3.2.1 Participants.

Recruitment was predominantly through university student cohorts or professional/elite sports teams, with participants aged in their late teens or twenties; 2 of these studies involved participants with a history of ankle injury, but participants were otherwise fit and healthy. One study explored the immediate effect of SVT on people with a neurological condition (multiple sclerosis; MS) [42]. Both male and female were recruited to the majority of studies (n = 12) except for 3 studies that recruited only male participants [18,43,44]. Three papers did not report specific gender characteristics [9,15,45]. (Table 1).

3.2.2 Eligibility criteria.

Seven studies recorded no exclusion criteria for their participants [15,43,44,46–49]. The remaining 8 studies identified seizures/epilepsy, migraines, and/or a history of neurological disorder as reasons for individuals not to take part (Table 1). Inclusion criteria were generally focused on level of sporting ability except in the case of the later studies where participants were recruited based on an underlying health condition [42] or a musculoskeletal injury [50].

3.3.1 Sample size.

Participant sample sizes varied greatly between the studies with numbers ranging from a pilot study of 3 [44] to the earliest study by Appelbaum and colleagues who had 157 participants [9].

3.3.2 Control groups.

All 15 studies included a control group. In an attempt to reduce bias, 4 of these studies had control participants wearing stroboscopic eyewear with the lenses in a permanently transparent (nonstrobing) state [9,15,46] and 1 study used lens-free glasses [42]. Wilkins and Gray applied a constant fast strobe setting to their control group participants in an attempt to combat the potential for differing motivation and effort levels compared to the variable SVT group [48].

3.4.1 Stroboscopic devices.

Five different commercially available brands of eyewear device were used to provide stroboscopic visual disruption across the 15 studies (Table 2). The most common devices to feature in the studies were the Nike Vapor Strobe (Nike, Beaverton, OR) (n = 7) [9,15,18,43,44,49] and the newer version of the eyewear, the Senaptec Strobes (n = 4) [40,41,47,50,51]. Other devices included the Visionup Strobe glasses (n = 1) [42], Plato goggles (n = 2) [46,48], and MJ Impulse (n = 1) [16].

3.4.2 Strobe settings.

Of the 15 studies, 7 studies used a single fixed frequency to deliver their stroboscopic training. There were a number of different set frequency settings employed in these studies: 4 Hz [18,46,49], 30 Hz (duty ratio: 50%) [42], 1.8 Hz [45], 0.1 Hz [51], and 120 Hz [47] (Table 2). Seven studies adopted a variable frequency approach [9,16,40,43,44,48,50]. Appelbaum and colleagues used both a fixed and a modulated strobe frequency depending on the drill [15].

3.4.3 Frequency and duration of interventions.

Four studies conducted a single, one-off session of SVT lasting from 5 minutes [18,46,49] to a maximum of 50 minutes [42]. The remaining studies (n = 11) provided multiple sessions of SVT over periods of up to 10 weeks with Wilkins and colleagues reporting a maximum accumulated total of 635 minutes of strobe time [44] (Table 2). There was also significant variability of training duration within studies. For example, in Appelbaum and colleagues’s study, total strobe time ranged from 54 to 300 minutes [15].

3.4.4 Training intervention.

Strobe glasses were worn while undertaking physical tasks such as throwing and catching drills (n = 4) [9,15,42,48] or sport-specific activities such as skating and passing drills in ice hockey [43], goalkeeping drills in football [44], or court-based badminton drills [16,40,41]. Two studies applied SVT while participants undertook activities specifically designed to challenge their balance. Kim and colleagues used a multimodal exercise programme performed over 6 weeks, while Symeonidou and colleagues used a single session of balance beam walking [50,51]. Four studies had participants undertaking SVT while performing machine or screen-based reaction time tasks [18,45,46,49], and 1 study had participants practicing surgical tasks under stroboscopic conditions [47] (Table 2).

3.4.5 Posttraining test points.

Table 3 demonstrates that across the reviewed studies, 14 posttests occurred immediately after the training with 6 of those studies also recording retention-test data ranging from delays of 10 minutes [18,49], 24 hours [15], 10 days [18,49], 4 weeks [44], and 6 weeks [40]. Mitroff and colleagues conducted a single posttest 24 hours after their last training session [43].

3.5.1 Outcome measures.

The majority of studies (n = 11) explored SVT outcomes in relation to either a specific visual ability such as reaction timing or hand–eye coordination [18,40,49], visual memory [15], or visual processing [9,16,46,48], or a battery of visual outcomes [42,44] (Table 4). Of the studies that addressed visual ability, some utilised a stand-alone unit such as a Bassin Timer to measure a single visuomotor outcome (anticipatory timing) [49], whereas other studies use computerised software (e.g., MATLAB/Psychophysics Toolbox/NeuroTrax Mindstreams) [9,15,16,42,45], online assessment tools [44,48] or specialist sensory assessment devices [18,40]. Hülsdünker and colleagues used neurophysiologic investigations (via electroencephalography) to identify modulations in the participants’ visuomotor performance and visual perception speed [16].

Of those studies, a further 3 also measured sport-specific performance tasks [16,40,44], and 1 captured gait and balance metrics as a secondary outcome [42]. Four studies included only motor performance measures; dynamic balance [50,51], precision of puck placement in ice hockey [43], and performance of suturing task [47] (Table 4).

3.5.2 Interpretation of outcomes.

SVT had a positive effect on several different visual and motor performance outcomes (Table 2), for example, central field motion sensitivity [9], short-term memory capacity [15], processing speed [48], and eye–hand coordination [18]. Conversely, SVT was found to have no significant effect on outcomes that were based on more sustained visual stimuli or stimuli appearing in the peripheral visual field [9].

4.2.1 Stroboscopic devices.

There was a lack of standardisation of stroboscopic instrumentation employed across the studies. The 5 devices used in the studies all had different manufacturer operating levels, so there was no consistent method of reporting the frequency of the strobe effect or “blink rate.” For example, the Senaptec Strobes operate at a level between 1 and 8, with level 1 representing the fastest/easiest setting at a frequency of 6 Hz (calculated according to the number of transparent-opaque cycles per second) and level 8 representing the slowest/hardest setting at a frequency of 1 Hz [52]. In comparison, the Visionup strobes are adjustable by 10 levels whereby level 1 equates to 1 Hz (based on the frequency of the opaque phase), and level 10 equates to 150 Hz [53]. Even for studies utilising the same strobe device, reporting of the strobe setting varied. For example, Kim and colleagues refer to the operating levels of the Senaptec Strobes (i.e., levels 1 to 8) [50], whereas Hülsdünker and colleagues record frequency settings (i.e., 15 Hz to 8 Hz) [40].

The illuminance value (i.e., degree of light transmission provided through the lenses; measured in lux) was only measured by Bennett and colleagues [45], in relation to the Nike Vapor Strobes and the PLATO goggles, but was not reported in any of the other papers in this review… The Visionup and MJ Strobe glasses have the option to adjust the “duty ratio” (or level of brightness of the lenses), to create darker conditions and, therefore, a more challenging visual environment.

Direct comparisons between studies using different brands of eyewear are therefore confounded by the variation in functionality of the different devices, description of lens characteristics, luminance values, and methods of reporting. This was highlighted by Wilkins and Gray who suggested that their use of PLATO goggles (as opposed to the Nike Vapor Strobes) may have contributed to the discrepancy in their findings with previous research [48]. It is evident from this review that more clarity around strobe frequencies and setting levels is required (both from manufacturers of the devices and researchers themselves) to provide clear understanding and accurate interpretation for future clinical application.

4.2.2 Strobe settings.

There was a lack of standardisation with regard to (1) the frequency of strobe settings and (2) the use of fixed versus variable strobe frequency settings. Smith and Mitroff examined the effect of SVT delivered at a fixed frequency of 4 Hz (Nike Vapor Strobes: level 3, 100 ms clear: 150 ms opaque) and found that participants in the intervention group had significantly better anticipation immediately after a single session of SVT [49]. Subsequent studies have since used these findings to justify their selection of strobe rate settings [18,46]. The most recent study by Symeonido and Ferris applied a much lower fixed frequency visual perturbation to participants as they walked on a treadmill-mounted balance beam for 30 minutes [51]. Balance performance, in number of step-offs of the beam, improved by 78% for the SVT group on the same day of the training. While the earlier studies in this review support a link between visual stimuli and motor learning, Symeonidou and Ferris attributed their much larger training effect to their use of lower frequency strobe settings [51].

For the 8 studies that adopted a variable frequency approach, training protocols generally started on the easiest (fastest) strobe setting before the strobe rate was reduced in response to the correct execution of training drills or to increase task difficulty over time. The exact duration with which participants trained at each strobe level was poorly documented in the majority of studies, making it difficult to extrapolate the findings (Table 2).

The study by Wilkins and Gray was the only one to directly compare training with a fixed strobe rate and a variable strobe rate [48]. Motion in depth sensitivity significantly improved posttraining, but no significant differences were found between the fixed and the variable rate groups. Further work is therefore required not only to determine whether there is an optimal frequency of strobe setting required for both visual and motor learning to occur but also to establish whether the effects of SVT are greater (or indeed reduced) if participants remain on a fixed strobe frequency [1].

4.2.3 Frequency and duration of interventions.

This review demonstrates that there was significant variability regarding the frequency, duration, and number of SVT interventions carried out within the studies (a finding that has been reported across other forms of vision therapy) [54,55]. Interventions ranged from a single session of SVT to multiple sessions over 10 weeks (Table 2). Further complicating interpretation in this review is the lack of clarity provided about the intervention protocols in some papers. For example, Zavlin and colleagues found that SVT had a significant positive effect on the performance of surgical suturing tasks by medical students after 5 sessions of training, but no detail is provided on how long each training session lasted [47]. This makes it difficult to make comparisons between studies because the exact duration of SVT is not known.

In this review, immediate visual improvements were reported after a single session lasting as little as 5 to 7 minutes in some studies [18,49]. This is a significantly lower duration than the 4- to 8-week length of visual training that has been shown to be effective in previous nonstroboscopic research involving athletes [5,56,57]. In contrast with the findings of Smith and Mitroff and Ellison and colleagues [18,49], however, 5 minutes of SVT was not sufficient to improve time to collision judgements of approaching objects in Braly and colleagues’ study [46]. The differing protocols in this review mean that currently no conclusions can be drawn regarding the optimal frequency and duration of SVT that is necessary for a successful outcome in any population.

4.2.4 Training interventions.

The lack of standardised training interventions across the studies in this review reflects the fact that SVT allows for training to take place in a variety of “real-world” contexts in addition to more controlled laboratory-based settings. Furthermore, the reporting of the precise details of the training interventions were vague in a number of the studies due to the logistical constraints imposed by using athletic populations, which limits interpretation of their findings [9,15,43]. While some studies using athletic populations did provide a detailed description of the SVT intervention (for example, Hülsdünker and colleagues described 2 badminton-specific protocols), making inferences on these studies remains challenging due to the sport-specific nature of the interventions provided.

Nike developed a series of videos showing exercises for SVT in 2010s. These mostly comprised variations on simple ball catching tasks (such as the wall-ball catch, the front catch, the turn and catch, and the power ball drop). These exercises formed the basis of interventions for 3 of the reviewed studies [42,44,48]. Kim and colleagues also used throwing/catching exercises alongside static and dynamic balance tasks as part of their SVT intervention [50]. Unlike the study by Shalmoni and Kalron who implemented ball drills intended for athletes in an intervention for people with MS [42], to our knowledge, Kim and colleagues’ study is the first to demonstrate how SVT might be used to increase the challenge of standardised exercises, thereby further improving sensorimotor control [50]. Their intervention is supported by the findings of Symeonido and colleagues, who also used SVT to enhance the effects of balance training [51].

Further work is required to determine how best to proceed with the introduction of stroboscopic training protocols in clinical populations. Evidence from this review suggests that one approach would be to use strobe glasses as an adjunct to existing rehabilitation practices whereby interventions are composed of standardised exercise programmes performed with and without stroboscopic glasses.

4.2.5 Posttraining test points.

Posttests were carried out within 24 hours in all the studies, providing evidence of the immediate effects of SVT on a range of outcomes. Retention-test data, which are arguably of more interest in relation to the clinical application of SVT, were collected in only 4 of the studies, which limits our understanding of the long-term effectiveness of SVT in any population. Future studies are required to determine whether the effects of SVT last beyond 4 to 6 weeks and, indeed, whether the duration and frequency of SVT affects retention.