Page 19 - JSOM Summer 2025

be knocked down to allow the participant to continue through during the dynamic shoot/no-shoot task was segmented out
the course. After completing both marksmanship tasks, the in- from the total visit based on the time duration of the task. The
ert pistol was holstered, the fNIRS sensors were removed, and data was then filtered for Mayer waves, respiration, and heart
the study visit was completed. pulsation through the examination of the power density spec-
trum and by applying a low pass filter at 0.14Hz.
Hemodynamic Monitoring of the Prefrontal Cortex
To quantify changes in hemodynamics within the PFC, a con- After filtering the data, the hemodynamic signals (HbO , de-
2
tinuous wave fNIRS probe (PortaLite MkII. Artinis Medical oxygenated Hb, and total Hb) from each of the three long
System, Elst, Utrecht, The Netherlands) was placed 2cm above channels (2.9, 3.5, and 4.1mm) were averaged together to rep-
the right eyebrow for each participant and adhered to the skin resent one regional relative concentration value of HbO , de-
2
using hypoallergenic double-sided tape. To further secure the oxygenated Hb, and total Hb of the PFC. To quantify the time
sensor to the participant’s head and protect the sensor from course of changes during the dynamic shoot/no-shoot task, the
contamination by ambient light exposure, the sensor was relative concentration values were bin-averaged into quartiles
wrapped with a black disposable foam elastic bandage around based on the total duration of the dynamic shoot/no-shoot
the participant’s head (3M Corporation, St. Paul, MN, USA). marksmanship task (Table 1). The segmentation, filtering, and
The fNIRS sensor consisted of three light-emitting diodes and analysis of the fNIRS-derived hemodynamic response during
two detectors, placed at inter-optode distances of 2.9, 3.5, and the dynamic shoot/no-shoot task were performed offline using
4.1cm for the three long channels and 0.70, 0.80, and 0.74cm custom-built LabView programs (LabView Professional 2022,
for the three short channels. The control unit was synced to a NI, Austin, TX).
laboratory desktop (HP Envy Desktop. HP Inc., Palo Alto, CA)
via a Bluetooth dongle and placed in a belt bag secured to the Statistical Methods
participant’s hip. All fNIRS-derived hemodynamic concentra- For participant characteristics, marksmanship screening score,
tions (HbO , deoxygenated Hb, and total Hb) were collected and total time of COF, an independent t test was performed to
2
at a sampling frequency of 25Hz and telemetrically sent to the compare groups. For fNIRS parameters, a linear mixed effects
desktop and visualized using Oxysoft software (version 3.4, model (LMM) with fixed effects for the marksmanship group
Artinis Medical Systems B.V., Elst, Utrecht, The Netherlands). (proficient marksmen and non-proficient marksmen) and
time (Q , Q , Q , and Q ) with random effects for intercept
3
4
1
2
Stationary and Dynamic Shoot/No-Shoot and participant were performed. Accounting for participant
Marksmanship Tasks differences across all groups as a composite allowed for the
The screening protocol and the shoot/no-shoot course of fire examination of group differences across the time course of the
(COF) were performed using an inert recoil-enabled Glock 17 shoot/no-shoot task.
pistol (Laser Ammo Ltd., Great Neck, NY), equipped with a
custom drop-in barrel. The barrel was customized to have an For all LMMs, post hoc analyses using Fisher’s Least Signifi-
infrared SureStrike vibration-activated laser integrated into cant Difference (LSD) were used on all pairwise comparisons.
the end of the barrel. To simulate recoil, pressurized green The effect sizes for independent t tests were calculated using
gas (Elite Force Airsoft, Fort Smith, AR) was loaded into the Cohen’s d with allocated benchmarks consisting of small
hollow magazine chamber before being inserted into the mag- (d=.2), medium (d=.5), and large (d=.8). All analyses were
33
azine housing. The screening protocol was a custom design conducted using SPSS (version 29, IBM SPSS Inc., Chicago,
course built into the “M-Range” software add-on using the IL), and an alpha level p≤.05 was considered statistically sig-
Smokeless Range 2.0 Judgmental and Marksmanship Shoot- nificant for all comparisons.
ing Simulator (Laser Ammo Ltd., Great Neck, NY). Similarly,
the shoot/no-shoot COF was also performed using the Smoke- Results
less Range simulator; however, a pre-designed course in the
Tactical Targets software (standardized Course #6) was used. There were no significant differences between age, height,
weight, resistance training, aerobic training, and time to com-
During both the screening and shoot/no-shoot COF, each par- plete the dynamic marksmanship task between marksmanship
ticipant stood approximately 3.05m from a white wall with a groups (d=.28–.68; p=.08–.24). There was a significant dif-
projected display area corresponding to 1.83x2.74m (View- ference in body mass index (BMI) between groups, indicating
Sonic Home theater PA503S DLP Projector. ViewSonic Inc., the proficient marksmanship group had a higher BMI than
Brea, CA, USA). To assess target hits and misses and display the non-proficient marksmanship group (d=.90; p=.04) (Table
the total number of shots fired, a short-throw camera (Laser 1). For the stationary marksmanship task, the average hit per-
Ammo Ltd., Great Neck, NY) was mounted behind the pro- centage for the proficient marksmanship group (86.7%) was
jector and calibrated to detect the appropriate display area. significantly greater than the non-proficient marksmanship
Prior to completing the stationary marksmanship task and the group (42.5%) (d=3.09; p<.001).
shoot/no-shoot COF, the marksmanship simulator was cali-
brated to adjust for light fluctuations and screen parameters. There was no significant group-by-time interaction, nor main
effect for time for HbO (Table 2). However, a significant dif-
2
Hemodynamic Processing ference in group indicated that the non-proficient marksmen
Throughout the entire visit, the relative concentration values of had a 26.3% greater HbO response on average than the profi-
2
HbO , deoxygenated Hb, and total Hb were calculated using cient marksmen throughout the dynamic marksmanship COF
2
the modified Beer-Lambert law equation using Oxysoft (ver- (Figure 2A). For deoxygenated Hb, there was no significant
sion 3.4, Artinis Medical Systems B.V., Elst, The Netherlands) group-by-time interaction nor main effects for time or group
with the differential pathlength factor calculated based on a (Table 2). For total Hb, there was no significant group-by-
standardized age-based equation. The fNIRS data obtained time interaction nor main effect for time (Table 2; Figure 2B).
32
fNIRS-differentiated Marksmanship Performance | 17