Backward Locomotion as a Novel Strategy for Enhancing Obesity Management

Featured Application

This study’s findings on the acute effects of forward and backward locomotion training in individuals with a BMI ≥ 30 provide practical recommendations for designing personalized exercise programs. Backward locomotion, which elicits higher cardiorespiratory, metabolic, and muscle activation responses than forward locomotion, can be integrated into fitness programs to increase energy expenditure in obese populations. This approach may improve cardiorespiratory fitness and metabolic health while potentially reducing joint stress, making backward locomotion suitable for rehabilitation or weight management programs. Exercise professionals can utilize these insights to create individualized protocols, adjusting speed and direction to optimize physiological benefits while considering the higher perceptual demands of backward locomotion. The results also support the development of therapeutic exercise interventions to improve overall health outcomes in obese individuals.

Abstract

Obesity is associated with reduced cardiorespiratory fitness and altered metabolic responses. However, the acute effects of forward and backward locomotion training in individuals with a body mass index (BMI) ≥ 30 remain underexplored. This study investigated this population’s cardiorespiratory, metabolic–perceptual, and muscle electromyography (EMG) responses to forward and backward locomotion at different speeds. Twenty-eight male participants were divided into four seven-member groups, following a randomized crossover design with a Latin Square-like counterbalancing approach. Participants completed four 10 min walking conditions (3 km/h forward, 3 km/h backward, 4 km/h forward, and 4 km/h backward) on separate days, with cardiorespiratory parameters (e.g., VO₂, VCO₂, and heart rate), metabolic responses (e.g., lactate and energy expenditure), and lower-limb muscle EMG activity measured. Statistical analysis using two-way repeated measures (MANOVA) revealed significant direction effects (p < 0.05) on VO₂, VCO₂, heart rate, energy expenditure, Borg RPE, final lactate, and the EMG activity of quadriceps, hamstrings, and tibialis anterior, but not on pre-lactate or soleus activity (p > 0.05). These findings provide valuable insights for optimizing exercise programs in obese individuals, supporting tailored movement strategies to enhance physiological outcomes.

1. Introduction

Obesity, defined by a body mass index (BMI) ≥ 30 kg/m², is a critical global public health concern linked to elevated risks of Type 2 diabetes, cardiovascular diseases, and hypertension [1]. According to the World Health Organization (WHO), obesity accounts for 80% of Type 2 diabetes cases, 35% of ischemic heart disease cases, and 55% of hypertension cases in Europe, contributing to over 1 million deaths annually [2]. In Türkiye, the prevalence of obesity is notably high, with 20.2% of adults classified as obese in 2022, marking the highest rate in Europe [2,3]. The global economic burden of obesity is substantial, with annual costs estimated at USD 2 trillion, equivalent to 2.8% of global economic activity [4]. Effective obesity management, aiming for a 10% weight loss over six months, is essential to mitigate these health and economic impacts. Exercise, combined with energy restriction, is a cornerstone of obesity treatment; however, obese individuals often encounter physical and perceptual barriers to sustained physical activity [5]. Walking is a widely recommended low-impact exercise, but backward locomotion, which requires greater coordination, may provide distinct physiological benefits, such as increased cardiopulmonary demands and reduced joint stress compared to forward locomotion [6,7]. Despite these potential advantages, the acute effects of backward locomotion in obese individuals remain largely unexplored, representing a significant gap in the literature [8].

This study aims to compare the acute cardiorespiratory, metabolic, and electromyography (EMG) responses of forward and backward locomotion in individuals with a BMI ≥ 30 to inform the design of optimized exercise interventions for obesity management. Our primary hypotheses are: (1) backward locomotion will elicit significantly higher cardiorespiratory (e.g., oxygen consumption [VO₂] or heart rate) and metabolic (e.g., energy expenditure or lactate levels) responses compared to forward locomotion; (2) backward locomotion will result in greater EMG activity in key lower-limb muscles (quadriceps, hamstrings, soleus, and tibialis anterior) due to increased coordination demands; and (3) backward locomotion will be perceived as more demanding, as measured by the Borg Rating of Perceived Exertion (RPE). Studies on backward locomotion in obese individuals are limited in the literature and have generally focused on long-term effects (e.g., body composition and cardiorespiratory fitness) or biomechanical parameters (e.g., joint loading) [6,8,9]. These studies have not comprehensively examined the acute cardiorespiratory, metabolic, and muscle activation responses of backward locomotion in the obese population. Our study provides a unique contribution by comparing the acute effects of forward and backward locomotion in sedentary male obese individuals with a BMI ≥ 30, measuring VO₂, VCO₂, heart rate, energy expenditure, lactate, Borg RPE, and quadriceps, hamstring, soleus, and tibialis anterior EMG activities. Separating the effects of speed and direction with a Latin Square design and two-way ANOVA fills this gap in the literature. It provides practical recommendations for optimizing exercise programs for obese individuals.

2. Materials and Methods

2.1. Study Design

This study employed a single-group crossover design with 28 participants. Each group was randomly assigned 7 participants. Four conditions were established: 3 km/h forward locomotion, 3 km/h backward locomotion, 4 km/h forward locomotion, and 4 km/h backward locomotion. The sequence of conditions was determined using a Latin Square design to minimize order effects. Each participant experienced all conditions in a different order. The four conditions were applied on four separate days, each lasting 10 min and providing sufficient rest periods between conditions. The procedure of the study is presented in Figure 1.

2.2. Determination of Sample Size

The a priori power analysis for a repeated measures ANOVA, conducted with an assumed medium effect size (f = 0.30), an alpha level of 0.05, and a desired statistical power of 0.95, indicates that a minimum total sample size of 26 participants is required. The analysis was based on a single group measured across four time points, with an assumed correlation among repeated measures of 0.5 and a nonsphericity correction (ϵ) of 1, indicating sphericity was assumed.

The computed noncentrality parameter (λ = 18.72) and the critical F value (Fcritical = 2.73) with numerator and denominator degrees of freedom of 3 and 75, respectively, confirm the adequacy of the sample size for detecting the specified effect. The actual power achieved (1 − β = 0.96) slightly exceeds the target, further supporting the robustness of the design. These results suggest that the study is well-powered to detect within-subject effects across the four measurements, minimizing the risk of Type II error.

Such power analyses are essential in repeated measures designs to ensure sufficient sensitivity, particularly when multiple measurements and potential correlations among repeated observations are involved [10].

The study included 28 sedentary male individuals studying at Erciyes University Faculty of Sports Sciences, aged between 18 and 34 (mean age: 21.93 ± 2.31 years), with a high body mass index (mean BMI: 31.88 ± 1.16 kg/m²), a large waist circumference (mean: 105.00 ± 6.13 cm), an average height of 176.54 ± 4.87 cm, and an average body weight of 100.35 ± 8.64 kg. Results are shown in

Table 1. Basic information on experimental subjects (n = 28).

Age (y)	Height (kg)	Weight (cm)	BMI	Waist (cm)
21.93 ± 2.31	100.35 ± 8.64	176.54 ± 4.87	31.88 ± 1.16	105.00 ± 6.13

.

Inclusion Criteria for the Study

−

Participants should not have any health problems that prevent them from exercising, and should not have an addiction to substances such as cigarettes, alcohol, or drugs;

−

A body mass index of at least 30;

−

A waist circumference of at least 94 cm.

The study and intervention protocol were prepared by the principles of the Declaration of Helsinki [11] and were approved by the Erciyes University Health Sciences Research Ethics Committee (IRB approval: 3074/CEIH/2022).

2.3. Measurements

2.3.1. Height Measurement

The participants’ heights were measured using a stadiometer scaled to 0.1 cm. The measurements were conducted without shoes, and the participants were instructed to stand with their backs against the measurement wall with their feet together. The stadiometer was adjusted to touch the participant’s head, and the resulting value was recorded as the height.

2.3.2. Body Composition Measurement

Bioelectrical impedance measurements were performed using the BC-418 (Tanita Corp., Tokyo, Japan). The system consists of a metal platform with four stainless steel rectangular footpad electrodes mounted on force transducers for weight measurement, and two handgrips with front and rear electrodes, providing eight electrodes. Measurements were conducted using a 0.8 mA sinusoidal constant current at 50 kHz. The participants’ age, gender, and height were entered, and the body type was set to ‘standard’ for all participants. Before testing, the participants were instructed to remove any metal accessories or similar items. A weight tare of 0.5 kg was applied for all participants. Subsequently, the participants were asked to stand barefoot on the machine’s metal footplates while holding the handgrips. The test surface was thoroughly cleaned and dried before each measurement.

2.3.3. Cardiopulmonary Exercise Test

The cardiopulmonary exercise test was conducted using a measurement system (Hp Cosmos, Bolzano, Italy). A suitable mask was fitted to the participant’s face, ensuring no external air leakage. A wireless receiver (Polar, Kempele, Finland) was attached to the chest to monitor heart rate. A safety belt was secured around the participant’s waist to prevent falling behind on the treadmill and ensure test safety. The participants warmed up on the treadmill at a 2.0 km/h speed with a 0% incline for 5 min. The test was performed at a 0% incline and predetermined speeds for 10 min. The system automatically analyzed the inhaled and exhaled air during the test, displaying the results on the screen. The average VO₂ (L/min), VCO₂ (L/min), VO₂ (mL/kg/min), heart rate, and energy expenditure (kcal/min) were reported. The participants were informed before the test that it should be terminated if they could not continue walking or experienced chest pain, dizziness, or nausea. Figure 2 illustrates the method used to measure VO₂ (L/min), VCO₂ (L/min), and VO₂ (mL/kg/min).

2.3.4. Blood Lactate Measurement

The participants’ lactate levels were measured using the handheld Lactate SCOUT device. The Lactate SCOUT is a portable analyzer that performs lactic acid analysis in 10 s using an enzymatic/amperometric method from 0.2 µL of capillary blood. Immediately before starting the exercise, a 0.5 µL blood sample (significantly less than one drop) was taken from the participant’s index finger, placed in the analyzer, and the result was recorded as resting blood lactate. Subsequently, the designated walking protocol was applied, and final measurements were taken 1 min after completing the exercise to determine lactate concentrations. The obtained results were reported as post-exercise blood lactate values. Figure 3 illustrates the method of blood lactate measurement

2.3.5. Perceived Exertion Rating

The Rating of Perceived Exertion (RPE) is a valid, simple, cost-effective tool for quantifying and monitoring training loads. The scale ranges from 1 to 10, with 1 representing the lowest level of exertion and 10 indicating the highest. After the test, the participants were asked, “How was the exercise?” and rated the level of exertion they felt during the walk based on the scale. The obtained data were recorded.

2.3.6. Anthropometric Measurement

Waist circumference was measured by identifying the midpoint between the participant’s lowest rib and the uppermost point of the iliac crest. The measurements were taken at this midpoint with the participants standing upright and without clothing around the waist. To avoid excessive pressure while maintaining consistent skin contact, measurements were obtained using a tape measure and recorded in centimeters.

2.3.7. Surface Electromyography Measurement

The tibialis anterior, soleus, quadriceps, and hamstring muscles were selected to analyze the study’s lower-extremity muscle groups. Surface Ag/AgCl electrodes were used for the measurements. Before electrode placement, the skin was shaved and cleaned with isopropyl alcohol wipes to reduce impedance values (<10 kΩ). Electrode placements were applied by the SENIAM (Surface Electromyography for the Non-Invasive Assessment of Muscle) standards and recommendations.

The surface electromyography (EMG) measurements encompassed the soleus, tibialis anterior, quadriceps femoris, and biceps femoris muscles. EMG signals from these four lower-extremity muscles were collected using surface electrodes during forward and backward locomotion on a treadmill. The EMG signal frequency range was filtered using the Bandpass MATLAB (version R2018b) function. Following filtration, the maximum voluntary contraction (MVC) value was calculated. MVC measurements for each muscle were conducted before the forward and backward locomotion trials, following the muscle testing method outlined by Hislop and Montgomery [12]. Maximum voluntary isometric contraction (MVIC) data were used to normalize EMG amplitude. The participants performed a maximum voluntary isometric contraction test for each muscle group tested.

The EMG data for each muscle group were normalized to the highest 1 s average EMG recorded during the maximum voluntary isometric contraction. Average EMG (AVG) and root mean square (RMS) values were calculated for each targeted muscle group. The EMG data were smoothed using a fourth-order Butterworth zero-phase low-pass filter to analyze muscle activation patterns. Figure 4 illustrates the method for surface electromyography measurement.

2.3.8. Par-Q Questionnaire

Before initiating the measurement protocol, a preliminary health screening was conducted using the Par-Q questionnaire to identify potential risks. The seven questions outlined in the Par-Q were carefully read to and by the participants, ensuring accurate responses. If a participant answered “yes” to one or more questions, they were referred to a doctor along with details of the specific question(s) to which they had responded affirmatively. These individuals were excluded from the study group.

2.3.9. Statistical Analysis

The statistical analysis of the data obtained from the study was performed using the SPSS v. 29 software package. The Shapiro–Wilk test was applied to assess the normality of the data distribution, and the results were evaluated. Differences between groups at baseline (pre-test) were analyzed using the independent samples t-test. A two-way repeated measures ANOVA was employed to examine the effects of speed (3 km/h and 4 km/h) and direction (forward and backward) factors on cardiorespiratory response, metabolic response, and muscle EMG activities.

Effect sizes for the ANOVA results were expressed as partial eta-squared (partial η²), while effect sizes for the dependent samples t-test results were reported as Cohen’s d (effect size, ES). The effect sizes were evaluated based on Cohen’s (1988) classification, with values < 0.4 considered small, 0.41–0.70 moderate, and >0.70 large, as outlined by Ulupınar and İnce [13]. The statistical significance level was set at p < 0.05. These analyses provided a reliable and systematic foundation for interpreting the study’s findings.

3. Results

Table 2. Acute Effects of Cardiorespiratory, Metabolic, Electromyographic, and Perceived Exertion Parameters During Forward and Backward Locomotion (n = 28).

					Direction			Speed			Direction * Speed
Parameters	3 km/h Forward	3 km/h Backward	4 km/h Forward	4 km/h Backward	F	p	ηp2	F	p	ηp2	F	p	ηp2
VO2 (mL/min)	1004.56 ± 150.85	1226.53 ± 190.76	1150.04 ± 175.52	1504.74 ± 210.48	398.4	0.001 **	0.94	120.74	0.001 **	0.82	49.62	0.001 **	0.65
VCO2 (mL/min)	793.09 ± 127.7	989.57 ± 198.48	893.02 ± 138.71	1208.27 ± 194.2	243.02	0.001 **	0.90	82.48	0.001 **	0.75	22.56	0.001 **	0.45
VO2 (mL/min/kg)	10.53 ± 2.09	13.25 ± 4.59	11.57 ± 1.38	15.19 ± 1.87	50.99	0.001 **	0.65	8.74	0.006 **	0.24	1.00	0.32	0.03
Heart Rate (beats/min)	90.86 ± 10.59	97.33 ± 10.71	96.74 ± 11.65	108.91 ± 12.63	181.79	0.001 **	0.87	50.41	0.001 **	0.65	16.22	0.001 **	0.37
Energy Expenditure (kcal/min)	4.71 ± 0.66	5.86 ± 0.91	5.51 ± 0.9	7.17 ± 1.03	331.07	0.001 **	0.92	141.74	0.001 **	0.84	23.05	0.001 **	0.46
Borg RPE (0–10)	1.21 ± 0.42	2.61 ± 0.74	2.04 ± 0.74	4.46 ± 1.04	289.71	0.001 **	0.91	93.35	0.001 **	0.77	32.48	0.001 **	0.54
Pre-Lactate (mmol/L)	1.67 ± 0.42	1.69 ± 0.37	1.89 ± 0.53	1.68 ± 0.45	2.38	0.13	0.08	1.59	0.22	0.05	3.06	0.09	0.10
Post-Lactate (mmol/L)	1.77 ± 0.56	2.08 ± 0.54	1.95 ± 0.77	2.51 ± 0.87	20.41	0.001 **	0.43	4.77	0.04 *	0.15	2.04	0.16	0.07
Quadriceps %MVC	22.02 ± 14.50	18.94 ±9.17	30.47 ± 18.55	25.03 ± 10.64	6.78	0.015 *	0.20	81.85	0.001 **	0.75	1.81	0.189	0.06
Hamstring %MVC	12.77 ± 5.60	13.18 ± 3.18	15.93 ± 6.19	18.02 ± 4.77	5.23	0.03 *	0.16	94.67	0.001 **	0.77	2.72	0.11	0.09
Soleus %MVC	23.08 ± 7.45	24.13 ± 6.64	28.18 ± 8.68	29.43 ± 6.81	1.88	0.181	0.06	67.29	0.001 **	0.71	0.26	0.87	0.01
Tibialis Anterior %MVC	20.57 ± 10.32	25.56 ± 9.75	26.74 ± 14.27	30.81 ± 11.60	19.87	0.001 **	0.42	73.64	0.001 **	0.73	0.93	0.34	0.03

Note: VO₂ = oxygen uptake per minute (mL/min); VCO₂ = carbon dioxide production per minute (mL/min); VO₂ (mL/kg/min) = oxygen uptake per minute normalized to body weight; kcal/min = energy expenditure in kilocalories per minute; Borg RPE = Rating of Perceived Exertion, 0–10 scale; (mmol/L) = millimoles per liter; %MVC = percentage of maximum voluntary contraction; $\overline{\mathrm{x}}$ ± S.D. = mean ± standard deviation; ηp² = partial eta-squared (effect size); * p < 0.05, ** p < 0.01.

presents the changes in cardiovascular, metabolic, electromyographic, and perceived exertion parameters of the participants during forward and backward movement at speeds of 3 km/h and 4 km/h.

At 3 km/h, backward locomotion significantly increased VO₂ (1226.53 ± 190.76 mL/min), VCO₂ (989.57 ± 198.48 mL/min), VO₂ (13.25 ± 4.59 mL/min/kg), and energy expenditure (5.86 ± 0.91 kcal/min) compared to forward locomotion (p < 0.001). Similarly, at 4 km/h, backward locomotion led to higher VO₂ (150.74 ± 210.48 mL/min), VCO₂ (1208.27 ± 194.2 mL/min), VO₂ (15.19 ± 1.87 mL/min/kg), and energy expenditure (7.17 ± 1.03 kcal/min) than forward locomotion (p < 0.001). Heart rate was also elevated during backward locomotion at both 3 km/h (97.33 ± 12.47 beats/min) and 4 km/h (108.91 ± 12.63 beats/min) compared to forward locomotion (p < 0.001). Post-lactate levels increased after backward locomotion at 4 km/h (2.51 ± 0.87 mmol/L, p = 0.04). Perceived exertion (Borg RPE) was higher during backward locomotion than forward locomotion at both speeds. At 3 km/h, Borg RPE was 2.61 ± 0.74 for backward locomotion compared to 1.21 ± 0.42 for forward locomotion (p < 0.001). At 4 km/h, it reached 4.46 ± 1.04 for backward locomotion compared to 2.04 ± 0.74 for forward locomotion (p < 0.001). Figure 5 shows the graph of the participants’ VO₂ (mL/min) mean values at different speeds and directions. Figure 6 shows the graph of the participants’ VCO₂ (mL/min) mean values at different speeds and directions. Figure 7 shows the graph of the participants’ VO₂ (mL/min/kg) mean values at different speeds and directions. Figure 8 shows the graph of the participants’ heart rate (beats/min) mean values at different speeds and directions. Figure 9 shows the graph of the participants’ energy expenditure (kcal/min) mean values at different speeds and directions. Figure 10 shows the graph of the participants’ Borg RPE (0–10) mean values at different speeds and directions. Figure 11 shows a graph of the participants’ mean pre-lactate and post-lactate (mmol/L) during forward motion at 3 km/h, and shows a graph of the participants’ mean pre-lactate and post-lactate (mmol/L) during forward motion at 4 km/h.

Forward locomotion significantly increased quadriceps activation compared to backward locomotion. At 3 km/h, the quadriceps %MVC was higher during forward locomotion (22.02 ± 14.50) than backward locomotion (18.94 ± 9.17, p = 0.015). Similarly, at 4 km/h, the quadriceps %MVC was greater during forward locomotion (30.47 ± 18.55) than backward locomotion (25.03 ± 10.64, p < 0.001). In contrast, tibialis anterior %MVC and hamstring %MVC were higher during backward locomotion. At 3 km/h, the tibialis anterior %MVC (25.56 ± 9.75) and hamstring %MVC (13.18 ± 3.18) were elevated in backward locomotion compared to forward locomotion (20.57 ± 10.32 and 12.77 ± 5.60, respectively, p < 0.05). At 4 km/h, this trend persisted with the tibialis anterior %MVC (30.81 ± 11.60) and hamstring %MVC (18.02 ± 4.77) being higher in backward locomotion compared to forward locomotion (26.74 ± 14.27 and 15.93 ± 6.19, respectively, p < 0.001). Figure 12 shows the graph of the participants’ quadriceps (MVC%) mean values at different speeds and directions. Figure 13 shows the graph of the participants’ hamstring (MVC%) mean values at different speeds and directions. Figure 14 shows the graph of the participants’ soleus (MVC%) mean values at different speeds and directions. Figure 15 shows the graph of the participants’ tibialis anterior (MVC%) mean values at different speeds and directions.

These findings indicate that forward locomotion leads to greater quadriceps activation. In contrast, backward locomotion imposes higher cardiorespiratory and metabolic demands, increases tibialis anterior and hamstring activation, and elevates perceived exertion at speeds of 3 km/h and 4 km/h.

4. Discussion

This study examined the acute effects of forward and backward locomotion training on cardiorespiratory, metabolic, and muscle electromyography (EMG) responses in sedentary male individuals with a body mass index (BMI) ≥ 30, providing novel insights into the physiological and biomechanical distinctions between these movement modalities.

The strengths of this study lie in its robust methodological design and comprehensive physiological assessments. The single-group crossover design with Latin Square counterbalancing effectively minimized order effects, ensuring reliable comparisons between forward and backward locomotion conditions. The inclusion of multiple outcome measures, including cardiorespiratory (VO₂, VCO₂, heart rate), metabolic (energy expenditure, lactate), electromyographic (EMG), and perceived exertion (RPE) data, provided a holistic understanding of acute responses in obese individuals. However, the study has limitations that warrant consideration. The sample was restricted to sedentary male participants aged 18–34 with a BMI ≥ 30, limiting generalizability to females, older adults, or nonsedentary individuals. The focus on acute responses precludes conclusions about long-term effects on fitness or weight management.

The findings confirm that backward locomotion imposes significantly higher cardiorespiratory and metabolic demands than forward locomotion, consistent with prior research, while offering new evidence specific to obese populations. Significant direction effects were observed for oxygen consumption (VO₂: F (1,26) = 398.40, p < 0.001, η² = 0.94), carbon dioxide production (VCO₂: F (1,26) = 243.02, p < 0.001, η² = 0.90), heart rate (F (1,26) = 181.79, p < 0.001, η² = 0.87), and energy expenditure (F (1,26) = 331.07, p < 0.001, η² = 0.92), indicating a greater physiological cost during backward locomotion. These results align with studies reporting a 30–40% higher metabolic cost in backward running due to increased step frequency and reduced stride length [14,15,16]. The elevated cardiorespiratory demand can be theoretically framed within the energy cost of the locomotion model, which posits that nonhabitual movement patterns, such as backward walking, require greater neural and muscular coordination, thus increasing metabolic expenditure [16]. In obese individuals, this effect is likely amplified due to the higher baseline energy costs associated with excess body mass, as noted in studies on forward locomotion [17,18].

The increased muscle activation during backward locomotion, particularly in the tibialis anterior (percentageMVC: F (1,26) = 19.87, p < 0.001, η² = 0.42), corroborates findings that backward running enhances the activation of anterior lower-limb muscles compared to posterior muscles like the soleus (percentageMVC: F (1,26) = 1.88, p = 0.181, η² = 0.06) [6,19]. The motor control theory may explain this selective activation, which suggests that backward locomotion alters muscle recruitment patterns due to reversed kinematics and increased demand for ankle dorsiflexion [20]. The lack of significant soleus activation aligns with reduced heel loading during backward locomotion, as heel-strike dominance diminishes [21]. For obese individuals, this shift in muscle recruitment could reduce stress on weight-bearing joints, supporting the hypothesis that backward locomotion is a joint-friendly exercise modality [22]. Although joint forces were not directly measured, the literature indicates lower patellofemoral compressive forces in backward running (3.0 ± 0.6 vs. 5.6 ± 1.3 body weight in forward running), suggesting potential benefits for obese individuals prone to joint overuse injuries [22,23].

The metabolic responses further highlight the distinct demands of backward locomotion. The significant direction effect on post-exercise lactate levels (F (1,26) = 20.41, p < 0.001, η² = 0.43) reflects greater anaerobic energy contribution, which is consistent with studies reporting elevated lactate during backward locomotion in varied contexts, such as aquatic environments [24]. This can be linked to the lactate threshold model, where higher exercise intensity in backward locomotion pushes metabolic demand beyond aerobic capacity, particularly in obese individuals with lower cardiorespiratory fitness [25]. The absence of a significant direction effect on pre-exercise lactate (F (1, 26) = 2.38, p = 0.13, η² = 0.08) confirms that the observed differences were exercise-induced, reinforcing the specificity of backward locomotion’s metabolic impact.

The perceptual responses, measured via Borg RPE (F (1,26) = 289.71, p < 0.001, η² = 0.91), indicate that backward locomotion was perceived as more challenging, aligning with reports of higher perceived exertion due to unfamiliar movement patterns and increased neuromuscular effort [16,24]. The significant speed–direction interaction for VO₂ (F (1,26) = 49.62, p < 0.001, η² = 0.65) and heart rate (F (1,26) = 50.41, p < 0.001, η² = 0.65) suggests that increasing speed amplifies the physiological cost of backward locomotion, a finding that extends prior work on graded backward walking at higher intensities [26].

While this study focused on acute effects, the observed increases in energy expenditure and muscle activation suggest that backward locomotion could enhance weight management and fitness in obese individuals, as supported by studies reporting reductions in waist-to-hip ratios and body fat percentage with backward training [8,9]. However, the paucity of experimental studies on obese populations limits direct comparisons. Conceptually, backward locomotion may serve as a novel stimulus within the exercise variety model, promoting engagement by introducing movement diversity, which is critical for sustaining exercise adherence in obese individuals [27].

This study’s novelty lies in its comprehensive analysis of acute responses in obese males. It addresses a gap in the literature where backward locomotion studies predominantly focus on nonobese populations or long-term outcomes [8,28]. Integrating cardiorespiratory, metabolic, and EMG data within a robust Latin Square design provides a foundation for optimizing exercise prescriptions for obese individuals.

5. Conclusions

This study provides robust evidence that backward locomotion elicits significantly higher cardiorespiratory, metabolic, and perceptual demands than forward locomotion in individuals with a body mass index (BMI) ≥ 30. The increased oxygen consumption (VO₂), heart rate, energy expenditure, and lactate production during backward locomotion, coupled with heightened activation of specific muscles such as the tibialis anterior, indicate that this movement modality is more metabolically demanding. Furthermore, the significant effects of speed and the direction–speed interaction underscore the importance of considering both factors when designing exercise protocols. While backward locomotion offers benefits such as reduced joint loading and increased energy expenditure, its higher perceptual demand (Borg RPE) should be considered when recommending it to overweight individuals.

Practical Recommendations: The findings suggest that backward walking can be an effective exercise modality for obese individuals to support weight management and enhance cardiorespiratory fitness. Exercise professionals may implement 10 min backward walking sessions at 3–4 km/h, 2 times per week, to increase energy expenditure while minimizing joint stress. However, due to the elevated perceptual effort, sessions should start at a low intensity and progressively increase based on individual tolerance. This approach can promote exercise adherence and serve as a joint-friendly alternative in rehabilitation programs for obese individuals.

Study Limitations: The study was limited to sedentary male participants, restricting the generalizability of findings to females or other age groups. Additionally, the lack of the direct measurement of joint forces necessitated reliance on indirect evidence regarding reduced joint loading in backward locomotion. As the study focused on acute effects, the long-term impacts of backward locomotion (e.g., on body composition or joint health) remain unexplored. Finally, although the Latin Square design minimized order effects, individual physiological variations may have partially influenced the results.

Recommendations for Future Research: This study provides valuable insights into the acute effects of forward and backward locomotion in obese individuals but identifies knowledge gaps that offer opportunities for future research. Below are structured recommendations addressing these gaps, outlining clinical and methodological directions.

Investigation of Long-Term Effects: The study’s focus on acute responses limits conclusions about long-term effects on fitness and weight management. Future research should employ longitudinal designs to examine the impact of backward locomotion training on body composition (e.g., fat mass or muscle mass), cardiorespiratory fitness (VO₂max), and metabolic health (e.g., insulin sensitivity). Clinically, such studies could optimize the integration of backward locomotion into long-term exercise programs for obesity management.

Broader and More Diverse Samples: The sample was restricted to sedentary males aged 18–34 with a BMI ≥ 30, limiting generalizability to females, older adults, or nonsedentary individuals. Future studies should include more diverse populations regarding gender, age, and activity level. This would enhance understanding of backward locomotion’s physiological and perceptual effects across demographics, supporting the development of inclusive exercise prescriptions. Clinically, this diversity could inform personalized exercise programs for obese individuals.

Incorporation of Biomechanical Analyses: Joint forces were not directly measured, hindering the full validation of backward locomotion’s joint-protective potential. Future research should incorporate biomechanical analyses (e.g., motion capture systems or wearable sensors for joint load measurements) to assess the impact of backward locomotion on joint stress. Methodologically, integrating these analyses with muscle activation (EMG) and cardiorespiratory data could provide a more comprehensive evaluation. This could support backward locomotion as a joint-friendly exercise in rehabilitation programs.

Clinical Application and Feasibility Studies: The feasibility of backward locomotion in clinical settings, particularly for rehabilitation or weight management programs in obese individuals, remains unexplored. Future studies should evaluate the integration of backward locomotion into structured exercise programs, assessing patient adherence and feasibility. For instance, the impact of backward locomotion’s higher perceived exertion (RPE) on exercise adherence in obese populations could be investigated. Methodologically, robust experimental designs like the Latin Square approach could enhance the reliability of such studies.

These recommendations aim to contribute to developing comprehensive strategies for optimizing exercise programs in obese populations and better understanding the clinical and practical potential of backward locomotion.