Journal of Clinical and Experimental Neuropsychology

ISSN: 1380-3395 (Print) 1744-411X (Online) Journal homepage: http://www.tandfonline.com/loi/ncen20

Speed of processing and strategic control of

attention after traumatic brain injury

Alicia Rhian Dymowski, Jacqueline Anne Owens, Jennie Louise Ponsford &
Catherine Willmott

To cite this article: Alicia Rhian Dymowski, Jacqueline Anne Owens, Jennie Louise Ponsford &
Catherine Willmott (2015) Speed of processing and strategic control of attention after traumatic
brain injury, Journal of Clinical and Experimental Neuropsychology, 37:10, 1024-1035, DOI:
10.1080/13803395.2015.1074663

To link to this article: http://dx.doi.org/10.1080/13803395.2015.1074663

Published online: 01 Sep 2015.

Submit your article to this journal

Article views: 488

View related articles

View Crossmark data

Citing articles: 10 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=ncen20

Download by: [Universitat Oberta de Catalunya] Date: 30 October 2017, At: 04:43
 Journal of Clinical and Experimental Neuropsychology, 2015
Vol. 37, No. 10, 10241035, http://dx.doi.org/10.1080/13803395.2015.1074663

Speed of processing and strategic control of attention

after traumatic brain injury
Alicia Rhian Dymowski1,2, Jacqueline Anne Owens1,2, Jennie Louise Ponsford1,2,3,
and Catherine Willmott1,2
1
School of Psychological Sciences, Monash University, Clayton, VIC, Australia
2
Monash-Epworth Rehabilitation Research Centre, Epworth HealthCare, Richmond, VIC, Australia
3
Centre of Excellence in Traumatic Brain Injury Rehabilitation, National Trauma Research Institute,
Alfred Hospital, Prahran, VIC, Australia
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017

(Received 19 May 2015; accepted 15 July 2015)

Introduction: Slowed information processing speed has consistently been documented after traumatic brain injury (TBI).
Debate continues as to whether deficits in strategic control are proportionate to, or remain after controlling for, reduced
speed of processing. The study aim was to investigate the association of speed of processing and strategic control of
attention with working memory, selective attention, response inhibition, and mental flexibility task performance after
TBI using traditional and novel clinical measures. Method: Twenty-five individuals with complicated mild to severe TBI
(post-traumatic amnesia duration, M = 39.52 days, SD = 38.34; worst Glasgow Coma Scale score, M = 7.33, SD = 4.35;
time post-injury, M = 392.64 days, SD = 537.19) and 25 matched healthy controls completed assessment of attentional
and executive functioning. Measures included the Symbol Digit Modalities Test (SDMT), the computerized Selective
Attention Task (SAT), the Ruff 2&7 Selective Attention Test (2&7), the visual n-back, Digit Span, Hayling Test, and
Trail Making Test (TMT). Results: t tests revealed that individuals with TBI demonstrated reduced processing speed on
the SDMT, n-back, SAT, 2&7, Hayling Test, and TMT-A (p .002 for all). Digit Span performance did not differ
between groups. Mixed-model ANOVAs revealed that individuals with TBI demonstrated a disproportionate increase
in reaction time with complexity, which was accounted for by speed on the SAT but remained on the Hayling Inhibition
Test after controlling for speed in ANCOVAs. Mann-Whitney U tests revealed that individuals with TBI also made
more errors on the Hayling Test, missed responses on the n-back and were unable to benefit from the automatic
condition of the 2&7. Conclusions: While slowed speed of information processing was pervasive across tasks after TBI,
residual difficulties in response inhibition remained after controlling for slowness, which suggests impaired strategic
control. These findings support targeted intervention for slowed speed of thinking and inhibition following TBI.

Keywords: Traumatic brain injury; Attention; Information processing speed; Strategic control; Working memory.

Attentional deficits are common and disabling
after traumatic brain injury (TBI; Olver,
Ponsford, & Curran, 1996) and can interfere with
everyday roles (Lewis & Horn, 2013). Slowed pro-
cessing speed and difficulty with ignoring
distractions, concentrating for extended periods,
and paying attention to more than one thing at a
time are commonly reported by individuals with
TBI and their family members (Olver et al., 1996;
Ponsford & Kinsella, 1991).

The authors would like to thank the participants and their families who generously gave their time. The button box for the computer
task was designed and built by Antonio Benci at the Monash Facility for Instrumentation & Technology Development. Thanks to
Christopher Hocking for programming the n-back task.
This work was supported by the Epworth Research Institute [Grant 80940].
The Epworth Research Institute was not involved in study design, analysis and interpretation of data, writing of the report or
decision to submit the article for publication. The authors report no conflicts of interest.
Address correspondence to: Alicia Dymowski, Monash-Epworth Rehabilitation Research Centre, 185187 Hoddle Street,
Richmond, 3121, VIC, Australia (E-mail: Alicia.Dymowski@monash.edu).

2015 Taylor & Francis

 PROCESSING SPEED AND ATTENTIONAL CONTROL AFTER TBI
The definition of attention remains controversial
(Styles, 2006; Whyte, Ponsford, Watanabe, &
Hart, 2010). Within limited capacity constraints,
attention serves to facilitate task performance by
selecting internal and external stimuli for active
processing (James, 1890; Raz, 2004). Attention is
not a unitary process, and numerous attentional
network models have been proposed. Whyte et al.
(2010) proposed that component processes of
attention include arousal, selection, strategic (or
executive) control, and processing speed. A two-
process model of attention distinguishing auto-
matic from controlled processing was proposed
by Schneider and Shiffrin (1977). Automatic pro-
cessing occurs in parallel for well-learned tasks not
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
requiring effortful control, whereas controlled pro-
cessing is serial in nature, involves conscious pro-
cessing, and has a limited capacity and rate. The
concepts of automatic and controlled processing
are incorporated in an information processing
model that describes contention scheduling and
the Supervisory Attentional System (SAS;
Norman & Shallice, 1980; Shallice, 1982).
Contention scheduling uses existing automatic sen-
sorimotor schemas to coordinate action for well-
learned behaviors and thoughts, whereas the SAS
is activated when tasks are novel or complex or
when automatic behaviors need to be overridden.
The SAS is akin to the central executive in
Baddeleys (1986, 2000) model of working mem-
ory. Strategic control (used in this paper to refer to
the executive processes of attention, incorporating
the concepts of the SAS and central executive)
directs, manages, and monitors limited attentional
resources in a goal-directed manner for tasks
requiring controlled processing, including sus-
tained, selective, divided, and shifting attention
and working memory (Whyte et al., 2010).
Hence, impairment of strategic control should
result in deficits in these cognitive domains
(McDowell, Whyte, & DEsposito, 1997; Serino
et al., 2006).
The high prevalence of attentional deficits fol-
lowing TBI is understandable, given the neuro-
pathology and neurochemistry of TBI and the
distributed nature of attentional networks.
Anterior regions of the brain, including the prefron-
tal cortex, are proposed to be involved in strategic
control of attention (MacDonald, Cohen, Stenger,
& Carter, 2000) and are particularly vulnerable to
insult (Bigler, 2001; Levin, Williams, Eisenberg,
High, & Guinto, 1992). Furthermore, diffuse axo-
nal injury (DAI) can result in widespread axonal
disconnection, resulting in the disruption of ascend-
ing catecholaminergic and serotonergic projections,
which are key neuromodulators of attentional
1025
networks (Povlishock, 1992). Acute cerebral eleva-
tions of dopamine and noradrenaline have been
reported in cerebrospinal fluid following TBI, and
return of these neurotransmitters to normal concen-
trations appears to be related to functional recovery
(Markianos, Seretis, Kotsou, Baltas, &
Sacharogiannis, 1992; Markianos, Seretis, Kotsou,
& Christopoulos, 1996).
Research has consistently demonstrated that
individuals with TBI experience reduced speed of
information processing across the spectrum of
injury severity (Felmingham, Baguley, & Green,
2004; Frencham, Fox, & Maybery, 2005; Mathias
& Wheaton, 2007; Ponsford & Kinsella, 1992;
Willmott, Ponsford, Hocking, & Schnberger,
2009). Researchers have postulated that difficulties
on complex attentional tasks demonstrated by indi-
viduals with TBI are largely accounted for by, and
proportionate to, slowed processing speed
(Felmingham et al., 2004; Ponsford & Kinsella,
1992; Spikman, van Zomeren, & Deelman, 1996;
Veltman, Brouwer, van Zomeren, & van
Wolffelaar, 1996). Others have argued that deficits
in strategic control of attention remain after
accounting for reduced processing speed, particu-
larly on tasks with greater complexity or increased
working memory load (Asloun et al., 2008; Azouvi,
Jokic, van der Linden, Marlier, & Bussel, 1996;
Park, Moscovitch, & Robertson, 1999; Serino
et al., 2006). Strategic control of attention has
been assessed with working memory tasks that
require controlled processing via maintenance and
manipulation of limited attentional resources.
Perlstein et al. (2004) parametrically varied working
memory load using a visual n-back task and found
severity-dependent and load-related working mem-
ory deficits in error rates, but no group differences
in reaction time (RT), suggesting deficits in strategic
control rather than speed. Yet evidence suggests
that both speed and strategic control may be
impaired after TBI. Willmott et al. (2009) found
that speed, but not working memory, accounted
for group differences in RT on a complex Selective
Attention Task (SAT); however, participants with
TBI also made more errors and missed responses,
which suggests impairments in strategic control. In
addition, Ros, Periez, and Muoz-Cspedes
(2004) found that group differences were accounted
for by reduced speed of thinking on an inhibition
task (Stroop Test) but not on a mental flexibility
task (Trail Making Test; TMT), suggesting that
attentional processes may be differentially influ-
enced by speed of processing and strategic control.
Individuals with TBI have demonstrated reduced
processing speed on the Hayling Test (Spitz et al.,
2013), a measure of response inhibition, but it is
 1026 DYMOWSKI ET AL.
unclear whether there is a differential effect of group
with increased executive load on this task.
The aim of the current study was to investigate
the association between speed of thinking and stra-
tegic control on traditional and novel attentional
tasks requiring controlled processing in people
with TBI relative to healthy controls. Given the
distributed nature of attentional networks and cor-
respondingly diverse neuropathological and neuro-
chemical disruption after TBI, it was hypothesized
that patients with TBI would demonstrate reduced
speed of information processing. Furthermore,
given the prevalence of lesions in frontal regions
responsible for strategic control, it was proposed
that on working memory, selective attention, and
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
mental flexibility tasks requiring controlled proces-
sing, participants with TBI would demonstrate a
disproportionate increase in RT, which could not
be accounted for by processing speed (i.e., after
controlling for the automatic processing condi-
tion), and would make more errors and missed
responses than controls. A particularly novel
aspect of this research was to use the Hayling
Test to investigate the association of speed and
strategic control on response inhibition. It was
hypothesized that individuals with TBI would
demonstrate a disproportionate increase in RT,
which could not be accounted for by initiation
speed, and would make more errors on this task
relative to healthy controls.
METHOD
Participants
Participants were recruited from the Acquired
Brain Injury Rehabilitation service at Epworth
HealthCare, Melbourne, Australia. Eligibility cri-
teria included: (a) history of complicated mild to
severe TBI; (b) age between 16 and 65 years; (c)
Demographics and injury characteristics
Gender (male)
Age (years): Mean (SD)
Education (years): Mean (SD)
% Employed
SES: Mean (SD)
NART FSIQ: Mean (SD)
GCS (worst score): Mean (SD) [Range]
PTA (days): Mean (SD) [Range]
Note. TBI = traumatic brain injury; SES = socioeconomic status as measured by the Scale of Occupational Prestige (Daniel, 1983);
NART = National Adult Reading Test; FSIQ = Full Scale Intelligence Quotient; GCS = Glasgow Coma Scale; PTA = post-traumatic
amnesia.
sufficient understanding of English to complete
the tasks; (d) sufficient cognitive ability to com-
plete the tasks; (e) dominant hand use; and (f) no
history of significant neurological or psychiatric
disturbance.
Twenty-five participants with TBI were
recruited. Median time post-injury was 216 days
(M = 392.64, SD = 537.19, range = 34 to 2497).
Considering PTA duration and worst GCS, parti-
cipants had sustained complicated mild (all of
whom had evidence of intracranial abnormality
on neuroimaging) to very severe injuries (PTA 1
day, n = 2; PTA >1 to <7 days, n = 2; PTA = 7 to
28 days, n = 10; PTA >28 days, n = 11). Worst
GCS scores ranged from 3 to 15 (GSC 1315, n =
6; GCS 912, n = 2; GCS 38, n = 16; not recorded,
n = 1). This study included two individuals with
complicated mild TBI, as research has demon-
strated poorer outcomes following complicated
compared to uncomplicated mild TBI (Iverson
et al., 2012; Lange, Iverson, & Franzen, 2009). All
participants were out of PTA when assessed. The
most common mechanism of injury was motor
vehicle accidents (n = 15), followed by falls (n =
4), cycling accidents (n = 3), pedestrians versus
motor vehicles (n = 2), and sporting injuries (n = 1).
Twenty-five age-, gender- and education-
matched healthy control participants were
recruited from the general community, and exclu-
sion criteria (b) to (f) above applied. There were no
significant differences between individuals with
TBI and controls for background demographics
(Table 1).
Measures
National Adult Reading Test (NART)
The NART (Nelson, 1982) comprises 50 words
with irregular phonemic pronunciations and was
TABLE 1
TBI Control
(n = 25) (n = 25) p
18 18 1.00
38.44 (16.25) 36.36 (14.83) .64
13.20 (2.72) 13.38 (2.51) .81
84 92 .39
4.48 (1.13) 4.33 (1.03) .63
106.62 (8.25) 107.22 (5.93) .77
7.33 (4.35) [315]
39.52 (38.34) [1142]
 PROCESSING SPEED AND ATTENTIONAL CONTROL AFTER TBI
used as a measure of pre-morbid intellectual func-
tioning. The number of errors and predicted Full
Scale IQ (FSIQ) were recorded.
Digit Span
Digit Span (DS) from the Wechsler Adult
Intelligence Scale 4th Edition (WAIS-IV;
Wechsler, 2008) was used to assess auditory work-
ing memory. Raw scores for digit span forwards
(DSF), backwards (DSB), sequencing (DSS), and
total (DST) were analyzed.
Trail Making Test
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
The TMT is a paper-and-pencil task consisting
of two parts: TMT-A and TMT-B (Reitan, 1958).
TMT-A requires individuals to connect sequen-
tially increasing numbers (measuring speed),
whereas TMT-B requires individuals to alternate
in ascending order between numbers and letters
(measuring speed and mental flexibility).
Completion time and number of errors for each
condition were recorded.
Hayling Test
The Hayling Test consists of two sets of 15
sentences, each with the last word missing
(Burgess & Shallice, 1997). Individuals must
replace the missing word with a logical response
(Hayling Initiation; measuring response speed) or
an illogical response (Hayling Inhibition; measur-
ing speed and response inhibition). Speed of com-
pletion and total error raw scores were analyzed.
Ruff 2 and 7 Selective Attention Test (2&7)
The 2&7 is a pencil-and-paper visual search and
cancellation task designed to measure selective
attention (Ruff & Allen, 1995). Participants can-
celled the digits 2 and 7 among either letters or
numbers, with the former being an automatic,
single-step retrieval condition and the latter requir-
ing controlled search employing working memory
abilities. Each of the 20 trials has a 15-s time limit.
Automatic speed (ASRS), controlled speed (CSRS)
and accuracy raw scores were calculated for both
conditions.
Selective Attention Task (SAT)
This computerized visual response contingency
paradigm is designed to assess selective attention
and the participants ability to inhibit automatic
responses (Ziino & Ponsford, 2006). It includes two
1027
conditions: Simple (SSAT) and Complex (CSAT)
Selective Attention Tasks. Participants responded
based on color for the SSAT (requiring automatic
processing), and color and content for the CSAT
(requiring controlled processing; see Willmott
et al., 2009; Ziino & Ponsford, 2006 for more
information). RT for correct responses, errors,
and missed responses were recorded for each con-
dition. The practice session of 20 trials was
repeated until an accuracy criterion of at least
70% was reached.
Symbol Digit Modalities Test (SDMT)
The written SDMT is a measure of processing
speed requiring the participant to examine a
sequence of symbols, then search a key for each
symbol and substitute the corresponding number
(Smith, 1991). Number of items correctly com-
pleted and errors within 90 s were recorded.
N-back
The n-back is a visual sequential letter memory
task, which parametrically varies working memory
load by asking participants to decide whether a
target matches the stimuli presented 0, 1, or 2
screens earlier, with 0-back requiring automatic
processing and higher working memory loads
requiring greater controlled processing. (Cohen
et al., 1997; Perlstein et al., 2004). The n-back
task was developed on E-Prime and comprised
single consonants centrally presented for 500 ms.
Participants responded with a dominant-hand
index finger press of one button for targets (p =
.33) and another for non-targets. A missed
response was recorded after 3000 ms. Conditions
were run in blocks of 18 stimuli, with three blocks
for each condition. Participants received visual
instructions and practice of each task prior to test
trials. Thereafter, order of the conditions was ran-
domized. RT for correct responses, errors, and
missed responses were recorded for each condition.
Procedure
This study was approved by Epworth HealthCare
and Monash University Human Research Ethics
Committees, and all participants provided
informed consent. Worst GCS scores and PTA
duration determined by prospective monitoring
with the Westmead PTA Scale (Shores,
Marosszeky, Sandanam, & Batchelor, 1986) were
recorded. Neuroimaging results were obtained
from medical records. Participants completed a
 1028 DYMOWSKI ET AL.
demographics questionnaire, followed by the atten-
tion tasks. Order of task presentation was rando-
mized between participants.
Data analysis
RTs greater or less than 2 standard deviations
from the mean for each group were removed (3%
for n-back across three conditions and 5% for SAT
across two conditions), resulting in similar percen-
tage of outliers to previous research (Willmott
et al., 2009). Univariate outliers with z > 3.29 (p
< .001, two-tailed) were identified (1 each in TBI
TMT-A and TMT-B) and assigned a score 1 unit
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
larger than the next most extreme score
(Tabachnick & Fidell, 2007). t tests were used to
identify differences between groups for normally
distributed test variables. Error data for the
SDMT, TMT, Hayling Test, 2&7 accuracy data,
and errors and missed responses for the SAT and
n-back did not meet parametric assumptions and
were analyzed using Mann-Whitney U Tests.
Bonferroni adjustments were made for multiple
comparisons. Two-way mixed model analyses of
variance (ANOVAs) were conducted on 2&7 and
SAT (selective attention), n-back (working mem-
ory), TMT (mental flexibility), and Hayling (inhi-
bition) data to test the hypothesis that participants
with TBI would demonstrate a disproportionately
greater increase in RT than controls at higher load
levels requiring controlled processing. Between-
subjects factor was group (control, TBI), and the
within-subject factor was condition/load. On tasks
demonstrating a significant interaction effect, one-
way analyses of covariance (ANCOVAs) were con-
ducted to determine whether disproportionate
group differences in RT remained after controlling
for slowed processing speed (i.e. the automatic
processing condition of each task).
RESULTS
A small number of individuals with TBI were pre-
scribed anticonvulsants (20%) or narcotic analge-
sics (12%). Separate multivariate analyses of
variance (MANOVA) were conducted to assess
whether medication influenced performance on
cognitive assessments. Dependent variables were
cognitive measures that demonstrated differences
between the TBI and control groups. There was
no significant effect of group on cognition for
anticonvulsants, F(11, 13) = 1.42, p = .27, Wilks
Lambda = .45, or for narcotics, F(11, 13) = 0.37,
p = .95, Wilks Lambda = .76.
Speed of information processing and
strategic control
Using a Bonferroni adjusted level of .002, bivariate
comparisons revealed that TBI participants com-
pleted fewer items on the SDMT and cancelled
fewer targets on both the 2&7 automatic and con-
trolled conditions than control participants (Table 2).
The TBI group also demonstrated significantly slo-
wed information processing relative to the control
group on TMT-A and on all conditions of the
Hayling Test, SAT, and n-back. There were, how-
ever, no differences in working memory performance
between the TBI and control group on any DS
condition.
Mann-Whitney U Tests revealed no significant
differences between TBI and control groups on num-
ber of errors on SDMT, TMT-A, or TMT-B, or
accuracy on 2&7 automatic and controlled condi-
tions (p > .05). In contrast, TBI participants made
significantly more errors on Hayling Inhibition than
did control participants (M = 15.84, SD = 18.52; M
= 4.24, SD = 6.65, respectively), U = 158, z = 3.01,
p = .002. There were no significant differences
between groups on SSAT or CSAT number of errors
or misses (p > .05). On the n-back, there were no
significant group differences for number of errors on
any condition (p > .05), but TBI participants made
significantly more missed responses than controls on
the 0-back (M = 0.64, SD = 2.20; M = 0.00, SD =
0.00, respectively), U = 238, z = 2.58, p = .02, 1-
back (M = 1.36, SD = 2.56; M = 0.08, SD = 0.40,
respectively), U = 176, z = 3.45, p = .001, and 2-
back conditions (M = 3.16, SD = 7.26; M = 0.04, SD
= 0.20, respectively), U = 133, z = 4.22, p < .001.
The 2&7, SAT, n-back, TMT and Hayling Test
were analyzed with mixed model ANOVAs to inves-
tigate the influence of load on attentional and execu-
tive functioning performance between groups.
Finding a significant Group Condition interaction
in the hypothesized direction would suggest that the
performance of individuals with TBI is disproportion-
ally poorer than that of controls under conditions of
increased working memory or executive load
(Kinsella, 2008; Perlstein et al., 2004; Willmott et al.,
2009).
On the n-back task (Figure 1), the TBI group was
significantly slower than the control group, F(1, 48) =
26.36, p < .001, 2p = .35. There was a significant
linear effect across loads, F(2, 47) = 45.02, p < .001,
2p = .66, with step-wise slower performance with
increased working memory load. However, an inter-
action effect was not observed, F(2, 47) = 0.74, p = .48.
On the SAT (Figure 2), participants with TBI
performed significantly more slowly than controls,
F(1, 48) = 26.95, p < .001, 2p = .36, and
 PROCESSING SPEED AND ATTENTIONAL CONTROL AFTER TBI 1029

TABLE 2
Means, standard deviations, significance and effects sizes for TBI and control groups

TBI Control

Task Measure Mean (SD) Mean (SD) p Effect sizea

Digit Span DSF RS 10.92 (2.47) 10.68 (2.64) .74

DSB RS 9.04 (2.70) 8.16 (2.29) .22
DSS RS 9.00 (2.25) 8.32 (2.23) .29
DST RS 28.96 (6.10) 27.44 (5.33) .36
SDMT Number correct 41.96 (10.74) 55.36 (9.95) <.001 1.29
TMT TMT-A time (s) 40.56 (16.68) 26.32 (9.33) .001 1.05
TMT-B time (s) 85.36 (39.47) 63.76 (21.71) .02
Hayling Initiation Time (s) 16.36 (14.67) 5.84 (5.01) .002 0.96
Inhibition Time (s) 46.20 (45.46) 9.80 (9.58) .001 1.11
2&7 ASRS 106.00 (33.56) 153.48 (25.25) <.001 1.60
CSRS 98.56 (27.08) 133.96 (19.21) <.001 1.51
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017

SAT SSAT RT (ms) 919.69 (224.26) 662.66 (99.47) <.001 1.48

CSAT RT (ms) 1663.58 (396.53) 1254.34 (199.29) <.001 1.30
N-back 0-back RT (ms) 842.00 (208.21) 598.07 (108.17) <.001 1.47
1-back RT (ms) 896.02 (187.19) 685.73 (137.11) <.001 1.28
2-back RT (ms) 1023.91 (213.87) 786.92 (161.15) <.001 1.25

Note. DSF = Digit Span Forward; RS = Raw Score; DSB = Digit Span Backward; DSS = Digit Span Sequencing; DST = Digit
Span Total; SDMT = Symbol Digit Modalities Test; TMT = Trail Making Test; ASRS = Automatic Speed Raw Score; CSRS =
Controlled Speed Raw Score; SAT = Selective Attention Task; SSAT = Simple Selective Attention Task; RT = Reaction Time; CSAT
= Complex Selective Attention Task.
a
Effect size = Cohens d.

participants performed more slowly on the com-

Figure 1. Mean reaction times (and standard error of the mean)
on the 0-, 1- and 2-back loads of the n-back working memory
task for TBI and control groups.
plex than the simple condition, F(1, 48) = 435.28,
p < .001, 2p = .90. There was also a significant
interaction effect, with participants with TBI show-
ing greater increases in RT from simple to complex
tasks compared to control participants, F(1, 48) =
5.65, p = .02, 2p = .11.
For the 2&7 (Figure 3), the TBI group cancelled
significantly fewer targets than did the control
group, F(1, 48) = 32.18, p < .001, 2p = .40, and
both groups cancelled significantly fewer targets
under the controlled than the automatic condition,
F(1, 48) = 45.90, p < .001, 2p = .49. There was a
significant Group Condition interaction, F(1, 48)

Figure 3. Mean number of items completed (and standard error

Figure 2. Mean reaction times (and standard error of the mean) of the mean) on the Automatic and Controlled conditions of the
on the Simple and Complex conditions of the Computerized Ruff 2&7 Selective Attention Test (2&7) for the TBI and control
Selective Attention Task (SAT) for the TBI and control groups. groups.
 1030 DYMOWSKI ET AL.
Figure 4. Mean response time (and standard error of the mean)
on the Initiation and Inhibition conditions of the Hayling Test
for the TBI and control groups.
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
= 9.22, p = .004, 2p = .16, with controls showing
greater decrement in speed on controlled relative to
automatic processing than the TBI group.
On the Hayling Test (Figure 4), the TBI group
performed more slowly than controls, F(1, 48) =
17.21, p < .001, 2p = .26, and both groups per-
formed more slowly on the Inhibition condition
than the Initiation condition, F(1, 48) = 17.85, p
< .001, 2p = .27. There was an interaction effect
whereby patients with TBI showed a greater
increase in speed on the Inhibition when compared
to the Initiation task relative to controls, F(1, 48) =
10.46, p = .002, 2p = .18.
For the TMT (Figure 5), the TBI group were
slower than the control group, F(1, 48) = 8.95, p =
.004, 2p = .16, and both groups took longer to
complete TMT-B than TMT-A, F(1, 48) = 140.65,
p < .001, 2p = .75. However, there was no sig-
nificant interaction, F(1, 48) = 1.13, p = .29.
Association of speed with group differences
We were interested to investigate the contribution
of slowed information processing to the
Figure 5. Mean response time (and standard error of the mean)
on Trail Making Test (TMT) A and B conditions for the TBI
and control groups.
differential impairments of RT in TBI and control
participants on the complex tasks, namely the
SAT, 2&7, and Hayling tests. ANCOVAs were
conducted to covary for speed of processing on
complex tasks (Felmingham et al., 2004). To con-
trol for speed of processing in selective attention,
we used the 2&7 automatic condition (ASRS) as a
covariate for the controlled condition (CSRS),
and the SAT simple condition (SSAT) as a cov-
ariate for the complex condition (CSAT). Hayling
Initiation was used as a covariate for Hayling
Inhibition when investigating the effect of speed
on response inhibition. The TMT and n-back
were not further analyzed, as no interaction
effects were found between group and task
complexity.
On the SAT, after adjusting for the simple con-
dition (SSAT), there was no significant difference
between the two groups on the complex condition
(CSAT), F(1, 47) = 0.99, p = .32. There was a
strong relationship between the simple and com-
plex condition times, with an 2p of .51.
On the 2&7, covarying for speed of processing
(ASRS) removed the significant group difference
on the controlled condition (CSRS), F(1, 47) =
0.29, p = .60. There was a strong relationship
between 2&7 automatic and controlled condition
scores, as indicated by an 2p of .79.
On the Hayling Test, a significant group effect
on the Inhibition Task remained after controlling
for speed on the Initiation Task, F(1, 47) = 4.72, p
= .04, 2p = .09. Again, there was a strong relation-
ship between the Hayling subtests, with an 2p
of .31.
These findings indicate that speed explained the
group differences in the complex conditions of the
selective attention tasks, but a specific deficit in
response inhibition remained after controlling for
speed.
DISCUSSION
The aim of this study was to investigate factors
contributing to attentional impairment after TBI,
particularly the influence of deficits in speed of
thinking and strategic control on a range of atten-
tional tasks requiring controlled processing, includ-
ing working memory, selective attention, mental
flexibility, and response inhibition. While reduced
information processing speed was pervasive across
tasks, impaired strategic control of attention was
only evident on a response inhibition task (i.e. the
Hayling Test) after accounting for slowed proces-
sing speed.
 PROCESSING SPEED AND ATTENTIONAL CONTROL AFTER TBI
Participants with TBI demonstrated reduced
speed of information processing on the SDMT,
TMT-A, and all conditions of the Hayling Test,
2&7, SAT, and n-back tasks relative to healthy
controls. This finding corroborates previous
research, consistently identifying reduced speed of
information processing as a common sequela post-
TBI (Felmingham et al., 2004; Frencham et al.,
2005; Mathias & Wheaton, 2007; Ponsford &
Kinsella, 1992; Willmott et al., 2009).
Working memory was assessed in the current
study with DS and the n-back tasks. There were
no significant differences between the groups on
any DS measure (DSF, DSB, DSS, or DST),
which is indicative of intact functioning of the
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
phonological loop (Serino et al., 2006). On the
n-back task, a measure of the central executive of
working memory, participants with TBI had
slower RTs than controls, and both groups were
slower with increased working memory load.
However, the TBI group was not disproportional-
ity slower at increased loads than was the control
group, which is contrary to the hypothesis, but in
line with previous research (Asloun et al., 2008;
Perlstein et al., 2004). Furthermore, participants
with TBI showed more missed responses than did
controls on all conditions of the n-back. These
results are in contrast to Perlstein et al. (2004),
who found that groups did not differ in the number
of missed responses on the n-back, but, rather, that
participants with moderate to severe TBI made
more errors at higher working memory loads (2-
and 3-back) relative to controls. Potentially,
Perlstein et al. (2004) elicited more errors through
inclusion of non-target repeats as foils. The
increase in missed responses in the current study
may also reflect the shorter stimulus onset asyn-
chrony of 3000 ms compared to 4000 ms used by
Perlstein et al. (2004), resulting in insufficient time
for participants with TBI to process and respond
appropriately. Alternatively, missed responses may
indicate greater inattention, distractibility, or dis-
engagement from all n-back conditions in partici-
pants with TBI relative to controls. Thus, these
results demonstrate impaired speed but not deficits
in strategic control on working memory tasks
after TBI.
The hypothesized disproportionate increase in
RTs with increased task complexity for partici-
pants with TBI relative to controls was observed
on two of five tasks. Specifically, participants with
TBI demonstrated disproportionately slower RT
on the SAT with additional working memory
load (CSAT: color and letter/number) relative to
the simple task (SSAT: color). Additional
1031
covariance analyses revealed that the difference
between groups on the CSAT was accounted for
by reduced speed of information processing
(SSAT). This is in line with previous research
using structural equation modelling, which showed
the main contribution to group differences on the
CSAT was slowed speed of thinking, with only a
weak contribution of impaired working memory
(Willmott et al., 2009). Yet in the current study
participants with TBI did not make more errors or
missed responses with increased complexity, as has
been previously demonstrated (Willmott et al.,
2009; Ziino & Ponsford, 2006). When compared
to Willmott et al. (2009), this sample of partici-
pants with TBI made fewer errors (M = 5.38, M =
2.52, respectively) and missed responses (M = 1.38,
M = 0.40, respectively) on the CSAT. Thus, the
accuracy profile of the current TBI group was
similar to that of the healthy control group, sug-
gesting that speed but not accuracy was impaired
with increased complexity.
On the 2&7 the control participants demon-
strated a greater reduction in speed with increased
complexity from the automatic to controlled pro-
cessing condition relative to TBI participants.
While this finding is opposed to the hypothesized
direction of change, it is commensurate with pre-
vious research (Willmott et al., 2009). Willmott
et al. (2009) suggested that such a finding poten-
tially indicates that individuals with TBI require
controlled processing for both the automatic and
controlled conditions, and thus are unable to ben-
efit from the reduced cognitive demands of the
automatic condition. Indeed, neuroimaging of par-
ticipants with DAI after TBI engaged in tasks
requiring strategic control has demonstrated
greater neural recruitment of regions of the pre-
frontal cortex and posterior cortices at earlier
stages of difficulty compared to controls (Turner
& Levine, 2008; Turner, McIntosh, & Levine,
2011). Alternatively, it has been suggested that
the findings may be indicative of greater motiva-
tion or mental effort in the group with TBI, or that
the task was not sufficiently complex to tap strate-
gic control (Asloun et al., 2008; Vallat-Azouvi,
Weber, Legrand, & Azouvi, 2007).
The Hayling Test results demonstrated that par-
ticipants with TBI were disproportionally slower
with increased complexity compared to controls
on Hayling Inhibition, which remained after cov-
arying for speed (Hayling Initiation). These results
suggest that the differences between groups were
due to impairment of a specific attentional compo-
nent rather than to deficits in speed of processing.
Participants with TBI also made more errors than
 1032 DYMOWSKI ET AL.
did controls, which is further indicative of
impaired response inhibition after brain injury.
The Hayling Test requires participants to inhibit
an automatic response before generating a novel
response, necessitating both response inhibition
and idea generation skills. The Stroop Test simi-
larly requires participants to inhibit an automatic
response in favor of a less familiar response, but it
has generally not been found to demonstrate a
disproportionate distraction effect (Felmingham
et al., 2004; Ponsford & Kinsella, 1992; Ros
et al., 2004). Potentially the addition of idea gen-
eration makes the Hayling more sensitive than the
Stroop Test to deficits in response inhibition after
TBI. Furthermore, the Hayling Test tapped two
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
executive processes simultaneouslynamely, inhi-
bition and idea generation (Strauss, Sherman, &
Spreen, 2006)which possibly made this task
more challenging and better able to tap everyday
functioning difficulties in individuals with TBI
than other tasks requiring controlled processing
used in this study (Spikman, Deelman, & van
Zomeren, 2000). Alternatively, examination of the
error profiles suggests that controls were more
likely than individuals with TBI to adopt a strategy
(i.e., naming objects in the room or listing items
from a predefined category) to avoid providing an
automatic response. This suggests that individuals
with TBI may demonstrate impairment in either
the generation or the implementation of strategies
(Shallice, 2002). Hence, the Hayling Test possibly
captures strategy production or implementation
better than do the other tasks assessing strategic
control employed in this study.
The results of the TMT indicate that participants
with TBI were slower than controls, but did not
perform differentially across conditions. Explicitly,
participants with TBI were not more disadvantaged
by increased working memory load and flexibility
requirements than healthy control participants.
Previous research findings on group differences in
attention flexibility (TMT-B) after controlling for
speed (TMT-A) are mixed. The current results are
in line with Felmingham et al. (2004), who found
that group differences on TMT-B were removed
after covarying for TMT-A, but contrast with find-
ings of Ros et al. (2004), who reported that deficits
in strategic control remained after controlling for
speed. The speed of completion of TMT-B in this
study (M = 85.36) was similar to Felmingham et al.
(2004) (M = 79.0 for focal pathology; M = 109.2 for
diffuse pathology), which were quicker than the
results of Ros et al. (2004) (M = 162.03). The
differences in speed of completion on TMT-B for
participants with TBI across studies likely explains
the differential findings.
GENERAL DISCUSSION, LIMITATIONS, AND
FUTURE DIRECTIONS
The SAT and Hayling Test demonstrated the pre-
dicted Group Condition interaction, with parti-
cipants with TBI more disadvantaged with
increased complexity than were controls. On the
2&7 participants with TBI appeared unable to
benefit from the minimal processing demands of
the automatic condition. Alternatively, the 2&7,
TMT and n-back may not have been sufficiently
challenging or complex to recruit strategic control
of attention. More challenging tasks tapping func-
tional attentional difficulties experienced in every-
day life by participants with TBI may be required
to elicit deficits in strategic control.
The present data strongly support the pervasive
influence of reduced speed of information proces-
sing on attentional performance after TBI.
Specifically, speed accounted for group differences
on the complex conditions of the selective attention
tasks, including the Ruff 2&7 CSRS and the
CSAT. Yet the findings also indicate that strategic
control influences performance beyond the influ-
ence of speed of thinking, whereby group differ-
ences on the Hayling Test Inhibition task remained
after controlling for speed. There were no differ-
ential group differences in RT on working memory
and mental flexibility tasks. The relative influence
of speed and strategic control on attention may be
task-dependent or attention-type-specific.
We acknowledge that the current results included
a small sample of participants who were treated at
Epworth HealthCare. Thus, the results may not be
representative of the broader population with TBI.
Furthermore, the small sample size may have
resulted in reduced power to detect deficits in strate-
gic control beyond the influence of speed. Lack of
significant findings in the covariance analyses should
therefore be interpreted cautiously. This study
included a heterogeneous sample, whereby the
majority of individuals had sustained a severe TBI,
and two individuals had sustained a complicated
mild TBI (both with intracranial abnormality evi-
dent on neuroimaging). While individuals with com-
plicated mild TBI typically have poorer outcomes
than those with uncomplicated TBI (Iverson et al.,
2012; Lange et al., 2009), it is possible that the
nature of the cognitive impairments differs across
this sample, and deficits in strategic control may be
limited to the most severely injured individuals.
Despite attempts to fractionate strategic control
into subcomponents both theoretically (Baddeley,
2002; Shallice, 2002) and via factor analytic tech-
niques (Mirsky, Anthony, Duncan, Ahearn, &
Kellam, 1991; Ros et al., 2004), the mechanisms
 PROCESSING SPEED AND ATTENTIONAL CONTROL AFTER TBI
underlying attentional control remain debated.
Hence, it is currently unclear whether various com-
ponent processes of strategic control are differen-
tially affected after TBI. Functional magnetic
resonance imaging (fMRI) and other imaging tech-
nologies provide advanced techniques for delineat-
ing component processes of strategic control as
well as exploring underlying neuropathology and
neurochemical alterations responsible for deficits
in complex attention and speed after TBI.
CONCLUSIONS
The current study strongly supports the vast array
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
of research demonstrating slowed information pro-
cessing after TBI. There was also evidence of
impaired strategic control of attention on response
inhibition tasks after controlling for speed. Hence,
various component processes of attention or atten-
tional tasks are differentially influenced by speed
and strategic control after TBI. Further investiga-
tion and replication of differential effects on tasks
requiring multiple executive processes and strategy
production and implementation is warranted in
order to guide rehabilitative efforts.
REFERENCES
Asloun, S., Soury, S., Couillet, J., Giroire, J.-M., Joseph,
P.-A.,Mazaux, J.-M., & Azouvi, P. (2008).
Interactions between divided attention and working-
memory load in patients with severe traumatic brain
injury. Journal of Clinical and Experimental
Neuropsychology, 30(4), 481490. doi:10.1080/
13803390701550144
Azouvi, P., Jokic, C., van der Linden, M., Marlier, N.,
& Bussel, B. (1996). Working memory and supervi-
sory control after severe closed-head injury. A study
of dual task performance and random generation.
Journal of Clinical and Experimental
Neuropsychology, 18(3), 317337. doi:10.1080/026990
50902970711
Baddeley, A. (1986). Working memory. New York:
Oxford University Press.
Baddeley, A. (2000). The episodic buffer: A new compo-
nent of working memory? Trends in Cognitive
Sciences, 4(11), 417423.
Baddeley, A. (2002). Fractionating the central executive.
In D. T. Stuss & R. T. Knight (Eds.), Principles of
frontal lobe function (pp. 246260). New York:
Oxford University Press.
Bigler, E. D. (2001). Quantitative magnetic resonance
imaging in traumatic brain injury. Journal of Head
Trauma Rehabilitation, 16(2), 117134.
Burgess, P., & Shallice, T. (1997). The Hayling and
Brixton Tests. Bury St Edmunds: Thames Valley
Test Company.
1033
Cohen, J. D., Perlstein, W. M., Braver, T. S., Nystrom,
L. E., Noll, D. C., Jonides, J., & Smith, E. E. (1997).
Temporal dynamics of brain activation during a
working memory task. Nature, 386, 604608.
doi:10.1038/386604a0
Daniel, A. (1983). Power, privilege and prestige:
Occupations in Australia. Melbourne: Longman
Cheshire.
Felmingham, K. L., Baguley, I. J., & Green, A. M.
(2004). Effects of diffuse axonal injury on speed of
information processing following severe traumatic
brain injury. Neuropsychology, 18(3), 564571.
doi:10.1037/0894-4105.18.3.564
Frencham, K. A. R., Fox, A. M., & Maybery, M. T.
(2005). Neuropsychological studies of mild traumatic
brain injury: A meta-analytic review of research since
1995. Journal of Clinical and Experimental
Neuropsychology, 27, 334351. doi:10.1080/13803390
490520328
Iverson, G. L., Lange, R. T., Wljas, M.,
Liimatainen, S., Dastidar, P., Hartikainen,
K. M.,. . . hman, J. (2012). Outcome from com-
plicated versus uncomplicated mild traumatic brain
injury. Rehabilitation Research and Practice.
doi:10.1155/2012/415740
James, W. (1890). The principles of psychology. New
York: Dover.
Kinsella, G. J. (2008). Traumatic brain injury and pro-
cessing speed. In J. DeLuca & J. H. Kalmar (Eds.),
Information processing speed in clinical populations
(pp. 173194). New York: Taylor & Francis.
Lange, R. T., Iverson, G. L., & Franzen, M. D. (2009).
Neuropsychological functioning following compli-
cated vs. uncomplicated mild traumatic brain injury.
Brain Injury, 23(2), 8391. doi:10.1080/
02699050802635281
Levin, H. S., Williams, D. H., Eisenberg, H. M., High,
W. M. J., & Guinto, F. C. (1992). Serial magnetic
resonance imaging and neurobehavioural findings
after mild to moderate closed head injury. Journal
of Neurology, Neurosurgery, and Psychiatry, 55,
255262.
Lewis, F. D., & Horn, G. J. (2013). Traumatic brain
injury: Analysis of functional deficits and posthospi-
tal rehabilitation outcomes. Journal of Special
Operations Medicine, 13(3), 5661.
MacDonald, A. W., Cohen, J. D., Stenger, V. A., &
Carter, C. S. (2000). Dissociating the role of the
dorsolateral prefrontal and anterior cingulate cortex
in cognitive control. Science, 288(5472), 18351838.
doi:10.1126/science.288.5472.1835
Markianos, M., Seretis, A., Kotsou, A., Baltas, I., &
Sacharogiannis, H. (1992). CSF neurotransmitter
metabolites and short-term outcome of patients in
coma after head injury. Acta Neurologica
Scandinavica, 86, 190193. doi:10.1111/j.1600-
0404.1992.tb05064.x
Markianos, M., Seretis, A., Kotsou, A., &
Christopoulos, M. (1996). CSF neurotransmitter
metabolites in comatose head injury patients during
changes in their clinical state. Acta Neurochirurgica,
138, 5759. doi:10.1007/BF01411725
Mathias, J. L., & Wheaton, P. (2007). Changes in atten-
tion and information-processing speed following
severe traumatic brain injury: A meta-analytic
 1034 DYMOWSKI ET AL.
review. Neuropsychology, 21(2), 212223.
doi:10.1037/0894-4105.21.2.212
McDowell, S., Whyte, J., & DEsposito, M. (1997).
Working memory impairments in traumatic brain
injury: Evidence from a dual-task paradigm.
Neuropsychologia, 35(10), 13411353. doi:10.1016/
S0028-3932(97)00082-1
Mirsky, A. F., Anthony, B. J., Duncan, C. C., Ahearn,
M. B., & Kellam, S. G. (1991). Analysis of the ele-
ments of attention: A neuropsychological approach.
Neuropsychological Review, 2(2), 109145.
doi:10.1007/BF01109051
Nelson, H. E. (1982). National Adult Reading Test
(NART): Test manual. Windsor: NFER Nelson.
Norman, D. A., & Shallice, T. (1980). Attention to
action: Willed and automatic control of behavior.
Center for Human Information Processing Technical
Report No. 99. San Diego, CA: University of
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
California.
Olver, J. H., Ponsford, J. L., & Curran, C. A. (1996).
Outcome following traumatic brain injury: A com-
parison between 2 and 5 years after injury. Brain
Injury, 10(11), 841848. doi:10.1080/02699059
6123945
Park, N. W., Moscovitch, M., & Robertson, I. H.
(1999). Divided attention impairments after trau-
matic brain injury. Neuropsychologia, 37, 11191133.
Perlstein, W. M., Cole, M. A., Demery, J. A.,
Seignourel, P. J., Dixit, N. K., Larson, M. J., &
Briggs, R. W. (2004). Parametric manipulation of
working memory load in traumatic brain injury:
Behavioral and neural correlates. Journal of the
International Neuropsychological Society, 10, 724
741. doi:10.10170S1355617704105110
Ponsford, J., & Kinsella, G. (1991). The use of a rating
scale of attentional behaviour. Neuropsychological
Rehabilitation, 1(4), 241257. doi:10.1080/09602019
108402257
Ponsford, J., & Kinsella, G. (1992). Attentional deficits
following closed-head injury. Journal of Clinical and
Experimental Neuropsychology, 14(5), 822838.
doi:10.1080/01688639208402865
Povlishock, J. T. (1992). Traumatically induced axonal
injury: Pathogenesis and pathobiological implica-
tions. Brain Pathology, 2(1), 112.
Raz, A. (2004). Anatomy of attentional networks. The
Anatomical Record (Part B: New Anatomy), 281B,
2136. doi:10.1002/ar.b.20035
Reitan, R. M. (1958). Validity of the Trail Making Test
as an indicator of organic brain damage. Perceptual
and Motor Skills, 8, 271276. doi:10.2466/
PMS.8.7.271-276
Ros, M., Periez, J. A., & Muoz-Cspedes, J. M.
(2004). Attentional control and slowness of informa-
tion processing after severe traumatic brain injury.
Brain Injury, 18(3), 257272. doi:10.1080/
02699050310001617442
Ruff, R. M., & Allen, C. C. (1995). Ruff 2 & 7 Selective
Attention Test. Odessa, FL: Psychological
Assessment Resources.
Schneider, W., & Shiffrin, R. M. (1977). Controlled and
automatic human information processing: I.
Detection, search and attention. Psychological
Review, 84(1), 166. doi:10.1037/0033-295X.84.1.1
Serino, A., Ciaramelli, E., Di Santantonio, A., Malag,
S., Servadei, F., & Ldavas, E. (2006). Central execu-
tive system impairment in traumatic brain injury.
Brain Injury, 20(1), 2332. doi:10.1080/
02699050500309627
Shallice, T. (1982). Specific impairments of planning.
Philosophical Transactions of the Royal Society B:
Biological Sciences, 298, 199209. doi:10.1098/
rstb.1982.0082
Shallice, T. (2002). Fractionation of the supervisory sys-
tem. In D. T. Stuss & R. T. Knight (Eds.), Principles
of frontal lobe function (pp. 261277). New York:
Oxford University Press.
Shores, E. A., Marosszeky, J. E., Sandanam, J., &
Batchelor, J. (1986). Preliminary validation of a
scale for measuring the duration of post-traumatic
amnesia. Medical Journal of Australia, 144,
569572.
Smith, A. (1991). Symbol digit modalities test. Los
Angeles, CA: Western Psychological Services.
Spikman, J. M., Deelman, B. G., & van Zomeren, A. H.
(2000). Executive functioning, attention and frontal
lesions in patients with chronic CHI. Journal of
Clinical and Experimental Neuropsychology, 22(3),
325338. doi:10.1076/1380-3395(200006)22:3;1-V;
FT325
Spikman, J. M., van Zomeren, A., & Deelman, B. G.
(1996). Deficits of attention after closed-head injury:
Sloweness only? Journal of Clinical and Experimental
Neuropsychology, 18(5), 755767. doi:10.1080/
01688639608408298
Spitz, G., Bigler, E. D., Abildskov, T., Maller, J. J.,
OSullivan, R., & Ponsford, J. L. (2013). Regional
cortical volume and cognitive functioning following
traumatic brain injury. Brain and Cognition, 83,
3444. doi:10.1016/j.bandc.2013.06.007
Strauss, E., Sherman, E. M. S., & Spreen, O. (2006). A
compendium of neuropsychological tests:
Administration, norms, and commentary (3rd ed.).
New York: Oxford University Press.
Styles, E. A. (2006). The psychology of attention (2nd
ed.). New York: Psychology Press.
Tabachnick, B. G., & Fidell, L. S. (2007). Using multi-
variate statistics (5th ed.). Boston, MA: Pearson
Education.
Turner, G. R., & Levine, B. (2008). Augmented neural
activity during executive control processing following
diffuse axonal injury. Neurology, 71, 812818.
doi:10.1212/01.wnl.0000325640.18235.1c
Turner, G. R., McIntosh, A. R., & Levine, B. (2011).
Prefrontal compensatory engagement in TBI is due to
altered functional engagement of existing networks
and not functional reorganization. Frontiers in
Systems Neuroscience, 5(9). doi:10.3389/
fnsys.2011.00009
Vallat-Azouvi, C., Weber, T., Legrand, L., & Azouvi, P.
(2007). Working memory after severe traumatic
brain injury. Journal of the International
Neuropsychological Society, 13, 770780.
doi:10.1017/0S1355617707070993
Veltman, J. C., Brouwer, W. H., van Zomeren, A. H., &
van Wolffelaar, P. C. (1996). Central executive
aspects of attention in subacute severe and very
severe closed head injury patients: Planning,
 PROCESSING SPEED AND ATTENTIONAL CONTROL AFTER TBI
inhibition, flexibility, and divided attention.
Neuropsychology, 10(3), 357367. doi:10.1037/0894-
4105.10.3.357
Wechsler, D. (2008). Wechsler Adult Intelligence Scale
Fourth Edition Australian and New Zealand Language
Adapted Edition: Sydney: Pearson Education.
Whyte, J., Ponsford, J., Watanabe, T., & Hart, T.
(2010). Traumatic brain injury. In W. R.
Frontera (Ed.), Delisas physical medicine and
rehabilitation: Principles and practice (5th ed., pp.
Downloaded by [Universitat Oberta de Catalunya] at 04:43 30 October 2017
1035
575623). Philadelphia, PA: Wolters Kluwer,
Lippincott Williams & Wilkins.
Willmott, C., Ponsford, J., Hocking, C., & Schnberger,
M. (2009). Factors contributing to attentional impair-
ments after traumatic brain injury. Neuropsychology,
23(4), 424432. doi:10.1037/a0015058
Ziino, C., & Ponsford, J. (2006). Selective attention def-
icits and subjective fatigue following traumatic brain
injury. Neuropsychology, 20(3), 383390. doi:10.1037/
0894-4105.20.3.383