Professional Documents
Culture Documents
Differential Neural Activation When Voluntarily Regulating Emotions
Differential Neural Activation When Voluntarily Regulating Emotions
Differential Neural Activation When Voluntarily Regulating Emotions
To cite this article: Michael N. Dretsch, Thomas A. Daniel, Adam M. Goodman, Jeffrey S. Katz,
Thomas Denney, Gopikrishna Deshpande & Jennifer L. Robinson (2017): Differential neural
activation when voluntarily regulating emotions in service members with chronic mild traumatic
brain injury, Applied Neuropsychology: Adult, DOI: 10.1080/23279095.2017.1362406
Article views: 7
Download by: [Australian Catholic University] Date: 23 September 2017, At: 04:54
APPLIED NEUROPSYCHOLOGY: ADULT
https://doi.org/10.1080/23279095.2017.1362406
none defined
ABSTRACT KEYWORDS
The objective of this study was to characterize the functional activation of the neural correlates of Attentional bias; emotional
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
voluntary regulation of emotion in soldiers both with and without chronic mild traumatic brain regulation; FMRI; mild
injury (mTBI). Using functional magnetic resonance imaging (fMRI) and a battery of cognitive and traumatic brain injury;
psychological health measures, we assessed differences between active-duty U.S. soldiers with military; TBI
chronic mTBI (n ¼ 37) and without (Controls, n ¼ 35). Participants were instructed to maintain
(passively view), enhance, and suppress emotions associated with negative and neutral visual
stimuli. The mTBI group showed significantly greater clinical symptoms, but only a mild decrement
in attention. Group contrasts, while controlling for posttraumatic stress disorder (PTSD) symptoms,
revealed a differential neural activation pattern compared to controls, but only during the enhance
condition. Specifically, the mTBI group showed greater activation in the precentral gyrus,
postcentral gyrus, inferior parietal lobe, insula, and superior temporal gyrus. Finally, the effect of
PTSD symptoms during the enhance condition was associated with accentuated activation of the
frontal and limbic regions implicated in both emotion regulation and PTSD. Hyperactivation of
neural regions in the mTBI group during the enhance condition may reflect vigilance towards
negative contextual stimuli and/or poor strategy that might result in suboptimal allocation of
resources to regulate emotions.
CONTACT Michael Dretsch dretschphd@gmail.com Human Dimension Division, HQ, TRADOC, 950 Jefferson Ave, Fort Eustis, VA 23604, USA.
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/hapn.
© 2017 Taylor & Francis Group, LLC
2 M. N. DRETSCH ET AL.
Liemburg, Aleman, Spikman, & van der Naalt, 2016; validated stimuli. Furthermore, many service members
van der Horn, Liemburg, Scheenen, et al., 2016). As experience psychological trauma, which contributes to
such, it is important to assess the neural correlates of the high levels of comorbidity. During combat, being
voluntary emotional regulation in this population. exposed to life-threatening events such as exposure to
Regulation of emotion is a complex process in that it blast from explosive devices, which results in mTBI,
implicates neuropsychological functions from multiple can be psychologically more traumatic than sustaining
cortical and subcortical regions and networks (for an injury from a noncombat event (e.g., sporting event,
reviews see Buhle et al., 2014; Kohn et al., 2014; Ochsner slipping in the shower). Because PTSD is linked to
& Gross, 2014). Furthermore, emotional regulation is an emotional dysregulation, and there is a high comorbid-
umbrella term that can refer to any number of processes ity of PTSD in service members with mTBI, a minimal
but generally is used to indicate shaping one’s emotion amount is known concerning the neural physiology
(Gross, 2014). In the laboratory, investigation of underlying mTBI in relation to emotional regulation.
emotional regulation requires systematic and focused Overall, emotional regulation following mTBI in mili-
studies for teasing apart the various underlying neural tary populations is understudied in terms of using
mechanism. Although the literature is replete with papers objective measures for emotion research.
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
on emotional regulation, only a few of the papers report The current study was aimed at exploring if combat-
using procedures that require voluntary emotional related mTBI is associated with emotional dysregulation
responses (Fitzgerald et al., 2016; Ochsner & Gross, when controlling for PTSD symptoms. This aim was
2005; Urry et al., 2006). In these studies, subjects are accomplished by comparing neural activation in prior
instructed to cognitively change their emotions by mod- deployed Army soldiers with and without mTBI during
ifying their thoughts. By doing so, subjects attempt to vol- an emotional regulation task adapted for functional
untarily perform a cognitive reappraisal of the situation. magnetic resonance imaging (fMRI; Urry et al., 2006),
Although, most frequently, subjects are asked to decrease and modified to include military-relevant, visual combat
negative emotions, cognitive reappraisal can be used to stimuli. The findings have important implications to the
increase or decrease both negative and positive emotions understanding of the underlying neurocircuitry
(Ochsner et al., 2004). Previous studies have shown that associated with emotional regulation in both healthy and
the medial (including the ventromedial PFC) and dorsal combat-injured military service members and veterans.
prefrontal cortex (including the ventral and lateral
regions), and cingulate gyrus are involved in voluntarily
reducing emotions (Ochsner & Gross, 2005; Phan et al., Method
2005; Urry et al., 2006).
Recruitment
An important and relevant emotional regulation
process involved in healthy functioning is the Active-duty male and female soldiers between the ages
suppression of negative emotions, or the attenuation of 18 and 50 years were recruited from Fort Rucker,
of the positive emotions, to mitigate emotions elicited AL and Fort Benning, GA to voluntarily participate in
by contextual cues associated with relevant psychologi- the current study. The participant sample (N ¼ 72)
cally traumatic events. As such, cognitive and behavioral was mostly male (90.3%) with a mean age of 31.5 years
therapies are aimed at improving one’s ability to old (SD ¼ 6.0). Ethno-racially, the sample was mostly
regulate emotions in response to trauma-related cues. Caucasian (65.3%) followed by African-American
There have been efforts to understand the underpinning (18.1%), Hispanic/Latino (12.5%), Asian (2.8%), and
neural functions of compromised emotional regulation Native American (1.4%). The majority of participants
in the laboratory (Cisler et al., 2016; Johnstone, van were right-handed (91.7%). Recruitment consisted of
Reekum, Urry, Kalin, & Davidson, 2007; Urry et al., advertisements distributed and posted at local facilities
2006; van der Horn, Liemburg, Aleman, et al., 2016). including the TBI Clinic. Soldiers receiving a
However, many studies only focus on populations comprehensive evaluation and/or treatment at the
suffering from mood and stressor/trauma-related instillation TBI clinic were referred to the study by the
disorders (e.g., depression, PTSD), fail to use an treating clinician and provided contact information to
appropriate control group, or do not use tasks that voluntarily pursue participation. Soldiers provided
require voluntary effort to regulate emotions. For written informed consent prior to their enrollment in
military service members, contextual cues can elicit the study. Data were collected at the Auburn University
altered emotional responses (e.g., Goodman, Katz, & MRI Research Center.
Dretsch, 2016). Most efforts to explore emotional regu- The study was carried out in accordance with the
lation in military populations fail to use salient and latest version of the Declaration of Helsinki and the
APPLIED NEUROPSYCHOLOGY: ADULT 3
protocol and procedures were approved by Auburn image on the screen in front of them, they would look
University’s Institutional Review Board (IRB) and the at it, and then they would receive one of three instruc-
IRB at Headquarters U.S. Army Medical Research and tions and regulate their emotions accordingly. If they
Materiel Command (HQ USAMRMC IRB). saw the word “ENHANCE,” they were told to attempt
to amplify the intensity of their negative feelings toward
the picture. They could do so by imagining themselves
Participants
or a loved one in the same situation or by imagining
Exclusion criteria were as follows: (a) pre-existing, non- a more extreme outcome than the one presented. If they
TBI neurologic disorder, psychotic disorder (e.g., saw the word “SUPPRESS,” they were told to reduce the
schizophrenia); (b) bipolar disorder; (c) history of mod- intensity of their negative feelings toward the picture.
erate-to-severe TBI; and (d) contraindications to MRI. They could do so by imagining the situation as fake
All participants were screened for MRI contraindica- or unreal, or by imaging the presented situation had
tions prior to recruitment and again the same day of an outcome better than the one within the context of
data collection. All participants were required to have the picture. If they saw the word “MAINTAIN,” they
a history of deployment(s) to a combat environment. were instructed to continue attending to the picture
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
Eligibility for the mTBI group included: (a) referred without positively or negatively changing their feelings.
by a neuropsychologist from the military TBI Clinic; After regulating their emotions according to the
(b) a documented combat-related mTBI that occurred instruction, participants were then asked to report
within the last five years; and (c) presenting symptoms how successful they were in regulating their emotions.
associated with a history of medically documented Using an MR-compatible 4-button box (Current
mTBI(s). Controls were healthy active-duty soldiers Design, Philadelphia) response device, participants
with no medical history of mTBI in the last five years responded by indicating whether they were “Not suc-
and no history of mild-to-moderate TBI. Soldiers’ cessful,” “Somewhat successful,” or “Very successful.”
electronic medical records were screened by the study If participants had no emotion to regulate, they were
physician to ensure they met the eligibility criteria. asked to choose a “No Emotion” option. After the
self-report period ended, an inter-trial interval (ITI)
with a fixation cue followed and the next trial began
Emotional regulation task
(See Figure 1 for a representative trial). Each trial lasted
Participants performed an emotional regulation task 17 s, with the fixation cue being presented for 1 s, the
modeled after the one implemented by Urry et al. view instruction (where participants passively viewed
(2006). Participants were told that they would see an the stimulus) lasting between 3–7 s, the regulation
Figure 1. A trial example of emotional regulation task for the maintain condition. Participants were instructed to maintain, enhance,
or suppress their emotional response to MAPS images that are military specific. After the regulation period ended participants
indicated how well they regulated their emotion using a button box.
4 M. N. DRETSCH ET AL.
instruction lasting between 5–8 s, and a 5 s time period 95% CI [4.62, 5.00]). Twenty-four neutral MAPS images
to self-report their regulation success. The duration of were selected on the basis of content normatively rated
the view and regulation instructions were counterba- as neither pleasant nor unpleasant (M ¼ 5.58, 95%
lanced to ensure that all trials were 17 s in duration. CI [4.94, 6.21]) with low arousal (M ¼ 2.63, 95%
An event-related design was used to present viewing CI [2.19, 3.08]). Independent samples t-tests, corrected
and regulation instructions while participants viewed for unequal variances when necessary, confirmed that
images during blood oxygen level-dependent (BOLD) the negative images were less pleasant, t(38.46) ¼ 8.00,
fMRI acquisition. There were 5 blocks of the emotional p < .05, and more arousing, t(73) ¼ 8.97, p < .05, than
regulation task. The first block was a practice block to the neutral images. Image presentation orders were
familiarize participants with the task and was conducted pseudorandomized in blocks of 4 image presentations,
outside of the scanner with 24 noncombat images with the constraint that each block contained 1 neutral
selected from the International Affective Picture System image. All experimental events and stimulus presenta-
(IAPS; Lang, Bradley, & Cuthbert, 2008). While tions were controlled and recorded using E-Prime 2.0
instructions were provided within the task, the instruc- software (Psychology Software Tools) on a standard
tions were also presented to the participant verbally by PC. All images and instructions were presented using
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
the researcher before the practice block began. The a rear-bore projection screen and projector (Avotec
remaining 4 blocks took place within the MRI scanner Silent Vision).
with the images from the Military Affective Picture
System (details reported below; Goodman et al., 2016).
Procedures
Once in the scanner, the instructions were presented
again before the first block. Each block contained Upon arrival at the Auburn University MRI Research
24-trials, with 6 repetitions of 4 trials: suppress Center for their scheduled testing appointment,
negative images, enhance negative images, maintain participants were re-screened for eligibility and
negative images, and maintain neutral images. re-consented to ensure full comprehension of the
The negative MAPS images were supposed to arouse a study’s procedures. Participants completed a battery of
negative emotion in the subject, while the neutral computerized neurocognitive tests and questionnaires
MAPS images were supposed to not alter the preexisting assessing demographics, medical history, and psycho-
emotional state of the subject. These trials were logical health. After completion, participants were then
pseudorandomized so that no more than three trials escorted to the MRI suite for the imaging portion of the
of the same regulation instruction were shown study. After completion of the study, participants
consecutively. received monetary compensation for their time.
administered the Test of Memory Malingering (Tom- subjected to multiple regression using a prewhitening
baugh, 2003), which consists of two learning trials and technique to account for intrinsic autocorrelations of
a retention trial that uses pictures of common, everyday the blood-oxygen-level-dependent (BOLD) signal.
objects (e.g., chair, pencil). A cut-off score (<45 correct)
was used to determine eligibility for participation in the
Functional data analysis
study. All participants passed the first trial.
First-level contrasts were created to examine the effect
of emotion regulation on participants’ neural activity.
Statistical analysis for self-report and
During the emotional regulation task, we considered
neurocognitive measures
only activity during the “regulation” phase of the trial
Nominal and ordinal data were analyzed using Chi (cf. Figure 1). Parameter estimates were generated for
squared (v2) and Kendall’s tau-b (Tb). Student’s t-tests each voxel independently for the various regulation
were used to explore group mean differences on conditions, and contrasts between the estimates were
self-report and neurocognitive measures (Benjamini- used to create the statistical images (i.e., to determine
Hochberg adjusted to control for false discovery rate). the effects of “suppression” and “enhancement,”
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
For the ratings of perceived success at regulating estimates from when participants were given the
emotion, two-way repeated-measures analysis of “suppress” or “enhance” instruction during negative
variance (ANOVA) of Group (mTBI, Control) x imagery were contrasted against estimates during the
Instruction (enhance, maintain, suppress) and Group “maintain” condition at each voxel independently). In
x Stimulus Type (negative, neutral) were used to analyze order to determine the effects of negative valenced
the data. images, we compared parameter estimates from when
participants were given the “maintain” instruction for
neutral imagery to the same instruction for negative
Imaging acquisition
imagery. After first-level analyses were completed,
Soldiers were scanned in a 3TVerio scanner (Siemens second-level analyses were conducted to combine each
Healthcare, Erlangen, Germany) using T2* weighted participant’s runs together. The resultant statistical
multiband echo-planar imaging (EPI) sequence with maps were then subjected to group level contrasts to
TR ¼ 600 ms, TE ¼ 30 ms, FA ¼ 55˚, multiband compare mTBI and Control groups using mixed effects
factor ¼ 2, voxel size ¼ 3.5 � 3.5 � 5 mm3, 680 volumes modeling, controlling for PTSD symptoms by using
per run and 4 runs per subject. Total acquisition time PCL scores as a covariate. Specifically, PCL scores were
was approximately 6 min and 48 s per scan and 27 min mean centered across all subjects allowing for resulting
and 12 s overall (for the ERT runs). A 32-channel head inferences to be adjusted for any PCL score differences
coil (Siemens Healthcare, Erlangen, Germany) outfitted between groups. We also examined the main effect of
with a mirror was used. Stimuli were presented using PCL score. Group maps were thresholded based on
an Avotec Silent Vision LCD projection system. the magnitude (z ≥ 2.3) and extent (cluster p < 0.05)
Anatomical data was obtained using the magnetization- of activation. This second-level analysis was implemen-
prepared rapid gradient echo (MPRAGE) sequence using ted to assess differential neural activity during emotion
the following parameters: TR ¼ 1900 ms, TE ¼ 2.5 ms, regulation conditions across groups. Resultant MNI
FA ¼ 9˚ and voxel size ¼ 1 � 1 � 1 mm3 coordinates were converted to Talairach space using
Lancaster et al.’s (2007) icbm2tal transform, and regions
were identified using the Talairach Client (http://www.
Functional data preprocessing
talairach.org/manual.html).
Functional data were preprocessed and analyzed using
FMRIB’s Software Library fMRI Expert Analysis Tool
Results
(FSL FEAT, Smith et al., 2004). Specifically, for prepro-
cessing, data were brain extracted, slice-time corrected, Group differences are shown in Table 1. There were
smoothed (5 mm FWHM Gaussian kernel), high-pass significant differences between the mTBI (n ¼ 37) and
filtered, motion corrected using FSL’s MCFLIRT Control (n ¼ 35) groups in the distribution of females,
(Jenkinson, Bannister, Brady, & Smith, 2002), realigned, v2 ¼ 8.20, p ¼ .004. There were no significant differ-
and registered to each individual’s high-resolution ana- ences between the groups in terms age or rank, p > .05.
tomical data, and normalized to Montreal Neurological There were significant group differences with the Con-
Institute (MNI) standard space using FLIRT (Jenkinson trol group being slightly more educated, t(67) ¼ 2.66,
et al., 2002; Jenkinson & Smith, 2001). Data were also p ¼ .010, than the mTBI group. As expected, the
6 M. N. DRETSCH ET AL.
Table 2. Mean (standard deviation) differences, effect sizes (d), and confidence intervals (CI) between Control group and mTBI group
on psychological health measures.
Group PCL NSI AUDIT LEC ESS ZAS ZDS CE
Control
M(SD) 30.06 (12.86) 9.35 (9.39) 4.74 (4.05) 37.89 (14.63) 9.21 (3.67) 30.11 (8.26) 32.26 (9.64) 54.15 (12.68)
mTBI
M(SD) 72.89 (13.74) 44.97 (15.17) 8.63 (7.96) 22.08 (10.47) 13.43 (5.43) 50.86 (9.12) 53.06 (9.53) 52.26 (12.54)
d 3.22* 2.89* 0.65* 1.26* 0.93* 2.39* 2.17* 0.15
95% CI 49.2, 29.5 9.8, 21.8 6.9, 0.8 9.8, 21.8 6.4, 2.0 24.9, 16.6 25.4, 16.2 4.2, 7.9
Note. PCL ¼ PTSD Checklist-5; NSI ¼ Neurobehavioral Symptom Inventory; AUDIT ¼ Alcohol Use and Dependency Identification Test; LEC ¼ Life Events
Checklist; ESS ¼ Epworth Sleepiness Scale; ZAS ¼ Zung Anxiety Scale; ZDS ¼ Zung Depression Scale; CE ¼ Childhood Environment.
*p � .05 (Benjamini-Hochberg adjusted).
APPLIED NEUROPSYCHOLOGY: ADULT 7
Table 3. Mean (standard deviation) differences, effect sizes (d), and confidence intervals (CI) between Control group and mTBI group
on standardized throughput scores on ANAM subtests.
ANAM Subtest
Group SRT SRT2 PRT CDS CDD MTS MTH
Control
M(SD) 103.86 (14.16) 104.32 (14.58) 104.98 (15.71) 105.45 (15.94) 104.37 (15.34) 102.16 (16.91) 101.35 (15.97)
mTBI
M(SD) 95.71 (16.73) 96.00 (16.19) 97.11 (13.95) 94.94 (13.66) 96.87 (14.92) 96.08 (15.09) 98.65 (15.82)
d 0.53* 0.54 0.53 0.71 0.50 0.38 0.17
95% CI 0.88, 15.43 1.08, 15.55 0.89, 14.84 0.39, 14.61 3.54, 17.47 1.44, 13.63 4.78, 10.17
Note. Higher standardized scores ¼ better performance. SRT ¼ Simple Reaction Time; SRT2 ¼ Simple Reaction Time-Delayed; PRT ¼ Procedural Reaction Time;
CDS ¼ Coded Substitution Test; CDD ¼ Coded Substitution-Delayed; MTS ¼ Matching to Sample; MTH ¼ Mathematical Processing.
*p � .05 (Benjamini-Hochberg adjusted).
Stokes, & Leal, 2007; Kontos et al., 2013; Walker, stimuli used in the task are negatively valenced (e.g.,
Franke, McDonald, Sima, & Keyser-Marcus, 2015). This perceived as threatening), individuals with PTSD show
is most likely due to qualitative differences in events what is described as attentional bias (Sipos, Bar‐Haim,
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
between combat-related mTBI in military personnel Abend, Adler & Bliese, 2014). In the current study,
versus noncombat mTBI in civilians. our mTBI group with elevated PTSD symptoms showed
There is evidence suggesting mTBI is associated with a mild decrement in attention as assessed via simple
cognitive impairments (Spira, Lathan, Bleiberg, & Tsao, reaction time. Since this task consists of neutral stimuli,
2014), but this is not always the case (Dretsch, we cannot rule out an attentional bias toward the
Silverberg, et al., 2015). More consistently, cognitive negatively valenced stimuli presented in the emotional
impairments have been associated with PTSD (Dretsch, regulation task.
Thiel, Athy, Irvin, et al., 2012). However, when the Prior evidence indicates soldiers with PTSD are
hypersensitive to aversive stimuli (Dretsch et al.,
2012), which appears to be linked with differences in
their neurocircuitry associated with hypersensitivity to
threat (Dretsch, Wood, et al., 2016). Considering this
evidence, it was necessary to control for PTSD symptom
severity when comparing groups during the specific
conditions of the emotion regulation task.
When controlling for PTSD symptoms, the mTBI
group had significantly greater activation in neural
regions during the enhance condition that might reflect
compromised cognitive-emotional processing; pre- and
postcentral gyri, inferior parietal lobe (IPL), insula, and
superior temporal gyrus (STG). The precentral gyrus
forms part of the supplementary motor area (pre-
SMA); with connections to the PFC, and functionally
linked with language and memory tasks (Johansen-Berg
et al., 2004; Kohn et al., 2014). The precentral gyrus has
been liked with employing self-focused or situation-
focused reappraisal strategies to increase or decrease
emotion (Ochsner et al., 2004) and viewing positive
valenced images (Seo et al., 2013). The IPL has been
implicated in the interpretation of sensory and
emotional information (Nomi et al., 2008; Pessoa,
Kastner, & Ungerleider, 2002). The postcentral gyrus
is the principle somatic sensory region, which shows
increased activation while enhancing emotions associa-
ted with negative valenced images (Ochsner et al., 2004).
Figure 2. Group differences during enhancement (Enhance - The STG has been linked with facial processing and
Negative Maintain) for mTBI > Controls. Warmer colors indicate social cognition (Adolphs, 2001; Haxby, Hoffman, &
higher z-statistics. Gobbini, 2000). Lastly, the insula has been implicated
8 M. N. DRETSCH ET AL.
in emotional-awareness and interoceptive awareness with increased BOLD signal in both prefrontal and lim-
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
(Garfinkel & Critchley, 2013; Gu, Hof, Friston, & bic regions. These regions included areas commonly
Fan, 2013). associated with regulation of emotion, especially the
It is unlikely the results of the current study were medial prefrontal cortex and cingulate gyrus (Ochsner
explained by PTSD symptoms in the mTBI group since & Gross, 2005; Urry et al., 2006), and have been
we controlled for this in the analysis. In support of this, implicated in regulation emotions in both civilian and
there was a main effect of PTSD symptoms on neural military populations with PTSD (Dretsch, Wood,
activation during the enhance condition. The findings et al., 2016; Greco & Liberzon, 2016).
suggest that elevated PTSD symptoms were associated Alternatively, other psychological health symptoms
could account for the significant neural hyperactivation
during the enhance condition, especially since
depression scores were greater than those reported by
controls. We did not control for depression due to
issues of multicollinearity and the potential of over-
fitting. However, prior findings using a similar task
have linked depression with greater activation in right
prefrontal cortex, but in the suppress condition only
(Johnstone et al., 2007). In contrast, our study
showed significant hyperactivation of multiple ROIs
that have been implicated in both cognitive and
affective processes, but only in the enhance condition,
not the suppression condition. As such, multiple
differences in clinical populations recruited by
Johnstone et al. and our study are likely to underlie
differences in findings.
Based on the functional imaging findings, an
attentional bias toward negatively valenced stimuli in
the mTBI group may be a cogent explanation. However,
since these findings were not observed during the
maintain condition, it is likely that hyperactivation in
the enhance condition is a result of a susceptibility to
heightened responses to relevant stimuli when
prompted. For example, the contextual images from
the MAPS may have elicited greater emotions associated
with prior combat experiences that resulted in the part-
icipants’ injuries. Although we controlled for PTSD
Figure 3. Main Effect of PTSD symptoms (PCL scores) on symptoms, this hyperactivation could be driven by
enhancement (Enhance - Negative Maintain) for mTBI > greater propensity for increased neural activity when
Controls. Warmer colors indicate higher z-statistics. prompted (explicitly) or potentially implicitly (i.e., when
APPLIED NEUROPSYCHOLOGY: ADULT 9
left to one’s own devices: counterfactual thinking). them more likely to sustain an mTBI (Dretsch, Silver-
Alternatively, or in combination, the results might be berg, et al., 2016; Dretsch, Williams, et al., 2016;
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
explained by dysregulation of the neurocircuitry due Panenka et al., 2017). Furthermore, some individuals
to injury pathophysiology (e.g., potentially a result of may be less resilient to the effects of mTBI and combat
frontal and anterior neuronal contusion, axonal injury, and, therefore, develop persistent present postconcus-
and/or from secondary biochemical and metabolic sive symptoms and continue to seek treatment. As such,
changes; Dretsch, 2010; Eierud et al., 2014; Little et al., it is feasible that patients recruited and enrolled in the
2014). Interestingly, there were no significant differ- mTBI group may have had a similar abnormal pattern
ences in self-reported success in enhancing emotional of neural activation prior to their injury(s). Finally,
responses to support perceived difficulty on behalf of given that some of the MAPS images contain faces of
the participants. Regardless of the precise mechanistic both civilians and soldiers, it might be valuable to
underpinning reason, to the best of our knowledge, this explore whether differences in neural activation was
is a first to illustrate a relationship between increased associated with a general response (vigilance) to combat
emotional load and the underpinning neurophysiologic images or moderated by specific image content (e.g.,
correlates in a military mTBI population while control- faces versus no faces).
ling for PTSD symptoms.
Conclusion
Limitations
The findings in the current study revealed that in
In the current study, we controlled for PTSD symptoms soldiers with a history of deployment, there is a different
in our imaging analysis. Given that the mTBI group also pattern of neural activation in those with mTBI
had significantly greater psychological health symptoms compared to those without when instructed to volun-
including anxiety and depression, and potential tarily enhance emotional responses to combat-related
for greater aversive medication interactions from visual stimuli. These differences were not explained by
polypharmacy use, it cannot be ruled out that the elevated PTSD symptoms in the mTBI group as this
findings are partially explained by factors other than was controlled for in our analysis. Significant group
mTBI pathophysiology alone. Future efforts should differences were not observed during either the main-
attempt to address the impact of these factors on tain or suppress conditions. Hyperactivation of neural
emotional regulation. An alternative explanation for regions in the mTBI group during the enhance
the findings could be due to unintentional selection bias condition may reflect underlying vigilance towards
during the recruitment phase rather than the product of negative contextual stimuli and/or poor strategy that
injury/trauma-related neurobiological changes. In this might result in poor allocation of resources to regulate
scenario, individuals with mTBI would have been emotions. It is likely that this abnormal pattern
vulnerable to persistent complaints based on their pre- contributes to the development and maintenance of per-
morbid ability to emotionally regulate. Hence, the find- sistent postconcussive symptoms following mTBI. How-
ing of abnormal neural activation associated with task ever, whether this pattern is associated with injury-
performance could reflect a specific phenotype or inter- related pathophysiology, comorbid psychological health
mediate phenotype of vulnerability. Evidence suggests conditions, an intermediate phenotype, or a
some individuals have a genetic predisposition making combination of these factors, remains unclear.
10 M. N. DRETSCH ET AL.
Gray, M. J., Litz, B. T., Hsu, J. L., & Lombardo, T. W. (2004). Lancaster, J. L., Tordesillas‐Gutierrez, D., Martinez, M.,
Psychometric properties of the life events checklist. Assess- Salinas, F., Evans, A., Zilles, K., … Fox, P.T. (2007). Bias
ment, 11(4), 330–341. between MNI and Talairach coordinates analyzed using
Greco, J. A., & Liberzon, I. (2016). Neuroimaging of fear- the ICBM‐152 brain template. Human Brain Mapping,
associated learning. Neuropsychopharmacology, 41(1), 28, 1194‐1205.
320–34. Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2008). Inter-
Gross, J. J. (2014). Handbook of emotion regulation, national affective picture system (IAPS): Affective ratings
New York, NY: Guilford. of pictures and instruction manual (Technical Report
Gu, X., Hof, P. R., Friston, K. J., & Fan, J. (2013). Anterior A-8). Retrieved from University of Florida, The Center
insular cortex and emotional awareness. Journal of Com- For The Study Of Emotion And Attention.
parative Neurology, 521(15), 3371–3388. https://doi.org/ Linnman, C., Zeffiro, T. A., Pitman, R. K., Milad, M. R.
10.1002/cne.23368 (2011). An fMRI study of unconditioned responses in
Guyker, W. M., Donnelly, K., Donnelly, J. P., Dunnam, M., post-traumatic stress disorder. Biology of Mood & Anxiety
Warner, G. C., Kittleson, J., … Meier, S. T. (2013). Dimen- Disorders, 1(1), 8.
sionality, reliability, and validity of the Combat Experiences Little, D., Geary, E., Moynihan, M., Alexander, A.,
Scale. Military Medicine, 178(4), 377–384. Pennington, M., Glang, P., Schulze, E., … Huang, J. H.
Haxby, J. V., Hoffman, E. A., & Gobbini, M. I. (2000). The (2014). Imaging chronic traumatic brain injury as a risk
distributed human neural system for face perception. factor for neurodegeneration. Alzheimer’s & Dementia,
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
functional connectivity of the human caudate: An appli- Zung, W. W. (1971). A rating instrument for anxiety
cation of meta-analytic connectivity modeling with beha- disorders. Psychosomatics, 12(6), 371–379.
vioral filtering. Neuroimage, 60(1), 117–129. https://doi. Zung, W. W., Richards, C. B., & Short, M. J. (1965).
org/10.1016/j.neuroimage.2011.12.010 Self-Rating Depression Scale in an outpatient clinic:
Saunders, J. B., Aasland, O. G., Babor, T. F., De la Fuente, J. R., Further validation of the SDS. Archives of General
& Grant, M. (1993). Development of the alcohol use Psychiatry, 13(6), 508–515.
disorders identification test (AUDIT): WHO collaborative
project on early detection of persons with harmful alcohol
consumption‐II. Addiction, 88(6), 791–804. Appendix A. psychological health measures
Schwab, K. A., Baker, G., Ivins, B. J., Sluss-Tiller, M., Lux, W.,
& Warden, D. (2006, March). The Brief Traumatic Brain a) PTSD Checklist-5 (PCL-5; Blevins et al., 2015). The
Injury Screen (BTBIS): Investigating the validity of a self- PCL-5 is a 20-item self-report measure that assesses
report instrument for detecting traumatic brain injury
(TBI) in troops returning from deployment in Afghanistan the 20 Diagnostic and Statistical Manual of Mental
and Iraq. Neurology, 66(5), A235–A235. Disorders, Fifth Edition (DSM-5) symptoms of
Seo, D., Olman, C. A., Haut, K. M., Sinha, R., MacDonald, PTSD. The PCL-5 has a variety of purposes,
A. W., & Patrick, C. J. (2013). Neural correlates of including screening individuals for PTSD, making
preparatory and regulatory control over positive and provisional PTSD diagnoses, and monitoring symp-
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
sustained a mild TBI. Subjects identify the mech- i) Alcohol Use Dependency Identification Test
anism (e.g., fragment, vehicular, blast), symptoms (AUDIT; Saunders, Aasland, Babor, De la Fuente,
immediately after the injury, and current symptoms & Grant, 1993). The 10-item instrument is used to
believed to be associated with an injury. A mTBI assess alcohol use. A score of 8 or more in men,
positive score required both the endorsement of an and 7 or more in women, indicates likelihood of
injury-related event, and at a minimum, an altered hazardous or harmful alcohol use. A score of 20 or
state of consciousness (e.g., being dazed, confused, more suggests potential alcohol dependence.
or seeing stars; posttraumatic amnesia; loss of j) Epworth Sleepiness Scale (ESS; Johns, 1991). This
consciousness <20 minutes). instrument asks the subject to rate his or her
f) Neurobehavioral Symptom Inventory (NSI; Cicerone, probability of falling asleep on a scale of
1995). The NSI is a 22-item measure designed to increasing probability from 0 to 3 in eight different
evaluate self-reported postconcussion symptoms situations.
(e.g., headache, balance problems, nausea, fatigue, k) Automated Neuropsychological Assessment Metrics
sensitivity to noise, irritability, sadness, nervousness, (ANAM) Version 4. The ANAM is a reliable auto-
difficulty concentrating, difficulty remembering, visual mated, computerized neurocognitive assessment that
Downloaded by [Australian Catholic University] at 04:54 23 September 2017
problems) on a scale from 0 to 4. The scale yields a includes six subtests: Simple Reaction Time (SRT),
total score ranging from 0–88. Procedural Reaction Time (PRT), Code Substitution
g) Zung Depression Scale (ZDS; Zung, 1971a). The ZDS Learning (CSL), Matching-to-Sample (M2S), Math-
is a 20-item scale measuring self- reported symptoms ematical Processing (MTH), Code Substitution
of depression. The ZDS produces scores ranging Delayed (CSD), and Simple Reaction Time Repeated
from 20 through 80; 20–44 for normal range, 45–59 (SRT2). The ANAM records accuracy, speed, and
for mildly depressed, 60–69 for moderately throughput (TP) performance for each subtest. TP
depressed, and 70 and above for severely depressed. is a single outcome measure produced from percent-
h) Zung Anxiety Scale (ZAS; Zung, 1971b). This ZAS is age correct (accuracy) divided by mean reaction time
a 20-item, self-report anxiety rating scale. The ZAS (speed). Therefore, TP scores represent the correct
has scores ranging from 20 through 80; 20–44 for number of responses per minute of available
normal range, 45–59 for mild to moderate anxiety, response time; thus, higher values indicate better
60–74 for marked to severe anxiety, and 75–80 for performance. Standardized TP scores were used in
extreme anxiety. the analysis.