An older adult walk side-by-side, without contact, with a younger adult without step synchronization.

Habitual (preferred) walking speed to cover 200m. Normal values for older men: 1.21m/s -1.32m/s), older women 1.07m/s -1.19m/s).

Habitual (preferred) walking speed to cover 200m. Normal values for older men: 1.21m/s -1.32m/s), older women 1.07m/s -1.19m/s).

Habitual (preferred) walking speed to cover 200m. Normal values for older men: 1.21m/s -1.32m/s), older women 1.07m/s -1.19m/s).

Average number of steps per second while walking at habitual (preferred) speed over 200m.

Average number of steps per second while walking at habitual (preferred) speed over 200m.

Average number of steps per second while walking at habitual (preferred) speed over 200m.

Average number of steps per second while walking at habitual (preferred) speed over 200m.

Average number of steps per second while walking at habitual (preferred) speed over 200m.

Average number of steps per second while walking at habitual (preferred) speed over 200m.

Average step length while walking at habitual (preferred) speed over 200m.

Average step length while walking at habitual (preferred) speed over 200m.

Average step length while walking at habitual (preferred) speed over 200m.

Average step length while walking at habitual (preferred) speed over 200m.

Average step length while walking at habitual (preferred) speed over 200m.

Average step length while walking at habitual (preferred) speed over 200m.

Left / right symmetry of the gait assessed via the autocorrelation method. A high value shows a symmetric gait.

Left / right symmetry of the gait assessed via the autocorrelation method. A high value shows a symmetric gait.

Left / right symmetry of the gait assessed via the autocorrelation method. A high value shows a symmetric gait.

Left / right symmetry of the gait assessed via the autocorrelation method. A high value shows a symmetric gait.

Left / right symmetry of the gait assessed via the autocorrelation method. A high value shows a symmetric gait.

Left / right symmetry of the gait assessed via the autocorrelation method. A high value shows a symmetric gait.

Left / right symmetry of the gait assessed via the autocorrelation method. A high value shows a symmetric gait.

Left / right symmetry of the gait assessed via the autocorrelation method. A high value shows a symmetric gait.

Regularity of the gait assessed via the autocorrelation method. A high value shows a less variable gait.

Regularity of the gait assessed via the autocorrelation method. A high value shows a less variable gait.

Regularity of the gait assessed via the autocorrelation method. A high value shows a less variable gait.

Regularity of the gait assessed via the autocorrelation method. A high value shows a less variable gait.

short-term Local dynamic stability (maximal Lyapunov exponent). A low value shows a more stable gait and a lower risk of fall.

short-term Local dynamic stability (maximal Lyapunov exponent). A low value shows a more stable gait and a lower risk of fall.

short-term Local dynamic stability (maximal Lyapunov exponent). A low value shows a more stable gait and a lower risk of fall.

short-term Local dynamic stability (maximal Lyapunov exponent). A low value shows a more stable gait and a lower risk of fall.

short-term Local dynamic stability (maximal Lyapunov exponent). A low value shows a more stable gait and a lower risk of fall.

short-term Local dynamic stability (maximal Lyapunov exponent). A low value shows a more stable gait and a lower risk of fall.

Attractor complexity index. A high value indicates that less attention is dedicated to control gait (automated gait).

Attractor complexity index. A high value indicates that less attention is dedicated to control gait (automated gait).

Attractor complexity index. A high value indicates that less attention is dedicated to control gait (automated gait).

Attractor complexity index. A high value indicates that less attention is dedicated to control gait (automated gait).

Attractor complexity index. A high value indicates that less attention is dedicated to control gait (automated gait).

Attractor complexity index. A high value indicates that less attention is dedicated to control gait (automated gait).

Attentional demand measured via detrended fluctuation analysis (DFA). A high value indicates that less attention is dedicated to control gait (automated gait).

Attentional demand measured via detrended fluctuation analysis (DFA). A high value indicates that less attention is dedicated to control gait (automated gait).

Attentional demand measured via detrended fluctuation analysis (DFA).A high value indicates that less attention is dedicated to control gait (automated gait).

Attentional demand measured via detrended fluctuation analysis (DFA).A high value indicates that less attention is dedicated to control gait (automated gait).

Attentional demand measured via detrended fluctuation analysis (DFA).A high value indicates that less attention is dedicated to control gait (automated gait).

Attentional demand measured via detrended fluctuation analysis (DFA).A high value indicates that less attention is dedicated to control gait (automated gait).

Variability of stride (gait cycle) duration over 200m walking. A high value shows a high stride-to-stride variability.

Variability of stride (gait cycle) duration over 200m walking. A high value shows a high stride-to-stride variability.

Variability of stride (gait cycle) duration over 200m walking. A high value shows a high stride-to-stride variability.

Variability of stride (gait cycle) duration over 200m walking. A high value shows a high stride-to-stride variability.

Variability of stride (gait cycle) duration over 200m walking. A high value shows a high stride-to-stride variability.

Variability of stride (gait cycle) duration over 200m walking. A high value shows a high stride-to-stride variability.

The Timed Up and Go (TUG) test is a clinical test used to assess balance and walking ability in older populations. The test measures the time a participant take to get up from an armed chair, walk three meters to a line drawn on the floor, turn around and sit back down in the same chair. The time is measured in seconds.

The results can be interpreted the following way:

• ≤ 10 seconds = normal, good general health
• ≤ 20 seconds = good mobility, can go out alone, mobile without a gait aid
• > 30 seconds = difficulties, cannot go outside alone, requires a gait aid

The Timed Up and Go (TUG) test is a clinical test used to assess balance and walking ability in older populations. The test measures the time a participant take to get up from an armed chair, walk three meters to a line drawn on the floor, turn around and sit back down in the same chair. The time is measured in seconds.

The results can be interpreted the following way:

• ≤ 10 seconds = normal, good general health
• ≤ 20 seconds = good mobility, can go out alone, mobile without a gait aid
• > 30 seconds = difficulties, cannot go outside alone, requires a gait aid

The Timed Up and Go (TUG) test is a clinical test used to assess balance and walking ability in older populations. The test measures the time a participant take to get up from an armed chair, walk three meters to a line drawn on the floor, turn around and sit back down in the same chair. The time is measured in seconds.

The results can be interpreted the following way:

• ≤ 10 seconds = normal, good general health
• ≤ 20 seconds = good mobility, can go out alone, mobile without a gait aid
• > 30 seconds = difficulties, cannot go outside alone, requires a gait aid

The Timed Up and Go (TUG) test is a clinical test used to assess balance and walking ability in older populations. The test measures the time a participant take to get up from an armed chair, walk three meters to a line drawn on the floor, turn around and sit back down in the same chair. The time is measured in seconds.

The results can be interpreted the following way:

• ≤ 10 seconds = normal, good general health
• ≤ 20 seconds = good mobility, can go out alone, mobile without a gait aid
• > 30 seconds = difficulties, cannot go outside alone, requires a gait aid

The Timed Up and Go (TUG) test is a clinical test used to assess balance and walking ability in older populations. The test measures the time a participant take to get up from an armed chair, walk three meters to a line drawn on the floor, turn around and sit back down in the same chair. The time is measured in seconds.

The results can be interpreted the following way:

• ≤ 10 seconds = normal, good general health
• ≤ 20 seconds = good mobility, can go out alone, mobile without a gait aid
• > 30 seconds = difficulties, cannot go outside alone, requires a gait aid

The Timed Up and Go (TUG) test is a clinical test used to assess balance and walking ability in older populations. The test measures the time a participant take to get up from an armed chair, walk three meters to a line drawn on the floor, turn around and sit back down in the same chair. The time is measured in seconds.

The results can be interpreted the following way:

• ≤ 10 seconds = normal, good general health
• ≤ 20 seconds = good mobility, can go out alone, mobile without a gait aid
• > 30 seconds = difficulties, cannot go outside alone, requires a gait aid

The Unipedal stance test measure the ability to stand on one leg and maintain balance. The participants are asked to stand as long as possible on one leg. The exercise is repeated three times for each leg. The time is measure in seconds. The average of the scores is calculated.

The mean time held for people aged 70-79 is 14.2 +/- 9.3 seconds.

The Unipedal stance test measure the ability to stand on one leg and maintain balance. The participants are asked to stand as long as possible on one leg. The exercise is repeated three times for each leg. The time is measure in seconds. The average of the scores is calculated.

The mean time held for people aged 70-79 is 14.2 +/- 9.3 seconds.

The Unipedal stance test measure the ability to stand on one leg and maintain balance. The participants are asked to stand as long as possible on one leg. The exercise is repeated three times for each leg. The time is measure in seconds. The average of the scores is calculated.

The mean time held for people aged 70-79 is 14.2 +/- 9.3 seconds.

The Unipedal stance test measure the ability to stand on one leg and maintain balance. The participants are asked to stand as long as possible on one leg. The exercise is repeated three times for each leg. The time is measure in seconds. The average of the scores is calculated.

The mean time held for people aged 70-79 is 14.2 +/- 9.3 seconds.

The Unipedal stance test measure the ability to stand on one leg and maintain balance. The participants are asked to stand as long as possible on one leg. The exercise is repeated three times for each leg. The time is measure in seconds. The average of the scores is calculated.

The mean time held for people aged 70-79 is 14.2 +/- 9.3 seconds.

The Unipedal stance test measure the ability to stand on one leg and maintain balance. The participants are asked to stand as long as possible on one leg. The exercise is repeated three times for each leg. The time is measure in seconds. The average of the scores is calculated.

The mean time held for people aged 70-79 is 14.2 +/- 9.3 seconds.

The Geriatric Depression Scale (GDS) is a self-rating scale designed for rating depression in older adults. The GDS questionnaire is composed of 15 questions. 10 of the questions indicate the presence of depression when answered positively, the 5 others indicate the presence of depression if answered negatively.

• A score between 0 and 5 indicate a normal state.
• A score between 5 and 11 indicate a mild to moderate depression.
• A score between 11 and 15 indicate a severe depression.

The Geriatric Depression Scale (GDS) is a self-rating scale designed for rating depression in older adults. The GDS questionnaire is composed of 15 questions. 10 of the questions indicate the presence of depression when answered positively, the 5 others indicate the presence of depression if answered negatively.

• A score between 0 and 5 indicate a normal state.
• A score between 5 and 11 indicate a mild to moderate depression.
• A score between 11 and 15 indicate a severe depression.

The Geriatric Depression Scale (GDS) is a self-rating scale designed for rating depression in older adults. The GDS questionnaire is composed of 15 questions. 10 of the questions indicate the presence of depression when answered positively, the 5 others indicate the presence of depression if answered negatively.

• A score between 0 and 5 indicate a normal state.
• A score between 5 and 11 indicate a mild to moderate depression.
• A score between 11 and 15 indicate a severe depression.

The FES-I questionnaire measure the "concern about falling". The questionnaire is composed of 16 questions with 4 choices valued from 1 to 4: "not at all concerned" (1), "a little concerned" (2), "somewhat concerned" (3), and "very concerned" (4). The minimum score is 16 and the highest 64.

• A score between 16 and 19 indicate a low concern about falling.
• A score between 20 and 27 indicates a moderate concern about falling.
• A score above 28 indicates a high concern about falling.

The FES-I questionnaire measure the "concern about falling". The questionnaire is composed of 16 questions with 4 choices valued from 1 to 4: "not at all concerned" (1), "a little concerned" (2), "somewhat concerned" (3), and "very concerned" (4). The minimum score is 16 and the highest 64.

• A score between 16 and 19 indicate a low concern about falling.
• A score between 20 and 27 indicates a moderate concern about falling.
• A score above 28 indicates a high concern about falling.

The FES-I questionnaire measure the "concern about falling". The questionnaire is composed of 16 questions with 4 choices valued from 1 to 4: "not at all concerned" (1), "a little concerned" (2), "somewhat concerned" (3), and "very concerned" (4). The minimum score is 16 and the highest 64.

• A score between 16 and 19 indicate a low concern about falling.
• A score between 20 and 27 indicates a moderate concern about falling.
• A score above 28 indicates a high concern about falling.

The "Questionnaire d'activité physique pour les personnes âgées (QAPPA)" evaluates the level of physical activity (LoPA) over the last 7days. Moderate (M) and Vigorous (V) physical activity (PA) are distinguished. The time spent per week (minutes) is multiplied by 8 for V activities and by 4 for the M activities and indicate the level of MET-min/week (Metabolic Equivalent of Task).

Moderate LoPA:

1. 150 minutes of PA (M, V or both) spread over at least 3 days of the last 7 days.
2. 800 MET-min/week of PA (M, V, or both) spread over 2 days of the last 7 days.
3. At least 20 minutes of V PA spread over 3 days of the last 7 days.

High LoPA:

1. At least 1500 MET-min/week of V PA spread over 3 days of the last 7 days
2. At least 3000 MET-min/week of M, V, or both, PA spread over 3 days of the last 7 days.

Low LoPA: cases that do not meet the classification for moderate or high level of activity.

The "Questionnaire d'activité physique pour les personnes âgées (QAPPA)" evaluates the level of physical activity (LoPA) over the last 7days. Moderate (M) and Vigorous (V) physical activity (PA) are distinguished. The time spent per week (minutes) is multiplied by 8 for V activities and by 4 for the M activities and indicate the level of MET-min/week (Metabolic Equivalent of Task).

Moderate LoPA:

1. 150 minutes of PA (M, V or both) spread over at least 3 days of the last 7 days.
2. 800 MET-min/week of PA (M, V, or both) spread over 2 days of the last 7 days.
3. At least 20 minutes of V PA spread over 3 days of the last 7 days.

High LoPA:

1. At least 1500 MET-min/week of V PA spread over 3 days of the last 7 days
2. At least 3000 MET-min/week of M, V, or both, PA spread over 3 days of the last 7 days.

Low LoPA: cases that do not meet the classification for moderate or high level of activity.

The "Questionnaire d'activité physique pour les personnes âgées (QAPPA)" evaluates the level of physical activity (LoPA) over the last 7days. Moderate (M) and Vigorous (V) physical activity (PA) are distinguished. The time spent per week (minutes) is multiplied by 8 for V activities and by 4 for the M activities and indicate the level of MET-min/week (Metabolic Equivalent of Task).

Moderate LoPA:

1. 150 minutes of PA (M, V or both) spread over at least 3 days of the last 7 days.
2. 800 MET-min/week of PA (M, V, or both) spread over 2 days of the last 7 days.
3. At least 20 minutes of V PA spread over 3 days of the last 7 days.

High LoPA:

1. At least 1500 MET-min/week of V PA spread over 3 days of the last 7 days
2. At least 3000 MET-min/week of M, V, or both, PA spread over 3 days of the last 7 days.

Low LoPA: cases that do not meet the classification for moderate or high level of activity.

The Investigating Choice Experiments for the Preferences of Older People (ICEPOP) CAPability (ICECAP-O) is a capability-based measure of the general quality of life of older people (≥ 65 years old). The ICECAP-O measures 5 attributes,each attribute has 4 level of answers. The 20 ICECAP-O attribute levels are estimated independently of one another such as the lowest level of capability (no capability on all attributes) was given a total value of 0 and the highest level of capability (full capability on all attributes) was given a total value of 1. The ICEPCAP-O range on a scale from 0 to 1.

The Investigating Choice Experiments for the Preferences of Older People (ICEPOP) CAPability (ICECAP-O) is a capability-based measure of the general quality of life of older people (≥ 65 years old). The ICECAP-O measures 5 attributes,each attribute has 4 level of answers. The 20 ICECAP-O attribute levels are estimated independently of one another such as the lowest level of capability (no capability on all attributes) was given a total value of 0 and the highest level of capability (full capability on all attributes) was given a total value of 1. The ICEPCAP-O range on a scale from 0 to 1.

The number of falls over the past year will be use to determine the retrospective fall rate.

The number of falls over three month will be used to determine the prospective fall rate. The measure will be taken every three month until 18 moths after the beginning of the training program or in the case the participants felt two times between the measure and the end of the training program.

The number of falls over three month will be used to determine the prospective fall rate. The measure will be taken every three month until 18 moths after the beginning of the training program or in the case the participants felt two times between the measure and the end of the training program.

The number of falls over three month will be used to determine the prospective fall rate. The measure will be taken every three month until 18 moths after the beginning of the training program or in the case the participants felt two times between the measure and the end of the training program.

The number of falls over three month will be used to determine the prospective fall rate. The measure will be taken every three month until 18 moths after the beginning of the training program or in the case the participants felt two times between the measure and the end of the training program.

The number of falls over three month will be used to determine the prospective fall rate. The measure will be taken every three month until 18 moths after the beginning of the training program or in the case the participants felt two times between the measure and the end of the training program.

The number of falls over three month will be used to determine the prospective fall rate. The measure will be taken every three month until 18 moths after the beginning of the training program or in the case the participants felt two times between the measure and the end of the training program.

Narrative feedback after transcription of semi-structured interviews. Quotes will be chosen to demonstrate themes which were common, or which represented a summary of a topic.

The number of participants withdrawing from the study will be used to measure the adherence to the training program.