czas i ADHD.pdf

2009, Vol. 23, No. 2, 265–269

0894-4105/09/$12.00 DOI:10.1037/a0014553

BRIEF REPORT

Contraction of Time in Attention-Deﬁcit Hyperactivity Disorder

David L. Gilden and Laura R. Marusich

The University of Texas at Austin

Attention-deﬁcit hyperactivity disorder (ADHD) has been associated with anomalies in dopamine

systems. Recent advances in the understanding of the core cognitive deﬁcits in ADHD suggest that

dopamine dysfunction might be expressed through shortened time scales in reward-based learning. Here

this perspective is extended by the conjecture that temporal span in working memory systems might

generally be shortened. As a test of this conjecture the authors focus on the implicit memory system

involved in rhythmic movement, assessing the minimum tempo at which rhythmic feeling can be

sustained in adults with diagnosed ADHD and in a control group of normal adults. The authors found that

people with ADHD do in fact have a rhythm cut-off that is faster in tempo than those without ADHD.

This ﬁnding is consistent with the idea that impaired dopamine dynamics have systemic consequences for

cognitive function, essentially recalibrating the clock that sets the time scale for the subjective experience

of temporal events.

Keywords: attention deﬁcit, timing, dopamine, musical performance, l/f noise

Attention-deﬁcit hyperactivity disorder (ADHD) is a syndrome

that impacts cognitive functions essential to the moment-to-mo-

ment apprehension of and response to the environment. An inﬂu-

ential cognitive theory of ADHD identiﬁes impairments in the

executive control of inhibition as being a principal deﬁcit (Barkley,

1997). Inhibitory control is operationalized through tasks that

require the withholding of response: so-called Go/No-Go methods.

In one version of the methodology (Conners, 1995), the dependent

measures have been standardized to serve as a diagnostic tool. Yet

Go/No-Go methods generally fail to generate consistent group

differences based on ADHD diagnosis (Castellanos, Sonuga-

Barke, Milham, & Tannock, 2006; Edwards et al., 2007). Never-

theless, one prediction of the inhibitory control theory, that ADHD

would lead to deﬁcits in temporal aspects of working memory

(Barkley, 1997), has led to a number of interesting ﬁndings re-

garding the perception of time and the planning of behavior, delay

aversion being most notable (Sonuga-Barke, Saxton, & Hall,

1998). The notion that time perception might be altered by ADHD

is supported by neuroimaging evidence of volumetric reductions

(Castellanos et al., 2002; Valera, Faraone, Murray, & Seidman,

2007) in areas known to control and regulate timing: prefrontal

cortex (Mangels, Ivry, & Shimizu, 1998; Smith, Taylor, Lidzba, &

Rubia, 2003) and cerebellum (Ivry & Spencer, 2004; Mangels et

al., 1998). Yet, direct psychophysical assessments of time percep-

tion have also failed to yield consistent group differences.

Timing behavior is generally assessed in ADHD populations

through explicit tests in which the participant must focus attention on

the passage of time per se. These tests are quite diverse and include

anticipating a future event scheduled some few seconds or minutes in

the future (Rubia, Taylor, & Taylor, 1999; Smith, Taylor, Rogers,

Newman, & Rubia, 2002; Sonuga-Barke et al., 1998), synchro-

nized tapping with an external signal (Rubia et al., 1999), discrim-

inating between two given durations (Rubia et al., 1999; Smith

et al., 2002; Toplak & Tannock, 2005), or generating a few

instances of a given interval (Barkley, Murphy, & Bush, 2001;

Kerns, McInerney, & Wilde, 2001; Smith et al., 2002). The results

of these studies have varied in terms of whether people with

ADHD showed timing biases toward shorter intervals, but ADHD

timing behavior consistently shows more variability than that of

controls. Such a result, however, provides little information about

timing mechanisms in ADHD because greater variability is uni-

versally found in ADHD cognitive assessment (Castellanos &

Tannock, 2002).

What is required here are implicit tests where timing behavior is

allowed to emerge as a byproduct of an activity that proceeds

without reﬁned judgment and discrimination. There is an implicit

aspect of timing behavior that is universally experienced and

eminently suitable for psychophysical assessment: the feeling that

emerges when we experience rhythm. Such feelings are a prime

example of Gestalt; the whole (rhythm) is greater than the sum of

the parts (individual time intervals). When we feel rhythm the

experience is of the feeling; the actual intervals that create the

feeling recede into the background. In this sense the data of

interest, the intervals so marked, arise implicitly. We refer to this

foreground/background distinction when we inquire if a person is

feeling rhythm. Having drawn the distinction, it must be remarked

that its relevance to ADHD cognition is not obvious. There are

manifestly many musicians who have ADHD; musical or dancing

ineptitude is not part of the symptom cluster of ADHD.

David L. Gilden and Laura R. Marusich, Department of Psychology, The

University of Texas at Austin.

Financial Support for this research was provided by National Institutes

of Mental Health Grant R01-MH58606 and National Science Foundation

Grants BCS-0226449 and BCS-0744989.

Correspondence concerning this article should be addressed to David L.

Gilden, 1 University Station A8000, Austin, TX 78712. E-mail: gilden@psy

.utexas.edu

265

Neuropsychology

266

BRIEF REPORT

That there may be a connection between ADHD and rhythmic

expression is suggested by evidence that dopamine pathways

are affected in ADHD (Volkow et al., 2007), and that dopamine

neurons are tuned to event salience, a key stimulus attribute in

the initiation of reward-based learning. Sagvolden et al. (2005)

have argued that loss of salience might be functionally expressed

by a reward system that is developmentally attuned to shortened

delays. At this level of explanation, the symptoms of ADHD (short

attention span, hyperactivity, impulsivity) are merely the macro-

scopic outcomes of steep delay-of-reinforcement gradients. In this

article we conjecture that time scale shortening in conditioned

learning may have consequences generally for the organization of

behavior, and in particular for the repetitive behavior experienced

as rhythmic movement.

The production of rhythm is a human capacity about which quite

a bit is already known. Practical musical performance mandates

the construction of metronomes that reﬂect true human capacities

in timekeeping, and most metronomes have a limiting largo setting

at 40 beats per minute (bpm), an interbeat interval of 1.5 seconds.

This is practically the slowest setting at which music can be

counted. Our conjecture is that rhythmic feeling may be subject to

a shorter limiting interval in ADHD. The suggestion is not that

people with ADHD cannot feel rhythm, but that there would be no

need to build a metronome that extends to 40 bpm if the intended

user has ADHD. A smaller pendulum with a shorter maximum

period would sufﬁce.

ADHD) whose task was simply to produce a steady train of

drumbeats at various tempi. In each tempo condition the partici-

pant drummed with a synchronizing signal for 16 beats and then

continued without the signal for another 60 beats. Most of the

pertinent facts about rhythm are deducible from simply noting

whether the performances appear stable or whether they wander

about. At tempi faster than 40 bpm the performances are mani-

festly stable; their associated time series seem to have a well-

deﬁned mean and variance. Performances at tempi slower than 40

bpm lack this apparent stability, showing a pronounced tendency

to meander. Meandering is a speciﬁc kind of losing one’s way that

is characteristic of a random walk process.

Random walks are generated by an imperfect copying mecha-

nism that has the recursive form: X(t 1) *X(t) ε (t), where

determines the rate at which successive copies lose their corre-

lation over generations, and ε is the random step that drives the

walk to meander. This relation provides an important insight into

drumming performances at tempi a little slower than the 40-bpm

metronome limit. Consider the options available to a drummer who

is asked to perform so slowly that he or she cannot feel the beat.

How does such a person know when to strike the drum when the

implicit body knowledge, the feel, is missing? One strategy is to

use the recollection of the most recent interval as a guide for when

it would be judicious to next strike the drum. This strategy natu-

rally generates the recursive pattern observed in the random walk.

From the point of view of an observer of drumming performance,

inferences about whether somebody else is feeling rhythm comes

down to deciding whether the performance indicates the recursive

use of explicit memories.

Statistics that are sensitive to recursion are quite different from

those that have been typically used to measure timing behavior.

Most assessments of the human capacity to produce regular pulses

are made at tempi well within the range of rhythmic feel (Allan,

1979; Wearden, 1991), and the sole focus has been on overall

accuracy as measured by the coefﬁcient of variation, (standard

deviation)/mean. When drumming performance is stable both the

mean and standard deviation are well-deﬁned in the sense that they

do not themselves change during the performance. In this case, the

coefﬁcient of variation is a true measure of relative error. How-

ever, in meandering performances the mean drifts, and this makes

the interpretation of the coefﬁcient of variation problematic. As-

sessments made at very slow tempi require statistics that do not

assume a stable mean but rather quantify the tendency to drift.

Such statistics focus on how well past performance predicts future

performance. Meandering performance tends to be less accurate

but more predictable than stable performance.

The most common statistic used in psychology to capture pre-

dictive accuracy is correlation. In time series analysis the concept

of correlation generalizes to the autocorrelation function, the cor-

relation of the series with a copy of itself that has been displaced 1,

2, . . ., k, trials. The autocorrelation function has demonstrated

utility in the assessment of rhythmic tapping (Madison, 2001).

Related to the autocorrelation function is its Fourier twin, the

power spectrum. The power spectrum is especially useful here

because it employs trial wavelength rather than trial separation to

partition the correlations, and drumming performance is more

easily described in terms of wavelength. In the power spectrum

global features such as hills and valleys are resolved at the long

wavelengths (low frequencies) while the inevitable beat-to-beat

Theoretical Issues in the Assessment of Rhythmic Feel

A recorded drumming performance is literally a succession of

marked moments in time: beats. The succession of intervals be-

tween beats forms a time series that may be used to assess the

feeling of rhythm. Examples of drumming performance displayed

as interbeat-interval time series are illustrated in Figure 1. These

data were produced by a normal adult (one not diagnosed with

Figure 1. Time series of successive interval estimates are shown for non-

synchronized drum beating. Target tempos of 15, 30, 40, 60, and 120 bpm

(target time intervals of 4, 2, 1.5, 1, and 0.5 seconds, respectively). Loss of

rhythm is implied by wandering estimates at tempo slower than 40 bpm.

BRIEF REPORT

267

variation is resolved at the short wavelengths (high frequencies).

Spectral representations are commonly used in the analysis of

stable timekeeping (Gilden, Thornton, & Mallon, 1995; Lemoine,

Torre, & Delignieres, 2006).

Beyond correlation are statistics that measure nonlinear aspects

of prediction. The sample entropy (Richman & Moorman, 2000) is

one such statistic that has proven worth as a clinical tool in

diagnosing cardiomypathy through wandering heartbeat (Norris,

Stein, Cao, & Morris, 2006), and promises to be quite useful in

assessing rhythmic feel. The sample entropy measures the ten-

dency for micropatterns in the time series to repeat. Formally, the

sample entropy is an average conditional probability; given that X

was followed by Y in the past, how likely is it that an event

resembling X will be followed by an event resembling Y. Signals

that drift generally have lower entropy than signals that ﬂuctuate

about a stable mean.

each condition, participants slapped the drum with their dominant

hand in time with a synchronizing signal for 16 beats and then

continued without the signal for another 3 minutes. To prevent

counting, participants drummed while reading aloud from a non-

technical book review printed in large clear type. 1

Results

Figure 2 displays key statistics at the three target tempi: (A)

coefﬁcient of variation, (B) power spectrum in log-log coordinates,

of variation as a measure of overall accuracy, recognizing that its

utility is compromised by drifting performance. The power spectrum

(see Thornton and Gilden (2005) for computational method), serial

correlation, and entropy are used to characterize drift. It is evident

from this ﬁgure that the statistics are resolving large group differences

in drumming performance at the target tempi. The tempo functions

are, however, fairly complex, and the data are best understood by

considering the conditions in the order of fastest to slowest tempi,

progressing from manifest rhythm to its complete breakdown.

At the fastest tempo, 60 bpm (1-s target), the ADHD and normal

control groups have similar performance characteristics. They have

similar accuracies, nonlinear predictability (entropy), and low fre-

quency power. As the power at low frequencies picks up the large-

scale hill-valley structure, equality between the two groups implies

that they have comparable stability in the long run. Where the two

groups differ in spectral power is at the high frequencies. Here the

normal control group actually shows greater ﬂuctuation magnitude,

and this is reﬂected by the marginally larger coefﬁcient of variation in

the normal control group. That the high frequency ﬂuctuations are

relatively smaller in the ADHD group makes their performances

slightly more predictable in the short run as evidenced by the en-

hanced serial correlation. The ﬁnding that ADHD performance is not

in any way compromised at 60 bpm is an important benchmark for

our method. If the secondary counting suppression task differentially

disrupted drumming performance in the ADHD group it might be

expected to enter as a main effect across tempo conditions.

Group differences at 40 bpm are quite large and appear in the three

measures of sequential correlation that are sensitive to the loss of

rhythmic feel and the onset of random walking. Comparing ADHD

Empirical Assessment of Rhythmic Feel

The following pilot study assessed whether adults with ADHD

lose rhythm at faster tempo than do normal controls. The study

focused on the range of tempo where people typically lose their

sense of rhythm and was designed to provoke both stable and

wandering behavior in the two groups.

Method

Participants

Eleven adults with a diagnosis of ADHD and 11 adults without

ADHD participated in the study, which was approved by the

institutional review board. All participants were students at

the University of Texas and were between 18 and 30 years of age.

The participants with ADHD were referred by the Ofﬁce for

Students with Disabilities. To register with the Ofﬁce for Students

with Disabilities, students must have a DSM–IV or ICD diagnosis

of ADHD from an external clinician, and they must have received

a psychological evaluation in the past 3 years to demonstrate that

their assessment is current. All of the participants with ADHD had

received a clinical diagnosis, and about half received this diagnosis

in childhood. Those who were diagnosed as adults all reported

childhood-onset symptoms, as required by DSM–IV criteria. The

participants with ADHD had not taken any medication in the 24

hours before the study. After complete description of the study to

the subjects, written informed consent was obtained.

1 Interval discrimination studies have demonstrated that counting mark-

edly improves performance when durations exceed about 1.2 seconds

(Grondin, Mielleur-Wells, & Lachance, 1999). Counting must be pre-

vented in assessments of rhythm at target tempi slower than 60 bpm, if we

wish to ensure that performers do not substitute their own counting beat for

the intended target. Covert counting is a problem not often encountered in

studies of timing behavior as durations exceeding a second are rarely

examined. Madison’s (2001) work is almost unique in this regard, and he

dealt with this problem by simply instructing the participants to not count.

Such an instruction places conﬂicting task demands upon the participant as

there is always the implicit requirement in any drumming study that

performance be as accurate as possible. An alternative strategy is to

suppress counting through a secondary task that requires articulation, one

that is highly practiced and relatively automatic. Reading elementary text

ﬁts this requirement and is the tactic used in this pilot study.

2 The sample entropy is a function of two parameters, m - the dimension

of the embedding space, and r - the neighborhood size. Here m 2, and

r .2 , where is the interval standard deviation. These values are typical

of practical application (Norris et al., 2006).

Materials

A Roland Handsonic percussion controller was used for the col-

lection of data. This device has time resolution comparable to the

keyboard but is superior to the keyboard in affording good tactile and

auditory feedback. The percussion controller was set to simulate

conga drums.

Procedure

Participants each completed three target-tempi conditions of 30,

40, and 60 bpm, corresponding to interbeat intervals of 2, 1.5,

and 1 second, respectively. These tempi were chosen to span the

critical interval where rhythm breaks down in normal function. In

268

BRIEF REPORT

Figure 2. Statistical measures of drumming performances produced by adults with ADHD (ﬁlled circles) and by normal

controls (open circles). Performance measures are given for the target interbeat intervals 1, 1.5, and 2 seconds, corresponding

to tempi of 60, 40, and 30 bpm. (A) Coefﬁcient of variation standard deviation/mean. (B) Log power as a function of log

frequency. (C) Autocorrelation at lag 1. (D) Sample entropy. Error bars depict standard error of the mean.

with normal control participants, their spectra were steeper, t (20)

2.5, p .01, their entropies were lower, t (20) 1.88, p .04,

and their serial correlations were larger, t (20) 2.77, p .01. All

three of these statistics suggest that 40 bpm deﬁnes a new regime

of behavior for the ADHD group, a regime where the sense of

rhythm has become so diminished that they substitute replicates of

their previous estimates for the target interval. The coefﬁcient

of variation in the ADHD group is also quite large at 40 bpm

compared to 60 bpm, t (10) 2.55, p .02, providing further

evidence that the nature of the performances at the two tempi are

quite different. It is equally evident that the normal control group

performance is rhythmically stable at 40 bpm, justifying the ex-

tension of the metronome to this tempo. On every statistical

measure of correlation as well as in the coefﬁcient of variation, the

differences between 40 bpm and 60 bpm in the normal control

group were small and not signiﬁcant.

Finally we consider performance beyond the metronome limit,

at 30 bpm. In this condition neither group produces competent

drumming. The coefﬁcient of variation in both groups is some 50%

larger than it was at 60 bpm, a sure sign that rhythmic feeling has

been compromised (normal control t (10) 4.8, ADHD

t (10) 3.2). Consistent with previous ﬁndings (Madison, 2001),

the normal control group meanders at 30 bpm: the serial correla-

tions are almost triple that in the two faster conditions ( r .2 vs.

r .08, t (31) 2.0, p .025). As at 60 bpm, the normal control

group is slightly less accurate than the ADHD group, as measured

by the coefﬁcient of variation. This difference is due, as it was

at 60 bpm, to the fact that the normal control group actually

ﬂuctuates a little more wildly from beat to beat. Again, the relative

suppression of power at the high frequencies in the ADHD group

gives them a larger serial correlation. However, the groups are not

distinguished by the sample entropy. Both groups are sufﬁciently

erratic in the short run that even in the presence of strong drift,

micropatterns in the time series are poor predictors.

The observation that ADHD performance is both slightly more

stable and more accurate than that of the normal controls at the two

bounding tempi, 60 and 30 bpm, suggests that the secondary task

of reading is not differentially diminishing ADHD performance. If

the presence of a secondary task is critical in producing group

differences at 40 bpm it must be that the ADHD group is more

vulnerable at this tempo. If the ADHD group is more vulnerable to

the effects of a secondary task only at 40 bpm it must be that their

sense of rhythm is compromised differentially at 40 bpm. As this

is essentially the conjecture we seek to test, the secondary task is

justiﬁed as a reasonable measure to prevent counting.

Discussion

This pilot study of ADHD rhythmic behavior resolves three distinct

performance regimes that exemplify two different aspects of working

BRIEF REPORT

269

memory. At fast tempo (60 bpm) both ADHD and normal control

groups execute accurate and stable performances. The interpretation

of this ﬁnding is that the system of implicit working memory that

produces rhythmic feel has a temporal span that exceeds 1 second in

both groups. At rhythmic the metronome limit (40 bpm), this implicit

memory system appears to be unavailable to people with ADHD, and

their performance reﬂects a different and more explicit usage of

memory, memory of their most recent estimates. As normal controls

display rhythmic feel at 40 bpm, this difference implies a difference

in the span of rhythmic feel. The effective span of rhythmic feel is

apparently contracted in ADHD adults to less than 1.5 seconds. At 30

bpm, both the implicit sense of rhythm and the explicit sense of prior

estimates are largely attenuated in both groups. This is presumably

why metronomes typically do not offer a setting at 30 bpm. If there is any

distinction between the groups at this tempo, it is that the ADHD group

has a better sense of what they are doing, as they seem to have a better

sense of their most recent estimates. Outside of this observation there is

little to distinguish normal control from ADHD performance at 30 bpm.

In this article we have considered one aspect of temporal integra-

tion, rhythmic feeling, providing preliminary evidence that this feel-

ing exists over a relatively restricted tempo range in ADHD. Rhyth-

mic feeling is a good place to begin an analysis of the more general

issue of temporal integration because it is methodologically simple, it

is supported by clear signatures in data, and it involves a system of

measurement that does not require sustained vigilance. These prelim-

inary results, however, do not remotely exhaust the range of cognition

that is involved in the perceptual organization of temporally based

events. Virtually any percept or activity that involves synthesis across

time is potentially of interest. In particular, ethological investigation of

human actions that imply universal time scales (Schleidt, Eible-

Eibesfeldt, & P¨ppel, 1987) may be a particularly productive way of

investigating ADHD temporality and in understanding what exactly is

implied by hyperactivity and inattentiveness.

Gilden, D. L., Thornton, T., & Mallon, M. (1995). 1/ƒ noise in human

cognition. Science, 267, 1837–1839.

Grondin, S., Meilleur-Wells, G., & Lachance, R. (1999). When to start

explicit counting in a time-intervals discrimination task: A critical point

in the timing process of humans. Journal of Experimental Psychology:

Human Perception and Performance, 25, 993–1004.

Ivry, R. B., & Spencer, R. M. C. (2004). The neural representation of time.

Current Opinion in Neurobiology, 14, 225–232.

Kerns, K. A., McInerney, R. J., & Wilde, N. J. (2001). Time reproduction,

working memory, and behavioral inhibition in children with ADHD.

Child Neuropsychology, 7, 21–31.

Lemoine, L., Torre, K., & Delignieres, D. (2006). Testing for the presence

of 1/f noise in continuation tapping data. Canadian Journal of Experi-

mental Psychology, 60, 247–257.

Madison, G. (2001). Variability in isochronous tapping: Higher order

dependencies as a function of intertap interval. Journal of Experimental

Psychology: Human Perception and Performance, 27, 411–422.

Mangels, J. A., Ivry, R. B., & Shimizu, N. (1998). Dissociable contribu-

tions of the prefrontal and neocerebellar cortex to time perception.

Cognitive Brain Research, 7, 15–39.

Norris, P. R., Stein, P. K., Cao, H., & Morris, J. J., Jr. (2006). Heart rate

multiscale entropy reﬂects reduced complexity and mortality in 285

patients with trauma. Journal of Critical Care, 21, 343.

Richman, J. S., & Moorman, J. R. (2000). Physiological time-series analysis

using approximate entropy and sample entropy. American Journal of Phys-

iology - Heart and Circulatory Physiology, 278, 2039–2049.

Rubia, K., Taylor, A., & Taylor, E. (1999). Synchronization, anticipation, and

consistency in motor timing of children with dimensionally deﬁned atten-

tion deﬁcit hyperactivity behaviour. Perceptual and Motor Skills, 89, 1237–

1258.

Sagvolden, T., Johansen, E. B., Aase, H., & Russell, V. A. (2005). A

dynamic developmental theory of attention-deﬁcit/hyperactivity disorder

(ADHD) predominantly hyperactive/impulsive and combined subtypes.

Behavioral Brain Science, 28, 397–419.

Schleidt, M, Eibl-Eibesfeldt, I., & P¨ppel, E. (1987). A universal constant

in temporal-segmentation of human short-term behavior. Naturwissen-

schaften, 74, 289–290.

Smith, A., Taylor, E., Lidzba, K., & Rubia, K. (2003). A right hemispheric

frontocerebellar network for time discrimination of several hundreds of

milliseconds. Neuroimage, 20, 344–350.

Smith, A., Taylor, E., Rogers, J. W., Newman, S., & Rubia, K. (2002).

Evidence for a pure time perception deﬁcit in children with ADHD.

Journal of Child Psychology and Psychiatry, 43, 529–542.

Sonuga-Barke, E. J. S., Saxton, T., & Hall, M. (1998). The role of interval

underestimation in hyperactive children’s failure to suppress responses

over time. Behavioural Brain Research, 94, 45–50.

Thornton, T. L., & Gilden, D. L. (2005). Provenance of correlations in

psychological data. Psychonomic Bulletin and Review, 12, 409–441.

Toplak, M. E., & Tannock, R. (2005). Time perception: Modality and

duration effects in attention-deﬁcit/hyperactivity disorder (ADHD).

Journal of Abnormal Child Psychology, 33, 639–654.

Valera, E. M., Faraone, S. V., Murray, K. E., & Seidman, L. J. (2007).

Meta-analysis of structural imaging ﬁndings in attention-deﬁcit/hyper-

activity disorder. Biological Psychiatry, 61, 1361–1369.

Volkow, N. D., Wang, G.-J., Newcorn, J., Telang, F., Solanto, M. V., Fowler,

F. S., et al. (2007). Depressed dopamine activity in caudate and preliminary

evidence of limbic involvement in adults with attention-deﬁcit/hyperactivity

disorder. Archives of General Psychiatry, 64, 932–940.

Wearden, J. H. (1991). Do humans possess an internal clock with scalar-

timing? Learning and Motivation, 22, 59–83.

References

Allan, L. G. (1979). The perception of time. Perception & Psychophys-

ics, 26, 340–354.

Barkley, R. A. (1997). Behavioral inhibition, sustained attention, and

executive functions: Constructing a unifying theory of ADHD. Pyscho-

logical Bulletin, 121, 65–94.

Barkley, R. A., Murphy, K. R., & Bush, T. (2001). Time perception and

reproduction in young adults with attention deﬁcit hyperactivity disor-

der. Neuropsychology, 15, 351–360.

Castellanos, F. X., Lee, P. P., Sharp, W., Jeffries, N. O., Greenstein, D. K.,

Clasen, L. S., et al. (2002). Developmental trajectories of brain volume

abnormalities in children and adolescents with attention-deﬁcit/hyper-

activity disorder. Journal of the American Medical Association, 288,

1740–1748.

Castellanos, F. X., Sonuga-Barke, E. J. S., Milham, M. P., & Tannock, R.

(2006). Characterizing cognition in ADHD: Beyond executive dysfunc-

tion. Trends in Cognitive Sciences, 10, 117–123.

Castellanos, F. X., & Tannock, R. (2002). Neuroscience of attention-

deﬁcit/hyperactivity disorder: The search for endophenotypes. Nature

Reviews Neuroscience, 3, 617–628.

Conners, C. K. (1995.) Conners’ Continuous Performance Test . Toronto,

Multi-Health Systems.

Edwards, M. C., Gardner, E. S., Chelonis, J. J., Schulz, E. G., Flake, R. A.,

& Diaz, P. F. (2007). Estimates of the validity and utility of the Conners’

Continuous Performance Test in the assessment of inattentive and/or

hyperactive-impulsive behaviors in children. Journal of Abnormal Child

Psychology, 35, 393–404.

Received July 18, 2008

Revision received October 15, 2008

Accepted October 27, 2008

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: