Virtual reality walking test
Hard- and software. This test is performed with the GRAIL system (Gait Real-time Analysis Interactive Lab, Motekforce Link, Amsterdam, the Netherlands) that comprises a 3D instrumented split-belt treadmill (0.8 x 1.5 m) with two embedded force plates, a semi cylindrical 240° projection screen (2.4 x 5 m), and a Vicon MX optical infrared tracking system (Vicon, Oxford, United Kingdom). The participant is secured by two handrails laterally attached to the treadmill, two laser barriers at the front and back end of the treadmill, and a safety harness that is attached to the ceiling to prevent participants from falling. In addition, the experimenter can press a stop button to instantly stop the treadmill in case of emergency. Four serially-connected RGB projectors project a virtual scenario on the projection screen. A photodiode is attached to the screen to accurately measure stimulus onsets and to prevent varying projection onsets. An ergonomic handheld key switch with a left and a right button and a voice recorder are used to record participant’s responses.
The virtual scenario is designed with D-Flow (Motekforce Link, Amsterdam, the Netherlands). The scenario roughly depicts an industrial-like environment with abstract objects placed laterally to a virtual walking path (see Figure 1). Motor and cognitive tasks are customized and added to the application (cf. below). All instructions and all stimuli are presented visually at eye level in a small, rectangular area in the middle of the projection screen.
Motor and cognitive tasks. Six different tasks with five trials each are presented in a mixed order. Task presentation and order is identical for each participant and at pre- and posttest. No given task is presented more than twice in a row. Every trial lasts 30 s with inter-trial intervals of 3 s to introduce the next trial (e.g., “Standing only” or “Walking only”, in German language). The scenario lasts 16.5 min in total.
One baseline task, three STs (one motor, two cognitive), and two cognitive-motor MTs are performed: (1) Standing task (baseline): Participants stand quietly with both feet on the treadmill while maintaining a straight direction of view by looking at a fixation cross. Posture is assessed via ground reaction forces. (2) Walking task: Participants walk at 1 m/s while maintaining a straight direction of view by looking at a fixation cross. As the treadmill accelerates and decelerates at 0.2 m/s², transitions between standing and walking take 5 s. Gait performance is assessed via ground reaction forces. (3) Serial Threes task: The Serial Threes task is an established measure of updating of working memory (59, 60). Participants stand quietly on the treadmill and look at a fixation cross at the center of the projection screen. At the beginning of each trial, a three digit-number is displayed for 5 s. Participants are asked to count backwards in steps of three loudly, and as quickly and accurately as possible. They are instructed to keep their eyes open during counting, and to stop counting when the next task is displayed on the screen. Participants are asked to spell out the whole number (e.g., “177” instead of “77”), and not to correct errors (i.e., to continue counting backwards from a possibly wrong number). Verbal responses (i.e., numbers) are protocolled by the experimenter, and are additionally recorded using a voice recorder. (4) Color Word Stroop task: The Stroop task is used to assess inhibitory control (61). In each trial, color-denoting words (i.e., yellow, red, blue, green) are sequentially presented for 500 ms in a mixed order with ten words per trial. Each stimulus is followed by a central fixation cross for 1800 to 2200 ms, such that mean ISI is 2500 ms. Font and meaning of color words match on congruent trials (e.g., the word “green” in green font), and do not match on incongruent trials (e.g., the word “green” in blue font). Two response words are presented for 1500 ms on the projection screen, time locked with stimulus onset. They are displayed in two rectangular areas, one located to the left below the stimulus word and the other located to the right below the stimulus word. Both response words are presented in white font; one names the font of the stimulus word, and the other names one of the three other fonts. Participants have to indicate which of the two response words names the stimulus font, by depressing either the left or the right button of a handheld key switch. Participants are instructed to give their responses as fast and as accurately as possible. The design of the Stroop task is balanced across congruency (50% congruent, 50% incongruent), stimulus font (25% of each font), stimulus meaning (25% of each meaning), position of correct and false answer boxes (50% left, 50% right), and frequency of correct and false response words per color (50% correct, 50% false). Reaction times and correctness of responses are recorded.. (5) Multitask 1: The walking task and Serial Threes task are executed concurrently. (6) Multitask 2: The walking task and Stroop task are executed concurrently. Participants are instructed to not give preference to either one of the concurrently executed tasks, while responding to cognitive tasks as fast and as accurately as possible. Outcome measures are the same as described above.
Procedure. Participants familiarize with the treadmill by walking through a virtual forest environment for about 5 to 10 min, while walking speed increases slowly up to 1 m/s. Familiarization ends when participants are able to walk securely while focusing their attention on the center of the projection screen. Physically low demanding tests (including MMSE and DSST), of about 12 to 15 min duration in total, are scheduled after familiarization to ensure that participants return to a physical resting state. After that, participants perform a short practice run of about 2 min including a shortened trial of each task in a fixed order.
Virtual reality driving test
Hard- and software. This test follows closely the driving test of Wechsler, Drescher (62), where a schematic drawing of the setup is provided. The setup consists of a VW Golf seat and three 48” monitors that are mounted at eye level, covering a visual field of 195°. A Logitech G27 steering wheel is located slightly to the left in front of the middle monitor. Gas and brake pedals are placed on the floor in a position similar to a real car. The car seat and pedals are adjustable to provide a realistic and comfortable driving position. A conventional numeric keypad is mounted to the right of the steering wheel, within participants’ easy reach. Numbers from 1 to 6 are visible on the keypad, other characters are covered by black tape. A regular headset is used for auditory task presentation and to present characteristic driving noise.
The driving simulation uses commercially available hard- and software (Carnetsoft version 8.0, Groningen, The Netherlands). Figure 2 shows the setup of the scenario. It displays 25.7 km of a slightly winding rural road, without intersections or traffic lights. The simulated environment pictures a typical landscape with clouds in a blue sky, mountains, little animal enclosures, grasslands, trees, traffic signs, gas stations, and construction sides. Regular oncoming traffic comprises cars and busses. The scenario does not involve any pedestrians, cyclists, or other road users. Participants drive a VW Golf with automatic transmission and a simulated dashboard that is presented at the bottom of the middle screen. Velocity is displayed on a speedometer. The participant's vehicle is continually accompanied by one car in the rear and another car in front. The lead car is programmed to drive at 70 km/h, and to slow down slightly if the distance to the participant’s car exceeds 100 m. The rear car is programmed to follow at a reasonable distance. In case of an accident, the front window shatters and the driver’s car is directly thereafter repositioned between the lead car and the rear car.
Motor and cognitive tasks. Participants perform a driving task and a battery of additional tasks that are designed to mimic cognitively effortful activities typically performed during driving. On three driving courses, they perform the driving task alone, the additional tasks alone, or the driving and the additional tasks concurrently.
Driving-alone course: participants are instructed to follow the lead car at a regular distance with about 70 km/h, to pay attention to posted speed limits, and to brake when the lead car brakes. Ten braking events are presented at irregular course locations, that are identical for all participants and at pre- and posttest. As the lead car approaches a 40 km/h speed limit sign, its brake lights flash up and the car slows down to 40 km/h within about 7 s. It keeps that velocity for about 6 s, and then accelerates within about 9 s back to 70 km/h. Lateral car position, car velocity, and distance to the lead car are continuously assessed. For braking responses, reaction times for gas-off and brake-on reactions are measured.
Cognition-alone course: the participants’ car drives in autopilot mode and responds automatically to braking maneuvers of the lead car. Two different types of tasks are presented at fixed course locations in a mixed order that is identical for all participants and at pre- and posttest. Both tasks are presented either visually on the windshield or auditory through headphones. (1) In the typing task, participants are asked to type a three-digit number with their right hand into the numeric keypad, as quickly and as accurately as possible. Stimuli presentation lasts about 5 s for visual trials, and about 3 s for auditory trials. (2) In the reasoning task, participants are asked to verbally provide arguments for or against issues of general interest, e.g., to state an argument for/against the use of electric vehicles. Each request is presented for about 5 s visually or 4 s auditory, and cannot be answered adequately by a simple “yes” or “no”. Requests are limited to 80 characters, and to two lines on the windshield. The experimenter protocols the participants’ answers. Answers are marked as correct if participants give a valid argument, and are marked as incorrect if participants give an invalid argument, or if they do not answer at all. The validity of arguments is agreed-upon among experimenters before the study. Both the typing and the reasoning task comprise 30 trials, 15 presented auditory and 15 presented visually. The concrete stimuli (three-digit numbers, reasoning questions) differ between trials, and they also differ between pre- and posttest; however, they are the same for all participants. Reaction times and correctness of responses are measured for the typing task, and correctness of answered requests is assessed for the reasoning task (as reaction time was not assessed).
MT course: participants actively drive and brake for the lead car, and they also respond to typing and reasoning tasks that are analogous to those on the cognition-alone course. No instructions are given regarding the preference for driving versus for additional tasks.
Procedure. One half of the driving-alone course and the complete cognition-alone course are scheduled on one day, in balanced order. The other half of the driving-alone course and the complete MT course are scheduled on another day, again in balanced order. The order of days is also balanced. On a separate day before testing, participants practice the driving-alone course and the cognition-alone course for approximately 5 min each, but they don’t practice the MT course.
A battery of five tests is used to measure a broad range of different cognitive functions. All tests follow standardized procedures and instructions. Four tests are computerized, one is a paper and pencil test. Computerized tests are conducted on a 24” computer screen with a display resolution of 1920 x 1080 pixel and a screen distance of about 65 cm. Each computerized test takes about 10 min with up to three practice trials of about 1 to 2 min each. Response feedback in provided after practice trials, but not after registered trials.
The N-back, Simon and Task switching tests are programmed in E-Prime 2.0 (Psychology Software Tools, Pittsburgh, PA) with stimuli presented in six blocks with inter-block breaks of 5 s (20 s after block 3). The maximum response window is 2000 ms. After a response is given or after 2000 ms a central fixation cross is presented for a variable response-stimulus interval between 800 to 1200 ms. All stimuli are black and presented on a white screen background. Participants respond by depressing the “X” or “M” key on a German keyboard with their left and right index finger, and they are instructed to respond as fast and as accurately as possible. Reaction times and correctness of responses are recorded.
Updating is assessed using the 2-back condition of a visuo-spatial N-back test (63). A black 4x4 grid is presented continuously. Dots (n = 19 per block) are presented sequentially in the center of different grid cells for 500 ms. Participants have to memorize the position of the dots and to depress the right key “M” when the position of the current dot is identical to the position of the second-to-last dot (target). They have to depress the left key “X” when the current dot appears at a different position as the second-to-last dot (non-target). In total, 30 targets and 72 non-targets are presented.
Inhibitory control is assessed using the Simon test (64). A black fixation cross is presented continuously on a white screen. Left- or rightward pointing arrows (n = 32 per block) are displayed sequentially for 500 ms either on the left or right side of the fixation cross. For one half of the stimuli the direction and position of the arrow are congruent (e.g., leftward pointing arrow on the left side), while for the other half of the stimuli, direction and position are incongruent (e.g., leftward pointing arrow on the right side, n = 96). Participants are instructed to press the left key “X” for leftward pointing arrows and the right key “M” for rightward pointing arrows.
Shifting is assessed using a modified visual task switching test (65). Geometrical shapes (n = 17 per block) are presented sequentially for 1500 ms. The geometrical shapes are either quadratic or circular and either big or small. Participants are instructed to indicate either the size of the shape (subtask A) or the form of the shape (subtask B) by pressing either the left key “X” for small or circular shapes or the right key “M” for big or quadratic shapes. In each block subtasks are presented in the following order: AABBAABBAABBAABBA. No external cues about subtask order are provided.
Dual-tasking (DT) is assessed using a dual-tasking test adapted from literature (66) where a manual tracking task and an auditory discrimination task are performed concurrently. Nine trials with about 45 s each are presented in three blocks: (a) three trials ST manual tracking (b) three trials ST auditory discrimination, and (c) three trials DT manual tracking and auditory discrimination with both tasks being performed simultaneously. The three blocks are presented in a randomized order across participants. In ST manual tracking trials, a small red target square moves from one side of the screen to the other following an unpredictable wave-shaped path. Participants are instructed to track the red target with a small white crosshair cursor that is controlled using a joystick. Only the vertical movement of the cursor can be controlled, the horizontal movement is aligned with the target. Participants are instructed to use the joystick with the right hand to keep the cursor as close as possible to the target. The vertical distance between cursor and target is continuously measured over the whole trial. In ST auditory discrimination trials, ten target sounds and 18 to 20 distractor sounds are presented per trial in a random sequence through headphones: the target sound is a high-pitched tone (1086 Hz), and two low-pitched tones (217 Hz and 652 Hz) are non-target distractor sounds. All sounds are presented for 75 ms with a jittered ISI of 1000 to 1300 ms. Participants are instructed to respond to the high-pitched tone only by depressing the “F12” key with their left index finger, and to react as fast and as accurately as possible. Reaction times and correctness of responses are assessed. In the DT manual tracking and auditory discrimination trials, participants perform both tasks simultaneously. No instructions are given regarding the prioritization of either one of the tasks. Same outcome measures are recorded as in the ST conditions.
Global cognition is assessed using the Digit-Symbol-Substitution test (DSST; 67). This test is part of the Wechsler Adult Intelligence Scale (68) and is performed as paper-and-pencil tests.
A battery of four established motor tests is used to determine different aspects of motor fitness (55, 69) following standardized procedures and instructions. Time is kept using a regular stop watch. Short practice trials with two to five repetitions are performed before each test.
Leg strength and endurance is assessed using the Chair stand test of the senior fitness test for older adults (70). Participants sit on a height-adjustable chair without armrests. Arms are crossed with hands on opposite shoulders. Participants continuously rise up to a straight standing position and sit down to a fully seated position with a straight back as often as possible within 30 s. They are asked to keep their arms crossed and both feet on the floor during the whole test. Correctly executed chair stands are registered.
Bimanual dexterity is measured with the Purdue Pegboard test (71, 72). The Pegboard consists of two rows of 25 small holes from top to bottom. Small metal pins (pegs) are located at the upper left and right of the board. Participants are instructed to simultaneously pick up a peg from the right side with the right hand and a peg from the left side with the left hand, to place both pegs into the top empty holes in the left and right row, and to repeat this procedure as often as possible within 30 s. Three trials are performed. The number of rows with two correctly placed pegs is assessed for each trial.
Static balance is assessed using the One-legged stand test with open and closed eyes (73). The test is performed in the GRAIL (cf. section on virtual reality walking test), but the waistcoat is not used as it could affect participants' posture. Eight trials are performed in total, the first four trials with eyes open, and the second four trials with eyes closed. Each leg is assessed twice, in alternating order. Participants stand on one leg with the other leg slightly flexed while looking straight ahead. The experimenter stands quietly sideways to the participant, to prevent falls. Participants are instructed to keep their arms at the side of their body, to not hop with their standing leg, not to put down their lifted feet, not to push the lifted leg against the standing leg during balancing, and not to open their eyes during eyes closed trials. Each trial is self-initiated. Participants are instructed to stand on one leg as long as possible. The experimenter starts time keeping when the participant lifts one leg, and stops when the participant is violating one of the above mentioned standards or after 20 s. Standing duration and ground reaction forces are assessed.
Psychomotor speed is measured using the Feet tapping test (69). Participants sit on a stationary chair (height adjustable) without armrests, and are asked to move both feet simultaneously back and forth across a mid-sagittal line on the floor. They are instructed to move both feet as fast as possible, while ensuring that both soles completely contact the floor at each tap. Two trials of 20 s duration are performed. The number of correct taps is registered using a hand clicker.
Cardiovascular fitness is measured by a spiroergometry (ZAN600 CPET, nSpire Health, Oberthulba, Germany) on a stationary bicycle (Lode Corival cpet, Groningen, the Netherlands). Participants are asked to avoid caffeine and alcohol intake for 12 hrs before testing and any vigorous exercise on the day before. Each measurement is accompanied by a physician or participants are required to bring a medical clearance certificate based on exercise electrocardiography (ECG) and clinical history. Respiration (oxygen (VO2) and carbon dioxide consumption (VCO2)) is measured breath-by-breath. Heart rate is assessed using an integrated digital twelve-lead electrocardiogram (Kiss, GE Healthcare, Munich, Germany). A Borg’s 6-20 scale (74) is used to assess the rate of perceived exertion (RPE, 6 = “No exertion at all”, 20 = “Maximal exertion”) every two minutes (74) as indicated by the participant by pointing on the number from 6 to 20 on an RPE sheet. Blood pressure is monitored via a sphygmomanometer. Participants undergo a ramp protocol. For male participants, the load starts at 20 W and continuously increases by 20 W/min. For female participants, the load starts at 10 W and increases by 15 W/min. All participants are instructed to maintain a cycling frequency between 60 to 80 revolutions per minute. Both protocols are preceded by a 3 min resting period and followed by 5 min cool-down period (1 min initial load, 4 min no load). Protocols are terminated when participants respiratory exchange ratio (RER = VCO2/VO2) remains > 1.05 for at least 30 s or exceeds 1.10, in case of volitional fatigue, or occurrence of physiological risk factors (i.e., blood pressure > 230/115 mmHg, dizziness, HR > about 220-age, cardiac arrhythmia, or other abnormalities). Each measurement is accompanied by an experienced sport scientist. Peak oxygen uptake (VO2 peak: VO2 consumption during the maximum load level achieved), RER, and the maximum load level (i.e., wattage) are analyzed and considered for rating the measurement validity.
Three different training programs (cognitive, motor, and simultaneous cognitive-motor training) are conducted over a period of twelve weeks in the facilities of the TUC and the WWU. Two training sessions are scheduled per week, for a total of 24 training sessions (total training time: 1080 min). Each training session has a duration of about 60 min, including 15 min for preparation (e.g., changing clothes, warm-up, hard- and software preparation). Each training program is conducted as circuit training with three 15 min blocks yielding a training sequence with a total of 72 training blocks. To ensure continuous training progress, difficulty level of the training is continuously adapted to the individual’s performances. Participants of the motor and the simultaneous cognitive-motor training wear a heart rate monitor during training to ensure that training intensity does not exceed 60% of VO2-peak, as determined by spiroergometry at pretesting. Training sessions are supervised by skilled trainers in group settings with a trainer-participant-ratio of at least 1:3 (motor and simultaneous cognitive-motor training) or at least 1:10 (cognitive training). The trainers provide instructions, help participants to sign into the software applications, answer questions and protocol participants’ performances. They are onboarded to the training procedure in a two-day workshop. To improve attendance and contribution, explanations about the possible benefits of the training are provided to the participants. Attendance and drop-outs are documented. To ensure a total attendance of 24 sessions for each participant, missing training sessions (e.g., in case of illness) are made up within the total training period of twelve weeks. Apart from training, participants are asked to continue their regular everyday activities. Table 3 illustrates exemplary training sessions for each of the three programs.