given in Table I.
At initial assessment the childrens parents and teachers
reported on behavior problems at home (CBCL) and at school
(Teachers Report Form, TRF; Achenbach 1991b). In the first
laboratory session all children underwent neuromotor assessment and several neuromotor tests. An abbreviated version of
the revised Wechsler Intelligence Scale for Children (WISC)
with German normative values (Brndler and Schallberger
1988) was used to determine IQ. In a second laboratory session a reaction-time test involving bilateral coordination was
performed in addition to other neuropsychological tests. All
children were free of medication at the time of testing; those
children with ADHD taking methylphenidate (n=7) had been
off medication for at least 24 hours.

Table I: Description of sample

Age
Sex (F/M)
IQ
CBCL attention problems
(T-scores)
Comorbidity (nr children)
Multiple comorbidities
Single comorbidity
ODD
CD
Phobias

ADHD
n=22

Full group
Control
n=20

10.93
3/19
98.67
66.13

10.55
3/17
105.56
43.10

4
10
11
2
5

ADHD
n=14

Subgroup
Control
n=14

>0.3
<0.05
<0.0001

10.54
0/14
101.43
64.31

10.78
0/14
104.10
43.11

>0.3
<0.0001

2
8
7
1
4

>0.5

CBCL, Child Behavior Checklist; ODD, Oppositional defiant disorder; CD, Conduct disorder.

Attentional and Neuromotor Deficits in ADHD J Steger et al. 173

NEUROMOTOR ASSESSMENT

Each child was examined by trained child psychiatrists using

a neuromotor assessment battery (Largo and Caflisch 1999)
based on Dencklas (1974) six tasks of repeated movements.
The children were instructed to execute each of the following tasks rapidly but precisely while the time needed to complete the movements was measured: (1) Finger repetition
(simple movement), the child touches index finger to
thumb (20 repetitions). (2) Finger sequention (patterned
movement), the child touches each finger successively to
thumb, moving from the index finger to the little finger each
time (5 repetitions). (3) Hand pat (simple movement), the
child pats the thigh with the palm of the hand, with the heel
of the palm resting on the thigh (20 repetitions). (4) Hand
pronationsupination (patterned movement), the child
pats the thigh with the back and palm of the hand alternately
(10 repetitions). (5) Toe taps (simple movement), the child
taps toe with the heel resting on the floor (20 repetitions).
(6) Heeltoe (patterned movement), the child alternates
touching heel and toe to the floor in a rocking motion (10
repetitions).
REACTION -TIME TEST

Stimuli and test

Figure 1 illustrates how pictures of an aeroplane or a man
(both subtending 0.95 horizontally and vertically) were presented on a computer monitor every 1.65 seconds for 180
seconds in black on a light grey background. Vertical lines

above and below the center of the picture served as fixation

marks. These pictures appeared with equal probability on
the left, right, or both sides simultaneously (1.33 from fixation). On target trials (p=0.67), the aeroplane which required
a response was presented. On non-target trials (p=0.33), the
man was presented at the same screen positions as the aeroplane (left, right, bilateral), and the child had to refrain from
responding. The six test conditions (target or non-target, each
in left unilateral, right unilateral, and bilateral conditions)
were presented in randomized order. Children were seated
1.2 metres from the monitor. They responded as quickly as
possible with the hand corresponding to the target position
(left, right, or simultaneously with both hands), using opposing pressure of thumb and index finger (precision grip). After
a short practise run, three experimental blocks, each containing 60 targets and 30 non-targets, followed.
Force recording and processing
Response force between thumb and index finger was measured by force transducers (strain gauges) and continuously
recorded for each hand at 256 Hz. The force pulses produced
by the responses were analyzed as shown in Figure 1.
The onsets of the force pulses (force-onset latencies)
were automatically marked as the time points at which the
force signals exceeded the noise level. In addition, the
time between force onset and force peak, and the peak force
were computed. To be coded as a correct trial, the force-onset
latency had to be within a time window of 200 and 800 ms and

10 (N)

Right hand
Left hand
0 ms

800 ms

200 ms

Force (N)
Right hand
Left hand

Time
0s

1.65 s

3.3 s

4.95 s

Figure 1: Selected stimuli (not drawn to scale) and force-pulse analysis in reaction-time test. Pulses in force curves for left
(bottom) and right (top) hand reflect correct responses for this arbitrary stimulus sequence. Inset: black line indicates,
correct; dashed small lines indicate incorrect; and dashed multipeak lines indicate imperfect right-hand responses to right
unilateral target. (a), parameters force-onset latency, (b) time from force onset to force peak, (c) peak force.

174

Developmental Medicine & Child Neurology 2001, 43: 172179

peak force had to be higher than 0.5 N (grey box in Fig. 1).
In addition to the traditional error types (omission and
commission errors, side errors, unilateral responses in the
bilateral condition, and bilateral responses in unilateral conditions), several imperfect but correct responses were coded
and analysed. These were: (1) trials with force-onset latencies shorter than 200 ms or longer than 800 ms, (2) trials with
mirror movements (correct but additional response pulses
below 0.5 N with the incorrect hand, see Fig. 1), (3) trials
with multiple force peaks (see Fig. 1), and (4) trials with
poor synchronization of bilateral responses (defined as
force-onset latency differences of more than 60 ms ). These
imperfect responses were not included in the total error
count in order to maintain compatibility with previous error
analyses. All errors were transformed into percentages and
time measures were log transformed (base 10).
DATA ANALYSIS

Group and condition effects were first tested separately for

the parameters of the neuromotor assessment, and the reaction-time test. Multivariate analysis of variance for repeated
measures covaried for IQ (MANCOVA/GLM) were used for
this purpose. Equivalent F values for Wilks Lambda are
reported. These multivariate analyses were followed by univariate analyses of covariance (ANCOVA) for specific tasks or
test conditions. Due to the fact that sex and IQ were suspected to correlate with an ADHD diagnosis, all analyses were
also performed on subgroups matched for sex and IQ
(instead of covarying for IQ). These subgroup results are
only reported if there were discrepancies with the full
group analyses. As no significant differences were obtained
between left- and right-sided responses, their mean was
taken for all analyses. For the neuromotor assessment, the
time (in seconds) needed to complete the movements in the
six tasks was analyzed with the factors group (ADHD versus
control), extremity (finger versus hand versus leg), movement mode (simple versus patterned), and with age as an
additional covariate. For the reaction-time test, the time and
error measures listed above were analyzed with the factors
group, bilaterality (mean unilateral versus bilateral), and target/non-target (targets versus non-targets, only for error
parameters). These analyses were performed both with and
without age as an additional covariate.
For the group prediction (ADHD or control group, full
groups only), the best group discriminator from the neuromotor analysis, and the best group discriminator from the
reaction-time analyses were selected. These two measures
and the IQ were entered together in a logistic regression.
Due to several measures in the reaction-time test indicating
high significance group differences (p<0.001) logistic regressions were computed for each of those measures alone to
determine the one with the best predictive value. Specific
settings of all these maximum-likelihood predictions included: forward and backward stepwise method, entry p=0.05,
removal p=0.1; classification cut off p=0.05; and 20 maximum iterations. In case of discrepancies between forward
and backward solutions, the window between entry and
removal criterion was reduced (entry p=0.05, removal
p=0.06). Subanalyses with single predictors were run to clarify the effects. Significance values of the Wald statistic are based
on the 2 distribution.

Results
Table I shows that the mean IQ was lower for children with
ADHD than for control children whereas the analysis of the
subgroups (n=14 for each group, males only) showed no
significant IQ differences. According to the CBCL, children
with ADHD had significantly higher attention-problem scores
than control children. No significant age differences between
the ADHD and control children were found.
NEUROMOTOR ASSESSMENT

The multivariate analysis of the neuromotor measures indicated that whether children with ADHD needed more time to
complete movements than control children depended on the
extremity (group extremity, F[2,37]=3.77, p<0.05). Posthoc analysis of this interaction indicated that the additional
time needed to complete finger compared to hand movements was increased in the ADHD group (group extremity,
F[1,38]=7.27, p<0.01), but none of the three extremities
revealed slowing when tested individually except for a trend
for slower finger movements (F[1,38]=3.22, p=0.08).
Univariate analysis also showed no significant group differences (see Table II) and only a trend for slower sequential finger movements in the ADHD group (F[1,38]=2.89, p=0.1).
A main effect of extremity (F[2,37]=4.96, p<0.05) indicated that finger and foot movements were slower than
hand movements. Patterned movements were slower than
simple movements (8.41 versus 5.84 seconds, F[1,38]=10.44,
p<0.01). In addition, the covariate age significantly affected
the time to complete movements, F[1,38]=8.00, p<0.01).
The negative correlations between age and time to complete
finger (r=0.260, p=0.048), hand (r=0.436, p=0.002), and
foot (r=0.318, p=0.02) movements indicated that younger
children were slower with all extremities. No significant
resutls were obtained for the covariate IQ.
REACTION -TIME TEST

The multivariate analysis of the values shown in Table III

indicated that in children with ADHD, force onsets were
more variable (force onset, SD 110.0 versus 84.6 ms ;
F[1,39]=12.51, p<0.001) but not significantly later than in
control children. In addition, the time from force onset to
force peak was prolonged in ADHD compared to control
children (225.2 versus 192.9 ms F[1,39]=4.42, p<0.05).
Bilateral responding was significantly slower than unilateral
responding (bilateral reaction-time deficit), but only in the
subgroup analysis (443.3 versus 422.4 ms , F[1,26]=12.74, p
<0.001). No significant group differences were found for the

Table II: Neuromotor task measures (unilateral), mean (SD)

Finger
Repetition
Sequention
Hand
Pat
Pronationsupination
Foot
Toe taps
Heeltoe

ADHD (n=22)

Controls (n=20)

5.85 (1.28)
10.37 (2.47)

5.68 (1.10)
8.87 (2.06)

4.84 (1.13)
7.00 (1.30)

5.07 (1.09)
6.88 (1.29)

7.20 (2.19)
9.23 (3.17)

6.39 (1.25)
8.10 (2.25)

Attentional and Neuromotor Deficits in ADHD J Steger et al. 175

bilateral reaction-time deficit. Children with ADHD made

more errors than control children (13.8% versus 4.7%;
F[1,39]=15.39, p<0.001); this was most pronounced for
bilateral non-targets (F[1,39]=4.89, p<0.05). They also
made more commission errors than control children
(F[1,39]=9.76, p<0.005) and responded more often with
multiple force peaks (F[1,39]=18.36, p<0.001), particularly
in the bilateral condition (F[1,39]=5.69, p<0.05). No significant group differences were found for omission errors or
responses outside the latency window (200 to 800 ms).
Compared to control children, those with ADHD also made
fewer synchronized bilateral responses in the bilateral
condition (F[1,39]=20.21,p<0.001) and made more bilateral
responses in the unilateral conditions (F[1,39]=15.36,
p<0.001). No significant group differences were found for side
errors, mirror movements, unilateral responses in the bilateral
condition, or peak force.
Significant effects of the covariate IQ indicated that children
with lower IQ made fewer synchronized bilateral responses
(F[1,39]=4.13, p<0.05) but they also made fewer erroneous
bilateral responses in the unilateral conditions (F[1,39]=7.52,
p<0.01). Introducing age as an additional covariate revealed
no significant age effects on reaction-time test parameters
except for two trends (force onset latency and variability
p<0.1). Group and condition effects remained unchanged.
GROUP PREDICTION

From the neuromotor assessment measures, only the time

difference between finger and hand movements (group
extremity, neuromotor interaction) discriminated between
the groups. In the reaction-time test, individual logistic regression analysis of several highly discriminating measures (Table
IV) indicated that response variability was the best predictor.
IQ was included because of the significant group differences.
Predictions based on these measures from neuromotor

assessment and the reaction-time test are shown in Table V.

The final equation contained overall response variability and
IQ, but not the neuromotor interaction measure, as significant predictors. This equation resulted in the correct classification of 85.71% of all participants. Twenty-one of 22
children with ADHD (95.45%) and 15 of 20 control children
(75%) were correctly classified. The initial backward solution
retained both the IQ and the neuromotor interaction at a
non-significant level, but more similar entry- and removalcriteria produced consistent forward and backward solutions
identical to the initial forward approach. When entering each
predictor separately, response variability resulted in 80.95%,
IQ 64.29%, and neuromotor interaction measure 69.5% correct classification.
Discussion
NEUROMOTOR ASSESSMENT

Neuromotor assessment indicated that 11-year-old children

with ADHD had no clear neuromotor-speed deficits except
that they were slower than control children when performing finger movements compared to hand movements. This
interaction suggests that children with ADHD have specific
problems with motor tasks that require fine motor skills.
Finding only a trend towards slowing across all finger movements, is consistent with the results of a recent report using
the same tasks and analyses controlling for age and IQ
(Schuerholz et al. 1997). Their finding of more pronounced
slowing with patterned compared to repetitive finger movements was not obtained in our data. However, the lack of
deficits for hand or foot movements alone contrasts with previous studies reporting differences for both hand and foot
movements (Denckla and Rudel 1978, Carte et al. 1996).
Considering that several standard findings (significant slowing for younger children, patterned movements, and finger/foot compared to hand movements) support the validity

Table III: Reaction time test measures, mean (SD)

ADHD (n=22)
Unilateral
Bilateral
Force onset (ms)
Variability of force onsets (ms)
Time from onset to peak (ms)
Total errors (%)
Errors (%)
Target
Non-target
Omissions
Commissions
Multiple peaks
Window
Side
Biuni
Mirror movements
Unibi
Synchronicity

Control (n=20)
Unilateral
Bilateral

421.7 (70.8)c
110.3 (26.7)c
228.8 (52.3)a
15.5 (9.0)a

445.2 (85.5)a
109.3 (27.1)a
221.6 (49.0)a
12.0 (10.0)a

415.2 (72.6)
84.2 (13.0)
197.7 (39.2)
6.8 ( 3.0)

439.9 (68.2)
85.5 (15.4)
188.1 (42.5)
2.7 (2.3)

13.0 (11.7)a
18.0 (11.3)c
4.5 (10.1)c
12.1 (7.7)b
6.9 (6.1)a
3.2 (3.5)a
1.0 (2.0)a
7.4 (5.3)d
3.8 (3.6)d

6.6 (15.3)d
17.4 (13.1)d
3.8 (10.4)d
5.5 (6.9)c
19.0 (9.2)c
2.5 (3.3)c

2.8 (5.2)dd
10.7 (5.0)d

5.3 (3.3)
8.2 (4.7)
0.8 (1.6)
6.5 (3.7)
2.2 (2.1)
2.0 (3.9)
0.6 (0.9)
3.9 (3.4)
2.9 (2.1)

0.8 (1.1)
4.7 (4.2)
0.3 (0.8)
1.7 (3.3)
8.8 (5.9)
0.6 (1.2)

0.5 (1.0)
5.5 (3.5)

Individual ANCOVA results with IQ covaried: a p<0.05; b p<0.01; c p<0.005; d p<0.001.

Multiple peaks, responses with multiple force peaks; Side, left hand response in right-side condition (vice
versa); Window, responses outside the latency window; Biuni, bilateral responses in unilateral conditions;
Unibi, unilateral responses in bilateral condition; Synchronicity, synchronicity of bilateral responses.
Multiple peaks, window, mirror movements, and synchronity are imperfect but correct responses.

176

Developmental Medicine & Child Neurology 2001, 43: 172179

of our neuromotor assessment, it appears likely that group

characteristics are responsible for these discrepancies. One
explanation for these discrepancies might be the older mean
age of the children in our groups. Gillberg and colleagues
(1983) and Gillberg and Gillberg (1989) showed that neuromotor deficits observed in children with ADHD at the age of 6
or 7 years often disappeared at age 10 and 13 years even
though behavioural problems persisted. Whereas the studies
of Denckla and Rudel (1978) and Carte and coworkers (1996)
included children younger than 8 years, the children participating in the present study were all older than 8 years, i.e. at
an age where neuromotor deficits in ADHD are less prominent. In addition, our ADHD group may have contained children with less severe ADHD than the sample of Carte (1996),
which consisted only of children with severe ADHD who had
been on stimulant treatment. Such a correlation between
motor deficits and severity was confirmed by Whitmont and

Clark (1996). Whereas additional analyses confined to children younger than 10 or to children with severe ADHD failed
to reveal neuromotor deficits for finger or foot movements
alone, the small samples (n<10) available for these analyses
do not allow firm conclusions.
REACTION-TIME TEST

The force-onset latencies from the reaction-time test also

indicated no general slowing in ADHD compared to control
children, which is consistent with other studies using simple
motor tests (Barkley et al. 1992, Leung and Connolly 1998)
and high event rates (van der Meere et al. 1992). Additional
motor coordination demands during bilateral responding
resulted in a bilateral reaction-time deficit, i.e. in slower
force onsets than with unilateral responding in the IQmatched subgroups. This effect was similar in size to that
observed in adults (Jeeves and Dixon 1970, Anson and Bird

Table IV: Evaluation of best predictor of reaction-time test for diagnostic status
Method
Enter
Enter
Enter
IQ
Enter
Enter

Variables

Parameter estimates

Standard error

Adjusted odds ratio

Overall classification %

Variability
IQ
Total errors
IQ
Synchronicity
0.1822
Multiple peaks
IQ
Biuni
IQ

0.1046
0.1019
55.0279
0.1526
45.1713
5.9702
14.6179
0.0634
37.0397
0.1755

8.0232
3.8771
7.7888
3.9819
8.3961
0.8334
9.3773
1.7476
9.0808
7.4096

1.1103
0.9032
7.912E+23
0.8585
4.146E+19
0.015
2.230E+06
0.9386
1.219E+16
0.8390

0.005
0.049
0.005
0.046
0.004

85.71
83.33
80.95

0.002
0.186
0.003
0.007

78.57
73.81

Multiple peaks, responses with multiple peaks; Biuni, bilateral responses in unilateral conditions; Synchronicity, synchronicity of bilateral
responses.

Table V: Predicting diagnostic status through logistic regression

Method

Forward

Backward

Entry/removal
probability

Variables

Parameter
estimates

Standard error
ratio

Adjusted odds

Overall
classification %

0.05/0.1
Movement

Variability

time
IQ
Variability
Movement
time
IQ
Variability
Movement
time
IQ
Variability
Movement
time
IQ
Variability
Movement
time
IQ

0.1046

8.0232

1.1103
0.0919

0.005

85.71

0.1019
0.0917
0.4620

3.8771
6.1202
2.5943

0.9032
1.0960
1.5872

0.05
0.0134
0.1072

88.10

0.0951
0.1046

3.0749
8.0232

0.9093
1.1103

0.0795
0.005
0.0919

85.71

0.1019
0.1046

3.8771
8.0232

0.9032
1.1103

0.05
0.005
0.0919

85.71

0.1019
0.0972
0.6229

3.8771
8.7194
7.0179

0.9032
1.1021
1.8644

0.05
0.0031
0.0081

80.95
69.05

0.0957

5.2194

0.9087

0.0223

64.29

0.05/0.1

Forward

0.05/0.06

Backward

0.05/0.06

Enter
Enter

0.05/0.1
0.05/0.1

Enter

0.05/0.1

Variability, overall variability of reaction time test; Movement time, increased time to complete finger movements compared to hand
movements (group extremity, neuromotor variables).

Attentional and Neuromotor Deficits in ADHD J Steger et al. 177

1993), which suggests that the underlying processes were

relatively mature in our 11-year-old children. Whether the
bilateral deficit in adults is also eliminated by covarying IQ
remains to be examined. The groups did not differ in the size
of their bilateral deficits. This contrasts with the reduced
bilateral deficit of patients with split-brains and suggests that
those callosal functions supporting bilateral motor coordination are unaffected in children with ADHD.
Measuring continuous-force output in this test provided
direct measures of motor function that are unavailable with
typical reaction-time testing. Due to the fact that peak force
was comparable between ADHD and control children, the
prolonged time from force onset to force peak in children
with ADHD must reflect a pure motor-speed deficit. Another
type of specific motor deficit was revealed by the increased
rate of multiple response peaks in children with ADHD
which was most pronounced under bilateral coordination
demands. Deterioration of the quality of motor output,
therefore, increased with demands. This may contribute to
the high rates of motor deficits in children with ADHD which
are obtained if qualitative aspects of motor performance in
more complex skills are also scored (Gillberg 1995, Kadesjo
and Gillberg 1998).
Mirror movements are another aspect of motor function
that can only be measured with continuous-force recordings
in reaction-time studies. The mirror movements were not
significantly increased in the group with ADHD, which contrasts with a neuromotor assessment study in younger children (Denckla and Rudel 1978) and parallels the lack of
significant neuromotor timing differences in this age range.
Future studies with younger children, who are known to
exhibit more mirror movements, could clarify the relationship between mirror movements derived from quantification of continuous-force output and from visual ratings
during neuromotor assessments.
Children with ADHD made more errors than control children and showed increased variability of force-onset latencies
during correct responding. These findings are consistent
with previous studies (Mitchell et al. 1990, Schachar 1991,
Oosterlaan and Sergeant 1998, Rubia et al. 1998, Fldnyi et
al. 2000) and are thought to reflect attentional deficits in children with ADHD. In particular, high response variability has
been associated with non-optimal regulation of attentional or
energetic states (Oosterlaan and Sergeant 1998). Similarly, the
fewer synchronized bilateral responses of children with
ADHD may be due to the increased variability of the participating unilateral responses. As neither the latency nor the variability of bilateral responses were specifically increased in
the group with ADHD, contributions from a specific transcallosal coordination deficit appear less likely.
Children with ADHD made slightly more unilateral instead
of bilateral responses than control children, but they made
many more erroneous bilateral responses on unilateral trials,
and particularly on unilateral non-targets (commission errors).
This suggests a bias towards bilateral responding throughout
the test and could reflect attentional responsecontrol deficits
in children with ADHD. Commission errors were more frequent in ADHD than in control children, which is consistent
with findings from other tests using similarly high response
rates (50%) either throughout or after cues (Michael et al.
1981, van Leeuwen et al. 1998). The lack of significant group
differences in omission errors is also consistent with previous

178

Developmental Medicine & Child Neurology 2001, 43: 172179

studies; omission errors are typically prominent only in vigilance tests with low-response probability.
Taken together, continuous-force measurements during
the reaction-time test revealed both specific motor and attentional deficits in children with ADHD but no clear age effects
and no general motor slowing. This finding is consistent with
recent event-related potential findings during the same test,
which indicate that both premotor and attentional brain
activity is reduced in boys with ADHD despite normal latencies (Steger et al. 2000). The neuromotor assessment also
indicated no general motor slowing in children with ADHD,
but instead resulted in clear age effects. Both modes of testing thus indicated that children with ADHD have no general
motor slowing but specific deficits which differ from those
characterizing younger children.
Correlating neuromotor with reaction-time parameters
provided further insight into the determinants of movement
speed during the neuromotor assessment. Finger movement
time correlated with force-onset latency during correct
responding (r=0.387, p=0.011) but not with the latency
between force onset and peak or with error rate in the reaction-time test. This suggests that the time needed to complete finger movements primarily varies with force-onset
latencies and is relatively unaffected by movement time and
errors of movement or sequencing in this child population.
GROUP PREDICTION

Prediction analysis indicated that force-onset variability and IQ

contributed to 85.7% correct group classification. Force-onset
variability provided the main contribution to group discrimination and IQ a minor contribution. Neuromotor interaction
(finger versus hand) not only led to poorer classification than
force-onset variability when tested separately, but was
removed from the final equation in the iterative procedures.
Similarly, the motor parameter in the reaction-time test led to
a lower classification than latency variability and errors did.
This contrasts with the results from neuromotor studies of
younger children and with significant group differences for
individual neuromotor measures. In the study by Carte and
colleagues (1996), neuromotor variables provided one of the
highest independent contributions to the 82% correct group
classification (using discriminant analysis of eight age-adjusted measures), similar to the 80.6% correct classification
obtained by Denckla and Rudel (1978).
In conclusion, even in simple motor reaction-time tests,
attentional measures seem to discriminate better than motor
measures between 11-year-old children with ADHD and control children. Strong support for the conclusion that motor
deficits are less pronounced in this age range comes from the
fact that subtle deficits, which are unlikely to allow visual scoring, could be measured reliably through continuous-force
recording. Further studies have to prove the clinical feasibility
of this new method of assessing neuromotor function.
Accepted for publication 8th June 2000.
Acknowledgments
We wish to thank the child psychiatrists of the clinic assessing the
neuromotor state of all children and H Haas from the Laboratory for
Biomechanics (ETH Zurich) for the construction and technical support
of the force measurement device. This work was supported by the
Swiss National Science Foundation (31-43790.95 and 32-43144.95).

References
Achenbach TM. (1991a) Manual for the Child Behavior CheckList/4-18 and 1991 Profile. Burlington, VT: Department of
Psychiatry, University of Vermont.
(1991b) Manual for the Youth Self-Report and 1991Profile.
Burlington, VT: Department of Psychiatry, University of Vermont.
American Psychiatric Association. (1987) Diagnostic and Statistical
Manual of Mental Disorders. 3rd revised edn. Washington DC:
American Psychiatric Association.
Anson JG, Bird YN. (1993) Neuromotor programming: bilateral and
unilateral effects on simple reaction time. Human Movement
Science 12: 3750.
Barkley RA. (1990) Attention Deficit Hyperactivity Disorder: a
handbook for diagnosis and treatment. New York: Guildford
Press.
Grodzinsky G, DuPaul GJ. (1992) Frontal lobe functions in
attention-deficit disorder with and without hyperactivity: a review
and research report. Journal of Abnormal Child Psychology 20:
16388.
Brndler M, Schallberger U. (1988) HAWIK-R Ergnzungsband mit
Normentabellen und Testanweisungen fr die deutschsprachige
Schweiz. Bern, Switzerland: Huber. (In German).
Carte ET, Nigg JT, Hinshaw SP. (1996) Neuropsychological functioning,
motor speed, and language processing in boys with and without
ADHD. Journal of Abnormal Child Psychology 24: 48198.
Castellanos FX, Giedd JN, Marsh WL, Hamburger SD, Vaituzis AC,
Dickstein DP, Sarfatti SE, Vauss YC, Snell JW, Lange N et al. (1996)
Quantitative brain magnetic resonance imaging in attention-deficit
hyperactivity disorder. Archives of General Psychiatry
53: 60716.
Chee P, Logan G, Schachar R, Lindsay P, Wachsmuth R. (1989) Effects
of event rate and display time on sustained attention in
hyperactive, normal, and control children. Journal of Abnormal
Child Psychology 17: 37191.
Denckla MB. (1974) Development of motor co-ordination in
normal children. Developmental Medicine & Child Neurology
16: 72941.
Rudel RG. (1978) Anomalies of motor development in
hyperactive boys. Annals of Neurology 3: 2313.
Fldnyi M, Imhof K, Steinhausen H-C. (2000) Zur klinischen
Validitt der computeruntersttzten Testbatterie (TAP) bei
Kindern mit Aufmerksamkeits- / Hyperaktivittsstrungen.
Zeitschrift fr Neuropsychologie. 11: 154167.
Gazzaniga MS, Sperry RW. (1966) Simultaneous double
discrimination response following brain bisection.
Psychonomic Science 4: 2612.
Gillberg C. (1995) Clinical Child Neuropsychiatry. New York:
Cambridge University Press.
Carlstrom G, Rasmussen P. (1983) Hyperkinetic disorders in
seven-year-old children with perceptual, motor and attentional
deficits. Journal of Child Psychology and Psychiatry
24: 23346.
Gillberg IC, Gillberg C. (1989) Children with preschool minor
neurodevelopmental disorders. IV: Behaviour and school
achievement at age 13. Developmental Medicine & Child
Neurology 31: 313.
Jeeves MA, Dixon NF. (1970) Hemisphere differences in response
rates to visual stimuli. Psychonomic Science 20: 24951.
Kadesjo B, Gillberg C. (1998) Attention deficits and clumsiness in
Swedish 7-year-old children. Developmental Medicine & Child
Neurology 40: 796804.
Kawabe S. (1989) Effects of force output and preparatory set on
premotor time of simultaneous bilateral response. Perceptual
and Motor Skills 68: 61925.
Largo RM, Caflisch JA. (1999) Neuromotor development from
kindergarten age to adolescence. In: AF Kalverboer and A
Gramsbergen, editors. Handbook on Brain and Behavior in
Human Development. (Forthcoming).
Lazarus JC, Todor JI. (1991) The role of attention in the regulation
of associated movement in children. Developmental Medicine
& Child Neurology 33: 329.
Leung PW, Connolly KJ. (1998) Do hyperactive children have motor
organization and/or execution deficits? Developmental Medicine
& Child Neurology 40: 6007.
McMahon SA, Greenberg LM. (1977) Serial neurologic examination
of hyperactive children. Pediatrics 59: 5847.

Michael RL, Klorman R, Salzman LF, Borgstedt AD, Dainer KB.

(1981) Normalizing effects of methylphenidate on hyperactive
childrens vigilance performance and evoked potentials.
Psychophysiology 18: 66577.
Mitchell S, Bradley VA, Welch JL, Britton PG. (1990) Coma arousal
procedure: a therapeutic intervention in the treatment of head
injury. Brain Injury 4: 2739.
Ohtsuki T. (1981) Increase in simple reaction time of knee
extension induced by simultaneous bilateral performance.
Perceptual and Motor Skills 53: 2730.
Oosterlaan J, Sergeant JA. (1998) Response inhibition and response
re-engagement in attention-deficithyperactivity disorder,
disruptive, anxious and normal children. Behavioral Brain
Research 94: 3343.
Overmeyer S, Simmons A, Santosh J, Andrew C, Williams SC, Taylor
A, Chen W, Taylor E. (2000) Corpus callosum may be similar in
children with ADHD and siblings of children with ADHD.
Developmental Medicine & Child Neurology 42: 813.
Piek JP, Pitcher TM, Hay DA. (1999) Motor coordination and
kinaesthesis in boys with attention-deficithyperactivity disorder.
Developmental Medicine & Child Neurology 41: 15965.
Rubia K, Oosterlaan J, Sergeant JA, Brandeis D, v. Leeuwen T. (1998)
Inhibitory dysfunction in hyperactive boys. Behavioral Brain
Research 94: 2532.
Schachar R. (1991) Childhood hyperactivity. Journal of Child
Psychology and Psychiatry 32: 15591.
Schaffer DM, Fisher PS-S, Cohen P, Piacentini J, Davies M, Conners
CK, Regier D. (1993) The diagnostic interview schedule for
children, revised version (DISC-R) I: Preparation, field testing,
interrater reliability, and acceptability. Journal of the American
Academy of Child and Adolescent Psychiatry 32: 64350.
Schuerholz LJ, Cutting L, Mazzocco MM, Singer HS, Denckla MB.
(1997) Neuromotor functioning in children with Tourette
syndrome with and without attention-deficithyperactivity
disorder. Journal of Child Neurology 12: 43842.
Semrud-Clikeman M, Filipek PA, Biederman J, Steingard R, Kennedy
D, Renshaw P, Bekken K. (1994) Attention-deficithyperactivity
disorder: magnetic resonance imaging morphometric analysis of
the corpus callosum. Journal of the American Academy of Child
& Adolescent Psychiatry 33: 87581.
Steger J, Imhof K, Steinhausen H-C, Brandeis D. (2000) Brain
mapping of bilateral interactions in attention-deficit
hyperactivity disorder and control boys. Clinical
Neurophysiology 111: 114156
Steinhausen HC, Metzke CW, Meier M, Kannenberg R. (1998)
Prevalence of child and adolescent psychiatric disorders: the Zurich
epidemiological study. Acta Psychiatrica Scandinavica 98: 26271.
van der Meere J, van Baal M, Sergeant J. (1989) The additive factor
method: a differential diagnostic tool in hyperactivity and learning
disability. Journal of Abnormal Child Psychology 17: 40922.
Vreeling HJ, Sergeant J. (1992) A motor presetting study in
hyperactive, learning disabled and control children. Journal of
Child Psychology and Psychiatry 33: 134754.
van Leeuwen TH, Steinhausen HC, Overtoom CCE, Pascual-Marqui
RD, vant Klooster B, Rothenberger A, Sergeant JA, Brandeis D.
(1998) The continuous performance test revisited with
neuroelectric mapping: impaired orienting in children with
attention deficits. Behavioral Brain Research 94: 97110.
Waber DP, Mann MB, Merola J, Bauermeister M, Cohen C, Ferber R,
Wolff PH. (1985) Motor overflow and attentional processes in
normal school-age children. Developmental Medicine & Child
Neurology 27: 4917.
Whitmont S, Clark C. (1996) Kinaesthetic acuity and fine motor
skills in children with attention-deficithyperactivity disorder: a
preliminary report. Developmental Medicine & Child Neurology
38: 10918.
Wyke M. (1971) The effects of brain lesions on the performance of
bilateral arm movements. Neuropsychologia 9: 3344.

Attentional and Neuromotor Deficits in ADHD J Steger et al. 179