Abstract
Objective
The purpose of this meta-analytic review is to examine the relationship between increased psychological pressure and Use of Force (UOF) behaviours, identifying current training methodologies and effectiveness of transfer of training interventions in high threat-simulated scenarios.
Background
Data from UOF performance within Law Enforcement indicates a low transfer of marksmanship training into real-world UOF, resulting in unnecessary damage to property, personal injury and increased risk to loss of life. This meta-analysis examines both the impact of increased pressure and current training interventions.
Method
A meta-analysis was conducted across a wide range of published research to answer the primary research questions.
Results
Increased levels of perceived pressure demonstrated an average decrease in marksmanship accuracy of 14.8%, together with a small increase in incorrect Decision Making (DM) and faster reaction Times (RT). Experience demonstrated a mitigating effect for pressure for marksmanship with a 1.1% increase for every one year of service but no effect on DM or RT. Training interventions utilizing a variety of early contextually relevant exposures to increased pressure improved performance over traditional training on average by 10.6%.
Conclusion
The outcomes illustrate the negative effect of pressure on marksmanship and UOF behaviours, and that early exposure to contextually relevant pressure may increase the transfer of training to real-world performance.
Application
Occupational experience is an important component in reducing the impact of pressure on UOF performance, and transfer of training may be enhanced through training methodologies that combine early exposure to contextually relevant pressure, that may replicate the benefits of experience.
Keywords: decision making, skilled performance, transfer of training, simulation based skill acquisition, perception-action
Introduction
Military, police, security guards, and similar tactical populations are required routinely to carry firearms with the potential to utilise them in spontaneous Use of Force (UOF) actions. These UOF actions can result in fatal consequences, but are rarely executed, and only in response to an individual’s perception of an imminent threat (Burrows, 2007). The increased level of both perceived and physical threat, complexity and constraints within these scenarios typically elicit an acute increase in physiological arousal, which in turn increases perceived pressure and states of anxiety (Anderson et al., 2002; Arble et al., 2019).
Increases in anxiety often constrain or delay cognitive processing, which can ultimately result in sub-optimal execution of the required motor skills and performance. These occur through changes in gaze behaviours, reaction and decision making times, the misidentification of relevant cues, and a reduced accuracy in skilled marksmanship (Lieberman et al., 2006; Murray & Janelle, 2003; Nieuwenhuys & Oudejans, 2010). Although members of tactical organisations undergo specific firearms training and routine recertification, the training and instruction is mostly designed around range based marksmanship training with little to no contextually relevant threat to create perceived pressure. The impact of this perceived pressure within a real word environment on their performance and the transfer of marksmanship skills from training to the real-world use of deadly force is still not completely understood. Therefore, the effectiveness of this training has come under increased scrutiny as there appears to be no clear link between performance competency demonstrated during training and real-world performance (Morrison & Vila, 1998; White, 2006).
The data available from police officer UOF incidents indicates there may be a lack of skilled transfer as police officers miss their intended target more often than not when discharging their weapons at an assailant (Morrison & Vila, 1998; White, 2006). Preliminary reviews indicate that performance in the use of deadly force actions may be substantially impaired under increased levels of anxiety. However, conclusions remain equivocal due to a lack of accurate record keeping in these instances. Despite this lack of conclusive evidence, marksmanship performance outcomes during UOF have been reported across a large number of studies as one of two variables: bullet hit rates and incident hit rates. Bullet hit rates are recorded as a comparison of accuracy from the number of shots fired to the number of hits on the intended target. Comparatively, incident hit rates have been defined as the striking of the opponent by at least one bullet. This is irrespective of hit or miss ratios and is the most commonly recorded measure, with performance on incident hit rates ranging from between 14% and 60% in accuracy.
Due to factors such as changes in incident frequency, data recording methods, UOF scenarios, and occupational policies and reviews, the underlying mechanisms through which marksmanship performance has been assessed, have been difficult to interpret (White, 2006). However, when marksmanship performance has been measured by bullet hit rates (i.e., accuracy instead of shots fired), the difference between training outcomes and subsequent performance execution has been more evident. Despite the lack of standardised data collection on UOF situations, the variability in performance may be explained through factors such as the perceived level of threat to officers’ welfare, officers’ experience and/or the proximity of the incident.
The intensity of a perceived threat appears to have a meaningful influence on performance execution (Dunsmoor et al., 2017; Rostker et al., 2008). For example, in incidents where the intensity of a threat may be perceived as high (i.e., a suspect is armed with a firearm and is returning fire over an unarmed individual), an inverse relationship between threat and performance is evident, resulting in a reduction in marksmanship performance.
The proximity of the threat during an engagement has also been highlighted as a significant factor that determines marksmanship accuracy, where optimal accuracy is achieved within three metres and significantly reduces beyond seven metres (Rostker et al., 2008). Reports from across a number of agencies have demonstrated that officers being fired upon during an incident reported that shooting incident hit rates reduced to as low as 14%. In comparison, when there was no return of fire, and the engagement range was within five metres performance, accuracy only decreased to approximately 37% (Morrison & Vila, 1998; Rostker et al., 2008).
Changes in the perceived intensity of a threat have also impacted marksmanship behaviours in terms of the number of shots fired (Rostker et al., 2008). Officers confronting an unarmed suspect fire 4.7 shots on average compared to 11.1 during ‘opposed encounters’ where suspects are returning fire (Rostker et al., 2008). Such reductions in accuracy, combined with an increase in the number of shots fired during reciprocal engagements could result in a significant number of projectiles missing their intended target. Increasing the risk of impacting unintended targets within a populated urban environment may lead to the damage of property, life changing injury, and/or fatal outcomes involving innocent bystanders.
Another consideration for accuracy measures is the deviation of the point of impact from the point of aim, as it enables an accurate measurement of marksmanship accuracy performance. This is a measure that has not been captured during reporting, but may add value in determining the full impact of performance from training environments to the real-world use of force. Drawing on the first author’s experiences during marksmanship training, performers are required to validate their competency by achieving a specific accuracy rate (such as 80% or 8 out of 10 shots) on a target, approximately the size of a A4 piece of paper. Considering the size in comparison to the much larger size of a suspect and variations in the accuracy of reporting, it might be argued that accuracy from range-validation to accuracy in hitting a suspect could be much lower, indicating that the transfer of training in use of force could be considerably lower than that reported, despite consistent advances in training methodology.
When reviewing the evolution of training practices and practical performance, there is reasonable evidence to question the effectiveness of these ‘advances’ in training techniques to marksmanship performance over the past century. For example, Morrison and Vila (1998) compared bullet hit rates or differences in shot accuracy performance between non-trained police officers from the nineteenth century (1863–89), contemporary opponents of police (1970–92), and contemporary trained officers (1975–98). Nineteenth century untrained officers achieved hit rates of between 15 and 22%, contemporary opponents achieved accuracy rates of between 12 and 28%, and the accuracy of contemporary police officers was reported as between 14 and 38% (Morrison & Vila, 1998). These changes in accuracy suggest that, only small increases in performance have been achieved in comparison to nineteenth century officers and untrained opponents. If the evolution of training has only led to small changes in performance, then the challenge to the lack of transfer may lie in cognitive factors and threat-response training, more so than motor skill acquisition. If incorporating these factors can improve the transfer of training to performance, it may recast the approach to training in threat-related scenarios.
The primary aim of this meta-analysis was to review the available literature from tactical populations on pre and post-training analysis using current intervention training strategies to assess the impact on marksmanship performance during low and high perceived threat environments. The secondary aim was to identify the factors that may constrain or negatively affect the transfer of training. This was intended to identify any secondary effects on individuals’ decision making ability in accurately targeting a threat from a non-threat in low and high perceived threat environments.
Methods
This systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) (Moher et al., 2009), and was registered with Open Science Framework (DOI 10.17,605/OSF.IO/V5NPU) prior to the completion of the final database search. The PRISMA checklist consists of a 27-item checklist and four-phase flow diagram for including items considered essential for the transparent reporting of systematic reviews and meta-analysis. The guidelines were created to improve transparency, accuracy, completeness and the frequency of documented systematic reviews and meta-analysis protocols (Moher et al., 2009).
Search Strategy
A systematic literature search was conducted through the databases: MEDLINE, EMBASE, CINAHL, PsycINFO, SPORTDiscus, Web of Science and AUSport. The goal was to identify full text original research studies that compared marksmanship performance between low and high pressure conditions within relevant tactical populations employing randomised controlled trials that involved marksmanship performance interventions. The search was limited to “English language” AND “Adult” AND “Humans”. The following keywords were used to define the population group: ‘military personnel’, OR ‘military’, OR ‘soldier’, OR law enforcement’, OR ‘police’, OR ‘special forces’; high perceived pressure condition: ‘anxiety’, OR ‘pressure’, OR ‘threat’, OR ‘simulated training’; and marksmanship outcome: ‘performance’, OR ‘training’, OR ‘accuracy’, OR ‘shoot’, OR ‘marksmanship’ OR ‘task performance’. These three search concepts were combined using the AND Boolean operator and were adapted for each database. A subsequent manual search of citations, recommendations, and known publications was conducted to include any relevant studies that may not have been identified within the initial search until no further studies were identified.
Study Selection
The titles and abstracts retrieved via the initial search were screened independently by the first author and at a minimum, one of the other co-authors to identify suitable full texts for secondary review. These studies were retrieved in full text format and reviewed for inclusion against the specified criteria by the first author and at least one co-author. Nomination for inclusion was conducted via the software program Covidence (Melbourne, Australia). Covidence is a systematic tool for uploading citations, screening tiles/abstracts, full text screening, and data extraction that allows for asynchronous collaboration between multiple reviewers (Babineau, 2014). Discrepancies in the inclusion of studies was resolved via collaborative deliberation with a third, independent reviewer.
Inclusion Criteria
Studies were eligible for inclusion if they: (1) utilised tactical populations who were required to carry firearms and respond in defence-of-life or property, (2) utilised a comparison between states of low and high perceived pressure, as validated through a state anxiety scale and/or relevant valid self-reporting instrument, (3) measured performance as an indicator of marksmanship or firearm activity, and (4) had a reportable outcome measure (i.e., marksmanship accuracy, reaction time). Studies were excluded if they: (1) utilised non-tactical populations or non-armed tactical populations, (2) did not use an adult population, (3) reported non-tactical firearm or marksmanship-related activities, including sports marksmanship or relevant sports activity (i.e., biathlon, skeet shooting), and/or (4) non-reportable outcomes.
Data Extraction
Once the final studies were confirmed for inclusion, data were extrapolated and checked for accuracy by a minimum of two reviewers. Variables of interest in this review were the level of perceived pressure and changes in marksmanship performance outcomes, as determined by shot accuracy (as a percentage of shots fired) in environments with low and/or high perceived pressure, decision making, reaction time, the influence of experience on performance and the impact of any relevant intervention protocol(s).
Quality Assessment
The risk of bias was assessed by two reviewers using the Mixed Model Appraisal Tool (MMAT) that appraises the methodological quality of five categories to studies: qualitative research, randomised controlled trials, non-randomised studies, quantitative descriptive studies and mixed methods studies (Pluye et al., 2011). Disagreements in scoring were discussed among all reviewers until consensus was reached.
Data Analysis
The Metafor statistical package in R software (version 3.4.3, R Foundation for Statistical Computing) was used to perform random effects meta-analysis using the restricted maximum likelihood approach with inverse variance weighting. The mean differences in marksmanship accuracy, reaction time, and decision making were compared across high and low pressure conditions, while shooting accuracy was compared between training and control groups using separate analyses. Occupational experience was included as a continuous moderator variable in the high versus low pressure analyses. Study ID was included as a random factor in all analyses to account for studies that reported multiple comparisons. Mean difference with 95% confidence interval (CI) was used as the effect size for marksmanship accuracy whereas standardised mean difference (SMD) and 95% CI was used as the effect size for reaction time and decision making due to differences in those outcome measures across studies. SMDs were described as trivial (<0.20), small (0.20 – 0.59), moderate (0.60 – 1.19), large (1.20 – 1.99), and very large (≥2.00) (Hopkins, 2000). Statistical heterogeneity within each meta-analysis was investigated using I2 statistic and considered low (I2 <25%), moderate (I2 = 25–49%) or high (I2 >50% (J. P. T. Higgins et al., 2003)).
Results
Search Results
From the database search, 7692 studies were retrieved with 4666 retained for initial title and abstract screening after duplicates were removed. Of these, 4504 studies were excluded, retaining 162 studies for full text assessment for eligibility with a final inclusion of 17 studies for review. From the final 17 studies, 10 were identified with results relevant to marksmanship performance between high and low perceived pressure states with sufficiently similar reportable results for inclusion in a meta-analysis (Figure 1). The results from the remaining seven studies are included for qualitative analysis. Despite the limited number of studies included in the meta-analysis, the Cochrane Consumers and Communication Review Group recommend that as little as two studies is sufficient to perform a meta-analysis with the provision that those studies can be meaningfully pooled and the results are sufficiently similar (J. P. Higgins et al., 2019; Ryan, 2016). Study characteristic data were extracted and are included in Tables 1, 2 and 3.
Figure 1.
PRISMA flow chart of study selection process for inclusion and exclusion.
TABLE 1:
Article characteristics of randomised control studies
| Intervention summary | Marksmanship outcome | |||||
|---|---|---|---|---|---|---|
| Author (year) | Population (n) | Training | Dose | Rounds; Distance | Pressure | Key Findings |
| Nieuwenhuys and Oudejans (2011) | Police (25M, 2F) | EXP: Practice against hostile opponent CON: P ractice against cardboard/mannequins |
1h/week 3 weeks |
10x4; 5m | HP: Opposed with soap cartridge LP: M annequin |
EXP performance unchanged between LP and HP after training and this was maintained after a retention period. CON performance decreased from LP to HP at all timepoints |
| Oudejans (2008) | Police (15M, 2F) | EXP: Practice against hostile opponent CON: P ractice against cardboard targets |
3 × training session over 2 weeks | 10 × 2; 5m | HP: Opposed with soap cartridge LP: Cardboard targets |
EXP performance unchanged between LP and HP after training. CON performance decreased from LP to HP at all timepoints |
| Colin et al. (2014) | Police (43M, 13F) | EXP: EI–mental imagery EEI–m ental imagery and emotional feeling CON: no imagery |
2 × 1 minute | 4 × 4; 5m | HP: Opposed with soap cartridge LP: Cardboard targets |
EXP performance unchanged between LP and HP after training. CON performance decreased from LP to HP at all timepoints |
| Liu et al. (2018) | SWAT (98m) | EXP: Shooting with live person beside target CON: Shooting with mannequin beside target |
1/h week 3 weeks |
Simulated hostage rescue | HP: Opposed with soap cartridge LP: Cardboard targets |
EXP performance unchanged between LP and HP after training. CON performance decreased from LP to HP at all timepoints |
| Nieuwenhuys et al. (2015) | Police (52M, 5F) | EXP: (VBP-LT) virtual trainer with no threat, (VBP-HT) virtual trainer with shoot back, (RLP-HT) practice against hostile opponent CON: P ractice against cardboard targets |
3 × 1/h sessions | 4 × 12; 5m | HP: Opposed soap cartridges, shoot back canon with soap cartridges LP: Cardboard targets |
Performance decreased from LP to HP with no significant effect of interaction |
| Shipley and Baranski (2002) | Police (40M, 14F) | EXP: Relaxation and visual rehearsal CON: no rehearsal | 1 × 30 min session | 4; 5–10m | HP: Single opposed reality-based scenario | EXP performance was greater than control post intervention |
CON, control group; EXP, experimental group; F, female participants; HP, high pressure condition; LP, low pressure condition; M, male participants; n, sample size; EI; Mental imagery; EEI, Mental imagery and emotional feeling; VBP-LT, Video based practice under low threat; VBP-HT, Video based practice under high threat; RLP-LP, Real-life practice under high threat.
TABLE 2:
Article characteristics of within subject’s comparison studies
| Task summary | Marksmanship outcome | |||||
|---|---|---|---|---|---|---|
| Author (year) | Population (n) | Task | Trials | Rounds; distance | Pressure | Key findings |
| Taverniers & DeBoeck (2014) | Police (25M) | Reality-based scenarios under a LP and HP | 2 LP 2 HP | 4 | HP: Opposed with soap cartridge LP: Mannequin | Performance decreased from LP to HP |
| Taverniers & Suss (2019) | Military (16M) | Less than lethal launcher under LP and HP | 1 × 3 LP 1 × 3 HP | 3; 10m, 20m, 30m | HP: Psychological stress LP: Lab condition | No change in performance between LP and HP at 10 and 20m. Performance decreased at 30m from LP to HP |
| Nieuwenhuys & Oudejans (2010) | Police (6M, 1 F) | Reality-based scenarios under a LP and HP | 5 LP 5 HP | 20; 5m | HP: Opposed with soap cartridge LP: Mannequin | Performance decreased from LP to HP for marksmanship accuracy and reaction time |
| Nieuwenhuys et al. (2012) | Police (33M, 3F) | Reality-based shoot, no – shoot scenarios under a LP and HP | 24 LP 24 HP | 48 | HP: Opposed with rubber cartridges LP: Firing sound | Performance decreased from LP to HP for marksmanship accuracy, decision making and reaction time |
| Nibbeling et al | Military (22M) | Opposed and cognitive tasks under LP and hp conditions | 8 LP 8 HP | 12; 5m | HP: Opposed with soap cartridge LP: Mannequin | Marksmanship performance decreased from LP to HP |
| Friedland & Keinan (2009) | Infantry (297M) | Simulate trench clearing for high, medium and low CE | 1 LP or 1 HP | 12; 10m | HP: Adjacent live firing LP: No firing | Performance was greater for high CE under HP |
| Gould et al. (1987) | Police (35M, 4F) | Progressive shooting competition | 2 LP 2 MP 1 HP | 5 × 6; 5m | HP: Individual competition MP: Team competition LP: No competition | Performance did not differ across conditions |
CON, control group; EXP, experimental group; F, female participants; HP, high pressure condition; LP, low pressure condition; M, male participants; n, sample size; practice under high threat.
TABLE 3:
Article characteristics of within subject’s comparison studies
| Task summary | Marksmanship outcome | |||||
|---|---|---|---|---|---|---|
| Author (year) | Population (n) | Task | Trials | Rounds; distance | Pressure | Key Findings |
| Giessing et al. (2019) | Police (16M, 3F) | Virtual reality-based scenarios under a LP and HP | 2 LP 2 HP | 24; 5m | HP: Opposed with soap cartridge LP: Mannequin | Performance did not differ between conditions |
| Gamble et al. (2018) | Military (26M) | Virtual shoot, no-shoot task | 64 LP 64 HP | 128; 5m | HP: Electric shock LP: Vibration belt | Performance decreased from LP to HP for marksmanship accuracy and decision making |
| Landman et al. (2016a, 2016b) | Police (37M, 5F) | Reality-based scenarios under a LP and HP | 5 LP 5 HP | 40; 5m | HP: Opposed with soap cartridge LP: Mannequin | Performance decreased from LP to HP for marksmanship accuracy and reaction time |
| Landman et al. (2016a, 2016b) | Police (40M) | Reality-based scenarios under a LP and HP compared against experience | 24 LP 24 HP | 48; 5m | HP: Opposed with soap cartridge LP: Mannequin | Performance decreased from LP to HP for marksmanship accuracy and reaction time. Experience officers performed better than inexperienced |
CON, control group; EXP, experimental group; F, female participants; HP, high pressure condition; LP, low pressure condition; M, male participants; n, sample size; practice under high threat.
Study Characteristics
A total of 960 participants (897 male, 63 female) with a mean age of 28.3 (5.76) years and mean occupational experience of 7.7 (5.5) years were involved in the studies. Of the 17 articles included in the review, 13 utilised participants in law enforcement (Colin et al., 2014; Giessing et al., 2019; Gould et al., 1987; Liu et al., 2018; Nibbeling et al., 2012; Nieuwenhuys et al., 2012, 2015; Nieuwenhuys & Oudejans, 2010, 2011; Oudejans, 2008; Saus et al., 2006; Shipley & Baranski, 2002) or infantry soldiers from within the military (6 of 18) (Friedland & Keinan, 1992; Gamble et al., 2018; Nibbeling et al., 2014; Roy et al., 2019; Taverniers & DeBoeck, 2014; Taverniers & Suss, 2019). The main forms of instigating the high pressure condition consisted of opposed scenarios (utilising soap marking cartridges fired at the participant) (14 of 17), electric shock belts (2 of 17), noise stressors (1), nearby live firing (1), and competition (1). From the final 17 studies, six were randomised control trials comparing various intervention strategies over control conditions, while 12 focused solely on the impact of the effect of condition from performance from low pressure to high pressure. The main intervention strategies employed in these studies were high threat opposed training (3 of 6), virtual high threat training (2 of 6) or mental imagery or rehearsal (2 of 6). Full characteristics are detailed in Tables 1, 2 and 3.
Quality Assessment
A summary of the studies included is provided in Table 4. Studies that met one methodological requirement scored 1 point, studies that met two methodological requirements scored 2 points, studies that met three methodological requirements scored 3 points, and studies that met four methodological requirements scored 4 points.
TABLE 4:
Quality assessment summary of included studies
| Author (year) | Design | Participants (n) | Age | Experience | Pressure Manipulation | Quality Assessment |
|---|---|---|---|---|---|---|
| Nieuwenhuys and Oudejans (2011) | Experimental | Police (25M, 2F) | 35 | 12 | Reality-based scenario with soap cartridge | 4 |
| Oudejans (2008) | Experimental | Police (15M, 2F) | 37 | 11 | Reality-based scenario with soap cartridge | 4 |
| Colin et al. (2014) | Within subjects, randomised crossover | Police (43M, 13F) | 29 | 6 | Reality-based scenario with soap cartridge | 4 |
| Liu et al. (2018) | Within subjects, randomised single case | SWAT (98m) | 21 | Not reported | Reality-based scenario with soap cartridge | 2 |
| Nieuwenhuys et al. (2015) | Experimental | Police (52M, 5F) | 37 | 13 | Shoot back canon with soap cartridges | 2 |
| Shipley and Baranski (2002) | Within subjects, randomised crossover | Police (40M, 14F) | 27 | 1 | Reality-based scenario with soap cartridge | 3 |
| Taverniers & DeBoeck (2014) | Within subjects, randomised crossover | Police (25M) | 29 | 8 | Reality-based scenario with soap cartridge | 3 |
| Taverniers and Suss (2019) | Experimental | Military (16M) | 23 | 6 | Psychological stress | 2 |
| Nieuwenhuys and Oudejans (2010) | Experimental | Police (6M, 1 F) | 23 | 3 | Reality-based scenario with soap cartridge | 3 |
| Nieuwenhuys et al. (2012) | Experimental | Police (33M, 3F) | 37 | 15 | Reality-based scenario with soap cartridge | 2 |
| Nibbeling et al | Single case, non-randomised | Military (22M) | Reality-based scenario with soap cartridge | 2 | ||
| Friedland and Keinan (1992) | Single case, non-randomised | Infantry (297M) | 18 | 1 | Adjacent live firing | 2 |
| Gould et al. (1987) | Single case, non-randomised | Police (35M, 4F) | 27 | 2 | Individual competition, team competition, no competition | 2 |
| Giessing et al. (2019) | Single case, non-randomised | Police (16M, 3F) | 22 | Not reported | Reality-based scenario with soap cartridge | 3 |
| Gamble et al. (2018) | Within subjects, randomised crossover | Military (26M) | 30 | Not reported | Electric shock | 3 |
| Landman et al. (2016b) | Within subjects, randomised single case | Police (37M, 5F) | 32 | 9 | Reality-based scenario with soap cartridge | 4 |
| Landman et al. (2016a) | Within subjects, randomised single case | Police (40M) | 32 | Not reported | Reality-based scenario with soap cartridge | 3 |
Marksmanship Accuracy
Nine studies provided sufficient information to be included in meta-analysis of shooting accuracy differences between high and low pressure conditions. Pooled results indicated that mean shooting accuracy was reduced by 14.8% (95% CI: −21.2%, −8.4%; p < 0.001; Figure 2) in the high, compared to low pressure condition. The level of statistical heterogeneity was high (I2 = 81.4%). Service experience was a statistically significant moderator (p = 0.012) and indicated that the mean reduction in shoot accuracy between high and low pressure conditions was improved by 1.1% (95% CI: 0.3%, 2.0%) for every one-year of experience.
Figure 2.
Forest plot of mean difference in accuracy scores from low perceived pressure to high perceived pressure marksmanship scenarios including experience in years.
A further five studies reported results using data that were not recorded as accuracy as a percentage of shots hit from shots fired or that could be converted to compatible data for meta-analytic review. Four individual studies reported a statistically significant reduction in performance from the low pressure to high pressure condition (Gamble et al., 2018), while one reported no statistically significant effect of condition on performance from LP to HP (Giessing et al., 2019) (Table 1, 2 and 3).
Intervention Results
Four studies provided sufficient information to be included in a meta-analysis of shooting accuracy differences between training and control groups (Colin et al., 2014; Nieuwenhuys et al., 2015; Nieuwenhuys & Oudejans, 2011; Oudejans, 2008). The pooled results indicated that mean shooting accuracy was improved by 10.6% (95% CI: 1.7%, 19.5%; p = 0.020) in the training compared to control group (Error! Reference source not found.). The level of statistical heterogeneity was moderate (I2 =35.8%) (Figure 3).
Figure 3.
Forest plot of mean difference in intervention strategies on reducing the impact of scores from low perceived pressure to high perceived pressure marksmanship scenarios. RLP-HT = Realistic practice–High threat, EI–Imagery rehearsal, EEI–Imagery.
A further three intervention-based research studies were identified without pre and post intervention data on marksmanship accuracy that were unsuitable for inclusion within the meta-analysis. One study resulted in a statistically significant effect in favour of the intervention (Liu et al., 2018), while no significant effects were evident in another study (Shipley & Baranski, 2002) (Table 1).
Decision Making
Four studies provided sufficient information to be included in the meta-analysis of decision making differences between high and low pressure conditions (Landman et al., 2016a; Nibbeling et al., 2014; Nieuwenhuys et al., 2012, 2015). The pooled results revealed a small reduction in decision accuracy for the high, compared to the low pressure condition (SMD: − 0.35; 95% CI: − 0.06, − 0.64; p < 0.019; Figure 4). The level of statistical heterogeneity was low (I2 =0%). Service experience was not a statistically significant moderator in this case (p = 0.223).
Figure 4.
Forest plot of mean difference in incorrect decisions on decision making from low perceived pressure to high perceived pressure shoot, no-shoot scenarios.
Reaction Time
Five studies provided sufficient information to be included in meta-analysis of differences in reaction time between the high and low pressure conditions (Landman et al., 2016a; Nieuwenhuys et al., 2012, 2015; Nieuwenhuys & Oudejans, 2010, 2011). The pooled results indicated a small reduction in reaction time for the high, compared to the low pressure condition (SMD:− 0.32; 95% CI: −0.19, −0.44; p < 0.001; Figure 5). The level of statistical heterogeneity was low (I2 = 0%). Service experience was not a statistically significant moderator for reaction time (p = 0.335).
Figure 5.
Forest plot of mean difference in reaction time in incorrect decisions on shoot, no-shoot targets from low perceived pressure to high perceived pressure scenarios.
Marksmanship Behaviours
The impact of high pressure on marksmanship behaviours that were different to those reviewed above were reported in seven studies. Three studies reported data identifying changes in visual gaze behaviours between low pressure and high pressure conditions. One article reported no statistically significant change in gaze behaviour between visual fixations, and time of fixation, on threat-related locations. Similarly, there was no statistically significant differences in target fixations between shoot and no-shoot scenarios, or between low and high pressure conditions (Nieuwenhuys et al., 2012). A statistically significant decrease in target fixations and a significant increase in fixations on threat-related locations was reported in one article from a low to a high pressure condition (p = 0.016, p = 0.023) (Nieuwenhuys & Oudejans, 2011). In contrast, another study comparing experience levels reported no significant difference between low pressure and high pressure conditions, with a statistically significant increase in threat-related fixations when comparing low, moderate and highly experienced participants (p = 0.049). Time spent on final aiming visual fixations also showed no significance between low pressure and high pressure conditions. However, experienced participants spent significantly longer aiming than did moderate or less experienced participants (p = 0.040) (Landman et al., 2016a). Body postural behaviour was assessed in two studies, with one article reporting participants turning their head away from the target significantly more frequently with their eyes closed significantly longer under the high pressure than the low pressure condition (Nieuwenhuys & Oudejans, 2010). The second article reported that participants exposed significantly less of their body area from a covered firing position during high pressure than the low pressure condition (p = 0.01) (Taverniers & DeBoeck, 2014).
Discussion
The outcomes of this review revealed that high pressure conditions are associated with a reduction in marksmanship, decision making accuracy and reaction time, indicating that increased pressure may bias the performer towards increases in the speed of response, rather than the accuracy of identification and execution. However, it is also evident that experience appears to moderate the impact of pressure on performance. Finally, intervention strategies that involved exposure to increased pressure, the development of cognitive coping strategies, and access to simulated scenario training, facilitated the transfer of training, particularly under high pressure conditions. However, the low number of studies available for the purposes of this meta-analysis and review, despite the broad database search, is also important as it highlights the need for more research to establish the relationship between pressure and outcomes in high risk, high consequence environments.
Effects on Marksmanship Performance
On the basis of the 10 studies analysed, the outcomes of the meta-analysis revealed a statistically significant reduction in marksmanship when participants were required to execute marksmanship skills in high pressure situations, with only two of the total 17 studies failing to reach significance. When this decrease was calculated as a proportion of initial accuracy, the reduction in performance from low to high pressure ranged from between 0.2% to 34.8% with a mean reduction of 14.8% (Figure 2). Across different studies, marksmanship data were recorded using different metrics, including qualitative trainer assessments, deviation from the centre of the target and hits to different parts of the body. However, despite the differences in the measures of marksmanship, the decrease in performance remained consistent in what participants perceived were high pressure environments.
The precise impact of pressure on performance is impossible to determine due to the non-standardised use of recording criteria, variations in threat stimuli, and the level of occupational training. Nevertheless, it remains clear that, as the perception of pressure increases with the relative intensity of threat, individual factors such as experience and prior violent encounters likely moderate the loss of performance that might otherwise be experienced in high pressure situations. Although the results ranged between a decrease of 0.2 to 34.8%, this remains lower than the reported decreases of between 40 to 86% in high pressure, real-world use of force encounters (Morrison & Vila, 1998; Rostker et al., 2008).
The reduction in performance from within the training and simulated reality environment to performance in a real-world context may also be explained through a graduated increase in pressure or the intensity of the threat confronting the officer. Specifically as the level of threat intensity increases, there appears to be a reduction in marksmanship performance accuracy. The nature of this relationship provides conceptual support for the Yerkes–Dodson Law and the more contemporary model involving catastrophe theory that explains the relationship between arousal and performance in tactical-specific performance (Krane, 1992; Stewart & Peregoy, 1983; Yerkes & Dodson, 1908). The decrease in performance from a traditional, low threat range or training environment to high threat, reality-based training has shown an average decrease of 14.8%. However, while limited, the data available from real-world encounters in unopposed, real-life UOF encounters, indicates a decrease of 40%–67% and up to an 86% decrease in performance in an opposed shooting incident (Morrison & Vila, 1998; Rostker et al., 2008; White, 2006). This would indicate a very low transfer of marksmanship training from the training environment to application in the real-world.
Effects of Experience
Increases in experience appeared to be associated with improvements in marksmanship under situations in which there was a high level of perceived pressure. The results suggest that occupational experience may begin to mitigate the negative impact of high perceived pressure environments following 10–13 years of service. With a 1.1% increase with every year of experience. Specific research relating to the role of experience revealed a statistically significant effect of experience on condition, with experienced officers achieving greater accuracy than novice officers under high pressure, with 81.4% and 54.2% accuracy, respectively.
Experienced officers also demonstrated superior decision making with lower, but non-significant rates of incorrect decisions (13% incorrect) compared to novice officers (58% incorrect) under high pressure. The strength of the relationship between experience and increased tolerance to perform under pressure is unclear due to insufficient research. However, a likely factor that explains the difference associated with experienced performers is the increase in the frequency of exposures to threat that is gained from experience that also associated with a reduction in the relative pressure during threat encounters. To assist with increasing the understanding of this realtionship, future research should be directed towards including environmental experience, prior experience to threat and violent encounters, previous occupational training, external training, and competency with firearms.
Effects of Intervention Strategies
The outcomes of intervention studies indicate that several intervention strategies may be effective in reducing the impact of high pressure conditions on marksmanship. Although there is limited data concerning the effectiveness of interventions, reality-based practices appeared to result in similar levels of effectiveness as did imagery and virtual reality–based interventions. A small but similar effect was associated with video based virtual training with an inverse relationship between effect of intervention and experience. When comparing the influence of experience in the context of these studies, the early exposure to contextually relevant stimuli, including the associated pressure, may provide a positive benefit in replicating experience that might be acquired in practice. This is especially evident when considering that the comparative results of the average intervention, incorporating contextually relevant pressure, is associated with an improvement in performance by 10.6%, or the equivalent 9.6 years of experience in service.
Differentiating the impact of reality-based training from imagery and emotionally based interventions also revealed a positive impact of intervention with the adverse impact of high pressure conditions reduced by 15% and 8% for mental imagery training (EI) and mental imagery and emotional training (EEI), respectively (Colin et al., 2014). However, it should be noted that two studies reported mixed results for interventions, one of which related to training Situational Awareness (SA, experience m = 6.09yrs) resulting in no significant changes to performance. However, training in mental imagery (MI, experience m = 15.11 years) was associated with a statistically significant, positive effect.
In combination, these results suggest that interventions may be more effective with inexperienced performers at the earlier stages of their career. However, the limited number of studies available, the lack of a standardised research design, variations in experience and the type of threat stimulus used, and the frequency and rate of exposure, make direct comparisons difficult in determining the optimal intervention strategy. Nevertheless, the results indicate that intervention strategies likely offer an improvement in performance, although further research is required to identify the optimal type and application of potential strategies.
Decision Making, Reaction Time and Behaviours
Although there is limited research focusing on the relationship between decision making, reaction time, and gaze behaviour amongst tactical populations responding to a high threat, it is inevitable that an association exists. For example, amongst the limited research available, the accuracy of decision making was reduced under high pressure, with higher rates of error associated with a greater frequency of false identification and the engagement of a no-shoot target or non-threat target (Nibbeling et al., 2012; Nieuwenhuys et al., 2012, 2015; Nieuwenhuys & Oudejans, 2010, 2011, 2011). A relationship also appears to be evident between increases in perceptual pressure and decreased reaction times to engage under high pressure conditions (Nibbeling et al., 2012; Nieuwenhuys et al., 2012; Nieuwenhuys & Oudejans, 2011). Changes in gaze behaviour evident under high pressure conditions followed a similar trend with participants more likely to focus on potential threat areas and information with less time spent on marksmanship-specific fixations, particularly amongst less experienced participants (Landman et al., 2016b).
Limitations
There are a number of limitations associated with the present study, including the relatively low number of studies available, together with variations in metrics that made it difficult to clearly determine the impact of high and low pressure conditions on marksmanship. There were also challenges in determining the effectiveness of intervention strategies due to variations in the tactical training and policies implemented across organisations and variations in the occupational environments. Therefore, it may be difficult to replicate the relevant and contextual threats within the preparation and training environments to fully analyse the impact of high pressure situations across tactical populations.
Conclusion
The evidence from the current meta-analysis indicates that, under increasing threat conditions, marksmanship performance appears to be negatively impacted, resulting in significant reductions in task execution. This effect is likely due to an increase in the fixation on threat information or given cues, resulting in a higher probability of false identification and a faster reaction time to engage. It is also associated with a reduction in the frequency of gaze fixations on task-relevant marksmanship information and a subsequent reduction in marksmanship accuracy. Experience appears to mitigate the impact of pressure. However, it remains unclear whether performance evident in the training environment translates to performance in real-world, life-threatening situations.
Intervention strategies that have been directed towards exposing participants to contextually relevant threats early in their learning through simulated or virtual training situations appear to elicit improvements in performance. However, these evaluations have been limited to training environments and lack real-world data to support the transfer of training into the operational environment. In the absence of standardised reporting, training and threat stimulus procedures, it is difficult to determine with certainty, the impact of high pressure situations on performance and the most effective interventions to enhance performance in relevant environments. Consequently, further research should be directed towards exploration the integration between contextually relevant training and high pressure environments early in the process of motor skill acquisition, particularly to address the relative paucity of available evidence on the impact of pressure on marksmanship performance in such high consequence environments.
Key Points
Marksmanship accuracy decreases by an average of 10.8% when performed under increased pressure
Increased pressure is associated with a small negative effect on decision making accuracy and faster reaction times for incorrect decisions
Experience constitutes a mitigating factor from the negative impact of pressure on marksmanship accuracy with an average of 1.1% improvement per year of service.
Early exposure to contextually relevant information and pressure, in place of traditional static, range-based marksmanship training, is associated with an increase in the transfer of training by an average of 10.6%.
Biography
Daniel Cooper, Macquarie University (MQ). Master of Research, Macquarie University, 2019.
Joel Fuller, Macquarie University. Doctor of Philosophy in Physiotherapy, University of South Australia, 2016.
Mark Wiggins, Macquarie University. Doctor of Philosophy in Psychology, University of Otago, 2001.
Jodie Wills, Macquarie University, Doctor of Philosophy in Biomechanics, Macquarie University, 2020.
Luana Main, Deakin University. Doctor of Philosophy in exercise physiology and Psychology, The University of Western Australia, 2010 Associate
Tim Doyle, Macquarie University. Doctor of Philosophy in Biomechanics, Edith Cowan University, 2006
Footnotes
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors acknowledge Macquarie University for providing PhD scholarship funding for this research.
ORCID iDs
Daniel Cooper https://orcid.org/0000-0002-0902-2746
Mark W. Wiggins https://orcid.org/0000-0002-6422-9475
Jodie A. Wills https://orcid.org/0000-0001-5397-4288
References
- Anderson G. S., Litzenberger R., Plecas D. (2002). Physical evidence of police officer stress. Policing: An International Journal of Police Strategies & Management, 25, 399–420. [Google Scholar]
- Arble E., Daugherty A. M., Arnetz B. (2019). Differential effects of physiological arousal following acute stress on police officer performance in a simulated critical incident. Frontiers in Psychology, 10, 759. 10.3389/fpsyg.2019.00759 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Babineau J. (2014). Product review: Covidence (systematic review software). Journal of the Canadian Health Libraries Association/Journal de l'Association des bibliothèques de la santé du Canada, 35(2), 68–71. 10.5596/c14-016 [DOI] [Google Scholar]
- Burrows C. (2007). Critical decision making by police firearms officers: A review of officer perception, response, and reaction. Policing, 1(3), 273–283. 10.18261/issn.2703-7045-2020-02-03 [DOI] [Google Scholar]
- Colin L., Nieuwenhuys A., Visser A., Oudejans R. R. D. (2014). Positive effects of imagery on police officers' shooting performance under threat. Applied Cognitive Psychology, 28(1), 115–121. 10.1002/acp.2972 [DOI] [Google Scholar]
- Dunsmoor J. E., Kroes M. C. W., Braren S. H., Phelps E. A. (2017). Threat intensity widens fear generalization gradients. Behavioral Neuroscience, 131(2), 168–175. 10.1037/bne0000186 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Friedland N., Keinan G. (1992). Training effective performance in stressful situations: Three approaches and implications for combat training. Military Psychology, 4(3), 157–174. 10.1016/j.ijpsycho.2018.03.017 [DOI] [Google Scholar]
- Gamble K. R., Vettel J. M., Patton D. J., Eddy M. D., Caroline Davis F., Garcia J. O., Spangler D. P., Thayer J. F., Brooks J. R. (2018). Different profiles of decision making and physiology under varying levels of stress in trained military personnel. International Journal of Psychophysiology, 131, 73–80. 10.1016/j.ijpsycho.2018.03.017 [DOI] [PubMed] [Google Scholar]
- Giessing L., Frenkel M. O., Zinner C., Rummel J., Nieuwenhuys A., Kasperk C., Brune M., Engel F. A., Plessner H. (2019). Effects of coping-related traits and psychophysiological stress responses on police recruits' shooting behavior in reality-based scenarios. Frontiers in Psychology, 10(1), 1523. 10.3389/fpsyg.2019.01523 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Gould D., Petlichkoff L., Simons J., Vevera M. (1987). Relationship between Competitive State Anxiety Inventory-2 subscale scores and pistol shooting performance. Journal of Sport Psychology, 9(1), 33–42. [Google Scholar]
- Higgins J. P., Thomas J., Chandler J., Cumpston M., Li T., Page M. J., Welch V. A. (2019). Cochrane handbook for systematic reviews of interventions. John Wiley & Sons. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Higgins J. P. T., Thompson S. G., Deeks J. J., Altman D. G. (2003). Measuring inconsistency in meta-analyses. Bmj, 327(7414), 557–560. 10.1136/bmj.327.7414.557 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hopkins W. G. (2000). A new view of statistics. : Internet Society for Sport Science. [Google Scholar]
- Krane V. (1992). Conceptual and methodological considerations in sport anxiety research: From the inverted-U hypothesis to catastrophe theory. QJM : Monthly Journal of the Association of Physicians, 44(1), 72–87. 10.1080/00336297.1992.10484042 [DOI] [Google Scholar]
- Landman A., Nieuwenhuys A., Oudejans R. R. (2016. b). The impact of personality traits and professional experience on police officers' shooting performance under pressure. Ergonomics, 59(7), 950–961. 10.1080/00140139.2015.1107625 [DOI] [PubMed] [Google Scholar]
- Landman A., Nieuwenhuys A., Oudejans R. R. D. (2016. a). Decision-related action orientation predicts police officers' shooting performance under pressure. Anxiety, Stress, and Coping, 29(5), 570–579. 10.1080/10615806.2015.1070834 [DOI] [PubMed] [Google Scholar]
- Lieberman HR., Niro P., Tharion WJ., Nindl BC., Castellani JW., Montain S J. (2006). Cognition during sustained operations: Comparison of a laboratory simulation to field studies. Aviation, Space, and Environmental Medicine, 77(9), 929–935. [PubMed] [Google Scholar]
- Liu Y., Mao L., Zhao Y., Huang Y. (2018). Impact of a simulated stress training program on the tactical shooting performance of SWAT trainees. Research Quarterly for Exercise and Sport, 89(4), 482–489. 10.1080/02701367.2018.1526003 [DOI] [PubMed] [Google Scholar]
- Moher D., Liberati A., Tetzlaff J., Altman D. G. (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. Annals of Internal Medicine, 151(4), 264–269. 10.7326/0003-4819-151-4-200908180-00135 [DOI] [PubMed] [Google Scholar]
- Morrison G. B., Vila B. J. (1998). Police handgun qualification: Practical measure or aimless activity? Policing. An International Journal of Police Strategies & Management. [Google Scholar]
- Murray N. P., Janelle C. M. (2003). Anxiety and performance: A visual search examination of the processing efficiency theory. Journal of Sport & Exercise Psychology, 25(2), 171–187. 10.1123/jsep.25.2.171 [DOI] [Google Scholar]
- Nibbeling N., Oudejans R. R. D., Daanen H. A. M. (2012). Effects of anxiety, a cognitive secondary task, and expertise on gaze behavior and performance in a far aiming task. Psychology of Sport and Exercise, 13(4), 427–435. 10.1016/j.psychsport.2012.02.002 [DOI] [Google Scholar]
- Nibbeling N., Oudejans R. R. D., Ubink E. M., Daanen H. A. M. (2014). The effects of anxiety and exercise-induced fatigue on shooting accuracy and cognitive performance in infantry soldiers. Ergonomics, 57(9), 1366–1379. 10.1080/00140139.2014.924572 [DOI] [PubMed] [Google Scholar]
- Nieuwenhuys A., Cañal-Bruland R., Oudejans R. R. D. (2012). Effects of threat on police officers' shooting behavior: Anxiety, action specificity, and affective influences on perception. Applied Cognitive Psychology, 26(4), 608–615. 10.1002/acp.2838 [DOI] [Google Scholar]
- Nieuwenhuys A., Oudejans R. R. D. (2010). Effects of anxiety on handgun shooting behavior of police officers: A pilot study. Anxiety, Stress, and Coping, 23(2), 225–233. 10.1080/10615800902977494 [DOI] [PubMed] [Google Scholar]
- Nieuwenhuys A., Oudejans R. R. D. (2011). Training with anxiety: short- and long-term effects on police officers' shooting behavior under pressure. Cognitive Processing, 12(3), 277–288. 10.1007/s10339-011-0396-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- Nieuwenhuys A., Savelsbergh G. J. P., Oudejans R. R. D. (2015). Persistence of threat-induced errors in police officers' shooting decisions. Applied Ergonomics, 48(5), 263–272. 10.1177/1747021820985819 [DOI] [PubMed] [Google Scholar]
- Oudejans R. R. D. (2008). Reality-based practice under pressure improves handgun shooting performance of police officers. Ergonomics, 51(3), 261–273. 10.1080/00140130701577435 [DOI] [PubMed] [Google Scholar]
- Pluye P., Robert E., Cargo M., Bartlett G., O’cathain A., Griffiths F., Boardman F., Gagnon M., Rousseau M. (2011). Mixed methods appraisal tool (MMAT) version 2011. Proposal: A mixed methods appraisal Tool for systematic mixed studies reviews. [Google Scholar]
- Rostker B. D., Hanser L. M., Hix W. M., Jensen C., Morral A. R. (2008). Evaluation of the New York City Police Department firearm training and firearm-discharge review process. Rand Corporation. [Google Scholar]
- Roy H., Wasylyshyn N., Spangler D. P., Gamble K. R., Patton D., Brooks J. R., Garcia J. O., Vettel J. M. (2019). Linking emotional reactivity between laboratory tasks and immersive environments using behavior and physiology. Frontiers in Human Neuroscience, 13(1), 54. 10.3389/fnhum.2019.00054 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ryan R. (2016). Cochrane Consumers and communication group: Meta-analysis. Cochrane Consumers and Communication Review Group. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Saus E.-R., Johnsen B. H., Eid J., Riisem P. K., Andersen R., Thayer J. F. (2006). The effect of brief situational awareness training in a police shooting simulator: An experimental study. Military Psychology, 18(sup1), S3–S21. 10.1207/s15327876mp1803s_2 [DOI] [Google Scholar]
- Shipley P., Baranski J. V. (2002). Police officer performance under stress: A pilot study on the effects of visuo-motor behavior rehearsal. International Journal of Stress Management, 9(2), 71–80. 10.1023/A:1014950602826 [DOI] [Google Scholar]
- Stewart I. N., Peregoy P. L. (1983). Catastrophe theory modeling in psychology. Psychological Bulletin, 94(2), 336–362. 10.1037/0033-2909.94.2.336 [DOI] [Google Scholar]
- Taverniers J., De Boeck P. (2014). Force-on-force handgun practice. Human Factors: The Journal of the Human Factors and Ergonomics Society, 56(2), 403–413. 10.1177/0018720813489148 [DOI] [PubMed] [Google Scholar]
- Taverniers J., Suss J. (2019). A user-centred assessment of a less-lethal launcher: the case of the FN 303 in a high-pressure setting. Ergonomics, 62(9), 1162–1174. 10.1080/00140139.2019.1626916 [DOI] [PubMed] [Google Scholar]
- White M. D. (2006). Hitting the target (or not): Comparing characteristics of fatal, injurious, and noninjurious police shootings. Police Quarterly, 9(3), 303–330. 10.1177/1098611105277199 [DOI] [Google Scholar]
- Yerkes R. M., Dodson J. D. (1908). The relation of strength of stimulus to rapidity of habit-formation. Punishment: Issues and Experiments, 18(5), 27–41. [Google Scholar]