Veterans with chronic
TBI demonstrate diminished Implicit Probabilistic Sequence Learning
Kiser, S., Chanpimol, S., Fioriti,
C., & Schwartz, B.
Background:
The goal of this study was to conduct the first experiment
to examine implicit probabilistic sequence learning (IPSL) in Veterans with chronic
TBI. Considerable published work has investigated the impact of various
neurologic disorders on IPSL, a cognitive process which plays an integral role
in motor-skill acquisition and inhibitory control. These are important to
rehabilitation/recovery after physical impairment. This line of research is particularly
important to our Veterans due to the significant risk for TBI. Presently, there
is neither an existing standard measure of IPSL for use in clinical populations
nor a clear understanding of what role nonconscious cognitive processes play in
recovery and functional outcomes.
Methods:
Veterans, 18-60 years of age, were recruited from the
Washington DC VA Medical Center. IPSL was assessed using the Triplets Learning
Task (TLT). In the TLT, learning is measured by calculating the relative
reaction-time between low and high frequency task conditions. Greater implicit
learning is demonstrated by higher reaction time difference between the task
conditions. Participants were unaware of the probabilistic nature of the TLT
and probed afterwards using a recognition test to confirm lack of awareness. A
mixed-factorial design was used to compare IPSL in Veterans with chronic TBI to
a Veteran control group (i.e. no diagnosis of TBI). Exclusion criteria included
a diagnosis of substance abuse within 3 months of enrollment, pre-morbid
history of neurologic disease (e.g., stroke), and prior history of bipolar
disorder, schizophrenia, or severe psychiatric illness as determined by the 22-item
Modified Mini Screen and the medical record. Data were analyzed using
repeated-measures ANOVA and, when necessary, corrected for sphericity and
multiple comparisons using Greenhouse-Geisser and Bonferroni adjustments,
respectively.
Results:
In this preliminary analysis, data were collected from 9
Veterans, 5 with a diagnosis of chronic TBI. Results from the mixed model ANOVA
demonstrated robust effect sizes between overall reaction-time to high and low
frequency events (F(1,7)=22.96; p=.003; η2p=.793) and reaction-times with
practice (F(5,35)=11.90; p<.001; η2p=.665). These results show that Veterans in both the experimental and
control groups demonstrate IPSL and inhibitory control in the experimental
task. However, metrics of implicit learning between groups showed marginal
significance (F(1,8)=4.57, p=0.07), where Veterans with TBI performed worse
than the control group. Statistical analyses of the recognition test revealed
that Veterans were unable to identify probabilistic event structures beyond
chance. These finding shows that IPSL is diminished in Veterans with chronic
TBI, adversely affecting motor-skill learning.
Discussion:
To our knowledge, the present study is the first to investigate
IPSL in Veterans with chronic TBI. The novelty of our findings is twofold.
First, both Veteran groups appear to learn probabilistic regularities in the
TLT, though those with chronic TBI perform worse than controls. Second, these
preliminary results serve as a proof of concept demonstrating the ability to
identify sensitive metrics of IPSL, motor learning, and inhibitory control in
Veterans with chronic TBI. Furthermore, these findings suggest cognitive
deficits not presently accounted for in neuropsychological assessment that
likely contribute to functional outcomes in TBI. Understanding the relationship
between TBI and IPSL should prove useful for informing rehabilitation techniques,
particularly when motor-skill acquisition is required. Additionally, IPSL is a
marker of striatal-dependent learning and could potentially characterize
disease progression following TBI. Previous studies have linked IPSL performance
to poor dopamine regulation in the caudate nucleus of the striatum, which is
shown to decrease into old age (65+ years) and in chronic diseases such as
Parkinson’s disease and Huntington’s disease. The pattern of learning exhibited
by Veterans with chronic TBI suggests untimely deterioration of striatal
function and could indicate dopamine dysregulation or depletion. This is
significant because pathophysiology of TBI is expected to impair this kind of
function. Further research is needed to understand why Veterans with chronic
TBI demonstrate reduced IPSL and whether TBI directly or indirectly affects
striatal-dependent learning.
