A total of 12 males with FAI participated in the study, and all provided informed consent. The Soochow University Ethics Committee Board approved this study. The demographic information of the participants is shown in Table 1. Self-reported instability and function were determined using the ankle joint functional assessment tool (AJFAT) (Ross et al., 2008; Ibrahim and Abdallah, 2020). Participants’ inclusion criteria were as follows: 1) one unstable extremity, with AJFAT scores ≤26 (Delahunt et al., 2006; de Noronha et al., 2008; Kim et al., 2021; Rosen et al., 2021); 2) one stable extremity, with AJFAT scores >26; 3) no vestibular, visual, or neurological disease; 4) no lower extremity injuries for ≥6 months; and 5) only one side of the lower extremities suffering from FAI.

Participants had had no strenuous activities or muscle fatigue within 24 h before the experiment. Initially, participants warmed up on a treadmill at 2.2 m/s for 3–5 min and stretched. In a single session, each participant with FAI wore experimental shorts and shoes and completed random drop landing trials without initial speed under four conditions, consisting of all combinations of each leg (unstable and stable extremities) and two conditions (VP and VD). Subsequently, participants completed a tiptoe slide from the platform’s edge at a height of 30 cm, with instructions to place the hands akimbo to reduce the influence of arm swing. Two examiners provided safety assistance throughout the experiment. Landing on a single leg and bending knees instinctively and toe-first is illustrated in Figure 1. The two visual conditions were tested randomly. The participants were required to wear black blindfolds when testing under VD conditions (Shoja et al., 2020).

Each condition repetition was recorded by one examiner, who was blinded to the AJFAT scores, with a rest of 30 s between landings until the participant had completed three successful trials. Successful trials were defined as those performed without any extraneous movement during the 3 s prior to the hop and with the participant subsequently remaining upright during the 2 s after the drop landing, with the tested extremity within the boundary.

Lower extremities’ kinetic data were collected at 100 Hz using a 3D motion capture system equipped with eight infrared cameras (Vicon Motion Analysis, United Kingdom) by tracking 18 infrared-reflective balls (reflective markers) of 14 mm diameter. Ground reaction force data were sampled at 1000 Hz using a 3D Kistler force measuring platform (model: 9287B, 90 cm × 60 cm×10 cm, Kistler, Inc., Switzerland). The plug-in gait was modeled using Visual3D (Version 6, C-Motion, Inc., United States) to calculate joint kinetics.

The following kinetic data were analyzed: ground reaction force (GRF, including vertical GRF [vGRF], medial GRF [mGRF], and lateral GRF [lGRF]), time to peak GRF (T_GRF, including time to peak vertical GRF [T_vGRF], time to peak medial GRF [T_mGRF], and time to peak lateral GRF [T_lGRF]), loading rate (LR), joint torque in the sagittal plane and the frontal plane, hip joint vertical length variation (ΔL), lower extremity stiffness (K_leg), and ankle stiffness. LR was calculated using Eq. 1. K_leg was calculated using Eq. 2. L R = V G R F / T _ v G R F , (1) K l e g = v G R F ∕ Δ L . (2)

The normality of each variable was determined using the Shapiro–Wilk test. Descriptive and dependent measures were calculated as mean ± SD. Dependent measures were used to examine the interaction effects of groups (unstable and stable extremities of patients with FAI) and conditions (VP and VD) that served as the repeated measure. Simple effect analysis was performed when significant interactions were found. If the simple effect analysis suggested a statistically significant difference in a metric in both groups between VP and VD, then the difference between VP and VD in each group was determined using the independent samples t-test. All statistical analyses were performed using a statistical software package (IBM SPSS Statistics, IBM, NY, United States), and the level of statistical significance was set at 0.05.

The experimental data are shown in Table 2. No significant interaction effects in the group were detected for GRF, T_GRF, and LR. Significant effects of the condition were detected for vGRF (p = 0.041), T_vGRF (p = 0.001), T_lGRF (p < 0.001), and LR (p = 0.001). A significant effect in the group was detected for T_lGRF (p = 0.004).

The results show that the T_lGRF was earlier in the unstable extremity than in the stable extremity, suggesting that the protective strategy for the unstable extremity may need to be sufficiently activated. Our results are similar to those of previous studies of patients with FAI during single-extremity jump landings, in that the severity of ankle instability correlates with the earlier frontal stabilization time (Ross et al., 2008). Some scholars (Lephart et al., 1998; Svinin et al., 2019) have suggested mechanisms responsible for the earlier T_lGRF in the unstable extremities and that the severity of ankle instability correlates with the shorter time to frontal stabilization. The proprioceptors in the ankle provide joint position sensation and motion signals along the sensory nerves to the spinal cord, where other sensory neurons may process and then transmit them to the brain for further processing. The proprioceptors damaged by recurrent sprains in the unstable extremities of patients with FAI may cause delayed signal transmission resulting in abnormal feedback (Lephart et al., 1998; Eechaute et al., 2009; Song et al., 2016; Ludvig et al., 2017; Ibrahim and Abdallah, 2020). This abnormal feedback may not be able to activate the protective strategy for landing in a timely, effective manner to reduce the injury risk. Moreover, studies (Horak et al., 1990; Patla et al., 2002; Chaudhry et al., 2005; Delahunt et al., 2006; Ergen and Ulkar, 2008; Kabbaligere et al., 2017; Ibrahim and Abdallah, 2020) have suggested that movement impairment could be modified by correcting muscle activations. Impaired proprioception causes a delay in the conduction of the afferent signal, and the consequent delay in the efferent signal causes a delay in the corrective muscle activation, which leads to a higher injury risk in the unstable extremity.

Higher hip extensor and ankle torque in the sagittal and frontal planes were revealed in the unstable extremity as compared with the stable extremity. Our results are similar to research on the effect of proprioceptive training on postural stability, that is, they show that a higher hip extensor torque increases ankle sprain risk (Hietamo et al., 2021). Joint torque represents the buffering function of energy absorption on the ground reaction force through muscle strength. The higher the hip extensor torque is, the less GRF is absorbed through the muscles, which means that the role of muscle strength in reducing ground reaction force is minor; the higher the load is, the less the extremity is buffered during drop landing. Based on kinetic chain theories, insufficient hip muscle function increases the likelihood of uncontrolled ankle displacements and ankle injury.

Previous studies have showed poor postural stability in ankle sprain patients (Delahunt et al., 2006; Doherty et al., 2014). Postural stability decreases when the unstable extremity is forced to support itself during drop landing. Corresponding to our results, studies have found that K_leg is larger in the unstable extremity than in the stable extremity. A study related to ankle and knee proprioception found that postural swing and center of body mass changes increase among ankle-injured individuals compared with the uninjured group and that a larger K_leg response might increase postural swing and center of body mass changes during movement (Lephart et al., 1998). Some studies (Ito and Gomi, 2020) have suggested the mechanisms of postural swing and center of body mass changes, in which impaired proprioception in the unstable extremity makes accurate adjustment signal transmission for joint position sense abnormal, then blocks joint pressure and tension feedback and further affects the nerve fiber processing of sensory information, thus aggravating postural instability.

The greater ankle stiffness in the stable extremity compared with the unstable extremity also supports the view that the unstable extremity has a higher injury risk. Ankle stiffness reduces energy consumption caused by joint activity through the rigid lever, and stored energy is released to cushion the ground impact force; ankle stiffness can prevent joint instability during drop landing (Gao et al., 2022). Ludvig et al. (2017) has suggested mechanisms for deficient levels of ankle stiffness, in which proprioception is impaired in the unstable extremity, leading to the muscle not activating. The ligaments and tendons become slack due to recurrent ankle sprain, resulting in energy consumption caused by joint activity. The energy available for cushioning the ground impact force is reduced, causing higher injury risk.

The result of increased vGRF and LR under VD in our paper shows the higher impact force and injury risk in the lower extremities of patients with FAI during drop landing with VD (Rosen et al., 2013). Under VD, the protective strategy may not be sufficiently activated to buffer load and reduce injury risk. Our results show that earlier T_vGRF and T_lGRF appear under VP than under VD. Previous studies have found that visual conditions affect postural stability (Zipori et al., 2018), and our research further found that VD increases the risk of injury in patients with FAI. Under VD, the brain is unable to conduct environmental image analysis and deliver control signals generated by visual information transfer to achieve a series of postural control outputs, such as muscle tension adjustment and joint angle change.

Moreover, our results show deficits in postural control in patients with FAI, revealing that hip and knee torque in the sagittal and frontal planes increase with VD and that the hip and knee strategies mobilize when the ankle is unresponsive. This phenomenon shows that the dynamic defense system of the joint is further limited due to the proprioception impairment caused by the recurrent ankle sprain, which leads to abnormal ankle strategies in postural control (Alghadir et al., 2020). In our paper, neither the unstable nor the stable extremities buffered the load through ankle torque. This result could reflect the inability of the unstable extremity to overcome the loss of two sources of sensory information (Song et al., 2016): proprioception, which is altered due to repeated ankle sprain, and visual inputs, which are lost under VD. Palm et al. (2009) studied the effects of vision and hearing on postural stability and suggested that other senses, such as proprioception and vestibular sense, can compensate for postural instability when visual information input is abnormal. Zipori et al. (2018) also suggested that proprioception dependence in individuals increases when visual information inputs are abnormal. However, our paper found that even the stable extremity, without proprioception impairment, does not show special performance compared with the unstable extremity during VD. Therefore, as for the stable extremity, proprioception dependence is reduced under the long-term influence of the unstable extremity deficits (Eechaute et al., 2009; Song et al., 2016).

Our results show that K_leg is larger under VD than under VP, indicating an increased instability and injury risk in the lower extremities. As mentioned previously, a larger K_leg may increase center of body mass changes and postural swing during drop landing. Eechaute et al. (2009) have supported this view and suggested that the center of body mass is used to maintain balance during motion. Palm et al. (2009) suggested that when visual control information decreases, postural control ability decreases and the lower extremity injury risk increases accordingly. In addition, muscle stiffness regulation is the first reflex mechanism to adjust posture stability. Neurofeedback modulates systems and adjusts muscle tone to adapt to changes in the spatial environment. A possible explanation for the lower ankle stiffness under VD could be that missed visual inputs impede information transmission, which activates the muscles (McKeon and Hertel, 2008; Ibrahim and Abdallah, 2020). Miller et al. (2018) have further reported the importance of visual feedback, which may allow users to receive external joint position cueing, which is also conducive to sending the protection strategy’s activation signal. The motion system subsequently manipulates ankle positioning and stiffness. When patients with FAI lose their normal perception of the external environment, ankle stiffness cannot be adjusted correctly.