Advertisement

Task-Specific Rehabilitation of Finger-Hand Function Using Interactive Computer Gaming

      Abstract

      Szturm T, Peters JF, Otto C, Kapadia N, Desai A. Task-specific rehabilitation of finger-hand function using interactive computer gaming.
      The present case study assessed the feasibility of using an interactive gaming system, coupled with the manipulation of common objects, as a form of repetitive, task-specific movement therapy. Three adults with moderate chronic motor impairments of the fingers and hand participated: one 36-year-old man with an incomplete cervical spinal cord injury, one 60-year-old man with a left cortical cerebro-vascular accident, and one 38-year-old woman with left hemiplegic cerebral palsy. Each subject received an intervention of 15 one-hour sessions, which consisted solely of interactive exercise gaming using a diverse range of objects. The objects provided graded and challenging training levels, which emulated the functional properties of objects used in daily life. This in turn produced positive effects on the recovery of active finger range of motion and hand function.

      Key Words

      List of Abbreviations:

      fMRI (functional magnetic resonance (imaging))
      NEUROLOGIC INJURIES frequently result in impairments limiting upper-extremity function.
      • Nakayama H.
      • Jørgensen H.S.
      • Raaschou H.O.
      • Olsen T.S.
      Recovery of upper extremity function in stroke patients: the Copenhagen Stroke Study.
      • O'Mahony P.G.
      • Thomson R.G.
      • Dobson R.
      • Rodgers H.
      • James O.F.
      The prevalence of stroke and associated disability.
      There is significant evidence that targeted physical training improves skills and reduces disability. However, these approaches (eg, constrained use
      • Tarkka I.M.
      • Pitkanen K.
      • Sivenius J.
      Paretic hand rehabilitation with constraint-induced movement therapy after stroke.
      • Wolf S.L.
      • Winstein C.J.
      • Miller J.P.
      • et al.
      Effect of constraint-induced movement therapy on upper extremity function 3 to 9 months after stroke: the EXCITE randomized clinical trial.
      • Liepert J.
      Motor cortex excitability in stroke before and after constraint-induced movement therapy.
      ) typically require extreme dedication and are often repetitive and boring.
      Several studies have provided preliminary descriptions of the benefits of virtual reality and video games in rehabilitation training.
      • Johnson M.J.
      • Feng X.
      • Johnson L.M.
      • Winters J.M.
      Potential of a suite of robot/computer-assisted motivating systems for personalized, home-based, stroke rehabilitation.
      • Colombo R.
      • Pisano F.
      • Mazzone A.
      • et al.
      Design strategies to improve patient motivation during robot-aided rehabilitation.
      • Merians A.S.
      • Poizner H.
      • Boian R.
      • Burdea G.
      • Adamovich S.
      Sensorimotor training in a virtual reality environment: does it improve functional recovery poststroke?.
      • Fasoli S.E.
      • Krebs H.I.
      • Hogan N.
      Robotic technology and stroke rehabilitation: translating research into practice.
      • Lambercy O.
      • Dovat L.
      • Gassert R.
      • Burdet E.
      • Teo C.L.
      • Milner T.
      A haptic knob for rehabilitation of hand function.
      • Broeren J.
      • Rydmark M.
      • Bjorkdahl A.
      • Sunnerhagen K.S.
      Assessment and training in a 3-dimensional virtual environment with haptics: a report on 5 cases of motor rehabilitation in the chronic stage after stroke.
      • Gelfond H.S.
      • Salonius-Pasternak D.E.
      The play's the thing: a clinical-developmental perspective on video games.
      The results of these studies suggest that many elements of the interactive games have a tremendous potential as rehabilitation tools. In their commercial configuration, however, video games and computer systems do not use a diverse range of objects as inputs to the games. Piron et al
      • Piron L.
      • Tonin P.
      • Piccione F.
      • Iaia V.
      • Trivello E.
      • Dam M.
      Virtual environment training therapy for arm motor rehabilitation.
      developed a virtual reality training system that uses multi-axial motion sensors to instrument a small sample of common objects (eg, an envelope used to complete a virtual mailing task). To extend this approach, we developed a custom interface device that uses a miniBIRD or microBIRD
      MiniBIRD/microBIRD; Ascension Technology, 107 Catamount Dr, Milton, VT 05468.
      Matlab; The MathWorks, 3 Apple Hill Dr, Natick, MA 01760-2098.
      miniature motion tracking sensor to transform nearly any object into a computer mouse or joystick. Therefore, our approach allows participants to play any commercial video game, by slaving the video game sprite to the object motion. For a full description of the system, see Otto's article.
      • Otto C.
      Magnetic motion tracking system for interactive gaming [dissertation].
      The system reported in this paper was designed to use any commercial game and this is important for many reasons. It is important to identify games that suit individual preferences; the wide range of activities available in commercial games makes this possible. It is important to have a large and ever changing variety of inexpensive games to maintain high levels of motivation and interest. Commercially available games do require a wide range of levels of precision, and movements that vary in speed, amplitude, and direction. By selecting objects used in activities of daily living, the therapy can target specific tasks. Hence, the games can be graded to match a client's current level of functioning. Finally, many commercially available games involve multi-tasking. Hence, the tasks also engage key attentional, perceptual, and cognitive skills.
      • Green C.S.
      • Bavelier D.
      Action video game modifies visual selective attention.
      • Green C.S.
      • Bavelier D.
      Enumeration versus multiple object tracking: the case of action video game players.
      • Akhutina T.
      • Foreman N.
      • Krichevets A.
      • et al.
      Improving spatial functioning in children with cerebral palsy using computerized and traditional game tasks.
      In this study, we evaluate the feasibility of our interactive gaming system that couples object manipulation with exercise. Through a variety of common objects, tools, and utensils, our system provides a basis for repetitive, task-specific therapy focused on finger and hand function.

      Methods

      Three case studies were conducted, and pre- and post-exercise changes in outcome measures were determined. We recruited the participants from an outpatient physical therapy clinic for upper-extremity re-education. Each volunteer gave informed consent. Ethics approval was obtained from the University of Manitoba Human Research Ethics Board.
      The participant in Case 1 was a right-handed, 32-year-old man who suffered an incomplete spinal cord injury (C6–C7) 15 months before the start of the study. Clinical assessment revealed: (1) the subject was dependent on tenodesis grip for grasping functions, (2) wrist, elbow, and shoulder movements were intact, (3) sensation was intact, (4) resting posture of the hands was finger flexion with moderate levels of finger flexor spasticity, (5) reduced active finger extension was limited by 45° from neutral, and (6) the subject had considerable difficulty with opening the hand for grasping and manipulating objects.
      The participant in Case 2 was a right-handed, 60-year-old man who suffered a left ischemic stroke involving middle cerebral artery 18 months before the start of the study. Clinical assessment revealed: (1) mild aphasia, (2) motor apraxia, (3) score of 5 on the hand and arm component of the Chedoke McMaster Stroke Assessment,
      • Barreca S.R.
      • Stratford P.W.
      • Lambert C.L.
      • Masters L.M.
      • Streiner D.L.
      Test-retest reliability, validity, and sensitivity of the Chedoke arm and hand activity inventory: a new measure of upper-limb function for survivors of stroke.
      and (4) moderate limitations and difficulty in handling and coordinating the use of most common household utensils, food, and drinks.
      The participant in Case 3 was a right-handed, 38-year-old woman with right-sided hemiplegic cerebral palsy. Clinical assessment revealed that the subject: (1) had decreased stereognosis of the right fingers, (2) typically would not use the right hand and had poor manual dexterity and difficulty handling and manipulating most common daily used objects, (3) was unable to rotate objects (eg, to turn a door knob or the top off a bottle), and (4) exhibited a number of visuospatial deficits including left-right confusion.

      Treatment Program

      Figure 1 shows a sample of the objects used during the interactive exercise gaming program. The program for subject 1 consisted of the manipulation of objects that required active finger extension (with wrist in neutral to preserve tenodesis). Examples included use of fingers to roll cylindrical and cone-shaped objects, and sports balls with different diameters and surface properties. The program for subjects 2 and 3 included objects varying in size, shape, and surface properties, manipulated with both hands, whole hand, progressing to 3- and 2-finger functions. Tasks were selected that required fine finger controls with graded precision.
      Figure thumbnail gr1
      Fig 1Sample of different therapeutic objects instrumented with the motion sensor and used in the interactive exercise gaming program. The tethered punching bag, different sized sports balls, and cylinders supported the weight of the arm while being moved. Different shapes and sizes of cups, glasses, bottles, and jugs were used and sometimes partially filled with fluids. Also used were objects of different surface properties (rough, sticky, slippery, or compliant).
      Figure 2 illustrates set-up of the interactive gaming system. Each exercise object is instrumented with the miniBird motion sensor. When the sports ball was rotated forward, the game sprite moved to the top of the monitor (bottom-left panel); when the ball was rotated to the right, the game sprite moved to the right edge of monitor (bottom-right panel). A short video is available that demonstrates the interactive exercise gaming system.
      Figure thumbnail gr2
      Fig 2Components and set-up for the interactive exercise gaming system. The miniature motion sensor is attached to each gaming object (front of Lego car; top right of sports ball) and connected (white cable) to the miniBird interface box situated to right of the laptop. MiniBird (output) connects to interface device (input), interface device (output) connects to the laptop through a universal serial bus cable (game controller). Rectangular box on stool situated to right of the object is the miniBird magnet (transmitter). A larger display monitor is connected to laptop for ease of viewing. Top 2 panels are full view, and bottom 2 panels are close-ups of top right panel with subject using sports ball as a multi-axial game controller slaved to computer game sprite.
      Participants could select from 25 different commercial video games (rated “everyone”), each offering different precision levels, speed constraints, visuospatial demands, and either single or dual axis movements. Participants received 15 one-hour therapy sessions, 3 sessions a week. Training consisted solely of the interactive exercise gaming; no other physiotherapy was received.
      Configurable difficulty level parameters were included in the system, which allowed participants to remain competitive by matching game play to the abilities of the participant. For example, the sensitivity of the physical object motion range mapping to the game sprite range could be adjusted. Most computer games have adjustable skill levels from novice to expert. Ergonomic properties of different objects also affect task difficulty.

      Outcome Measures

      We performed the following assessments pre- and post-treatment: Jebsen-Taylor Hand Function Test,
      • Jebsen R.H.
      • Taylor N.
      • Trieschmann R.B.
      • Trotter M.J.
      • Howard L.A.
      An objective and standardized test of hand function.
      passive and active range of motion at the metacarpal-phalangeal joint, and the Chedoke-McMaster Stroke Assessment
      • Barreca S.R.
      • Stratford P.W.
      • Lambert C.L.
      • Masters L.M.
      • Streiner D.L.
      Test-retest reliability, validity, and sensitivity of the Chedoke arm and hand activity inventory: a new measure of upper-limb function for survivors of stroke.
      (subject 2 only).
      A computerized visual-tracking task was used to quantify fine motor manipulation skills.
      • Carey J.R.
      • Greer K.R.
      • Grunewald T.K.
      • et al.
      Primary motor area activation during precision-demanding versus simple finger movement.
      • Yamanaka H.
      • Kawahira K.
      • Arima M.
      • et al.
      Evaluation of skilled arm movements in patients with stroke using a computerized motor-skill analyser for the arm.
      A custom software program was created to control the motion of an on-screen cursor (large bright-colored square). The cursor moved in a predictable sinusoidal manner either horizontally (left to right) or vertically (top to bottom). Participants were instructed to move a test object in concert with the moving target cursor: 9 cursor movement cycles appeared. The miniBird motion sensor was attached to each object at a specified location. The position coordinates of both the target cursor and object motion were synchronously recorded at a sampling rate of 100Hz. The test objects selected for each participant corresponded to his/her motor impairment, functional ability, and personal needs. For subject 1, a light-weight piece of styrofoam (10×5×2.5cm) was used to evaluate hand function. The styrofoam was grasped between the 4 fingers and the thumb. While maintaining his grasp of the styrofoam, subject 1 was instructed to flex and extend his wrist in concert with a vertically moving target (subject 1, task 1); and supinate and pronate his forearm in concert with a horizontally moving target (subject 1, task 2). In both tasks, his elbow was flexed at 90°, his forearm was supported, and external-internal shoulder rotation was restricted. For subject 2 and subject 3, the following 2 tasks were used. The first task used a thin piece of Plexiglas (6 × 12cm), which was grasped between the 4 fingers and the thumb. While maintaining grasp, subject 2 and subject 3 were instructed to flex and extend their wrist in concert with a vertically moving target (subject 2, task 1; subject 3, task 1). The second task involved a pencil held between the thumb, index, and middle fingers, which the subjects used to push/pull a 4-wheeled Lego car forwards/backwards in concert with a vertically moving target (subject 2, task 2; subject 3, task 2). These test objects were not used during the regular therapy sessions. A cross-correlation analysis was used to quantify the similarity between the movement trajectories of the target (reference) and manipulated test object cursors. The peak correlation coefficient represented a global performance index.
      We constructed a questionnaire to obtain feedback about the client's perception of the exercise gaming program, including the following questions: (1) Were the video game exercises fun to play? (2) Did the video games increase your motivation to perform your exercises? (3) Were the video game exercises challenging? (4) Did the video game's difficulty level enhance the exercises? (5) Did you prefer interactive video game exercise to traditional physiotherapy received in the past? Each question was ranked: (1) I strongly agree; (2) I agree; (3) neutral; (4) I disagree; and (5) I strongly disagree.
      Assessments were performed by an independent physiotherapist, who was not informed of the treatment program.

      Results

      Figure 3 presents the time taken to complete the Jebsen-Taylor Hand Function Test. Posttreatment, all subjects exhibited substantial decreases in the total time. Peak correlation coefficient results from the tracking tasks are shown in figure 4. There was a substantial increase in peak correlation coefficient posttreatment for all subjects.
      Figure thumbnail gr3
      Fig 3Pre- and posttreatment Jebsen-Taylor Hand Function Test total test scores for each subject.
      Figure thumbnail gr4
      Fig 4Pre- and posttreatment peak cross correlation coefficients (r value) for task 1 (top histogram) and task 2 (bottom histogram) for each subject.
      Example target cursor and object motion trajectories recorded for one tracking task are presented in figure 5. Pretreatment, subject 2 was only able to produce 4 of the 9 tracking cycles; compared with the reference waveform, the amplitudes of his movements were relatively small. Posttreatment, subject 2 was able to produce movements for all tracking cycles; his movements were more consistent and of much greater amplitude. In addition, his posttreatment Chedoke-McMaster hand score improved from 5 to 7. Similar results were obtained for subject 3 (tested with the same object). Pretreatment, only 5 of the tracking cycles were evident; posttreatment, she exhibited a marked improvement in movement amplitude and all movement cycles were produced. For subject 1, movement cycles were evident for all tracking cycles. However a 30% increase in the extension amplitude was seen for subject 1, task 1, and a 25% increase in the supination amplitude was seen for subject 1, task 2. Posttreatment, subject 1 increased his active metacarpal-phalangeal extension by 35°, was able to fully extend his fingers, and to grasp most objects.
      Figure thumbnail gr5
      Fig 5Presented for subject 2 are movement trajectories of the test object pretreatment (bottom left plot) and posttreatment (bottom right) during task 1; using an implement to push/pull Lego car. Top plots are trajectories of reference target motion. Bottom plots, linear translation in primary axis of object motion; in this case up is forward (pushing) car motion and downward is back (pulling) car motion.
      The questionnaire results were positive, with subjects answering “strongly agree” to all 5 questions. All participants expressed that they enjoyed the video game-based tool, preferring it to exercise programs, and would like to continue the treatment.

      Discussion

      The present findings show that the interactive exercise program produced positive effects on the recovery of finger motion and hand function. These findings are consistent with studies that have evaluated efficacy of constraint-induced movement therapy.
      • Tarkka I.M.
      • Pitkanen K.
      • Sivenius J.
      Paretic hand rehabilitation with constraint-induced movement therapy after stroke.
      • Wolf S.L.
      • Winstein C.J.
      • Miller J.P.
      • et al.
      Effect of constraint-induced movement therapy on upper extremity function 3 to 9 months after stroke: the EXCITE randomized clinical trial.
      • Liepert J.
      Motor cortex excitability in stroke before and after constraint-induced movement therapy.
      The configurable system parameters, (eg, speed and level of accuracy) permitted the exercises to be challenging. In turn, this allowed the participants to be competitive and enjoy the games.
      Specificity of training is an important component in therapy regimens and this is dependent on the quality and quantity of feedback (finger-object contact and slippage) required for handling and manipulation.
      • Zatsiorsky V.M.
      • Gao F.
      • Latash M.L.
      Motor control goes beyond physics: deferential effects of gravity and inertia on finger forces during manipulation of hand-held objects.
      • Winges S.A.
      • Soechting J.F.
      • Flanders M.
      Multidigit control of contact forces during transport of handheld objects.
      • Flanagan J.R.
      • Bowman M.C.
      • Johansson R.S.
      Control strategies in object manipulation tasks.
      A recent study by Carey et al
      • Carey J.R.
      • Greer K.R.
      • Grunewald T.K.
      • et al.
      Primary motor area activation during precision-demanding versus simple finger movement.
      investigated fMRI signals during 2 rhythmic finger flexion-extension tasks. In the first task, subjects had to track a moving cursor; in the second task, the movements were free form. Their findings showed that during the tracking tasks, there was significantly greater fMRI signal intensity in both the contra-lateral and ipsilateral primary motor cortex. These findings support the view that precision, goal-directed movements are a preferred method to promote neuro-adaptation during the recovery from central nervous system disorders. This is consistent with the present therapy approach, which incorporates task dynamics and precision movements during object handling and manipulation.
      Learning to restore function of fine and gross manual skills requires movement error reduction by the timely processing of tactile and proprioceptive sensory feedback.
      • Quaney B.M.
      • Cole K.J.
      Distributing vertical forces between the digits during gripping and lifting: the effects of rotating the hand versus rotating the object.
      • Scheidt R.A.
      • Dingwell J.B.
      • Mussa-Ivaldi F.A.
      Learning to move amid uncertainty.
      Task variability during practice is an important factor and studies have shown that introducing variability (random practice) improves performance in subsequent sessions (retention) and is more effective than blocked repetition of a constant task.
      • Emken J.L.
      • Reinkensmeyer D.J.
      Robot-enhanced motor learning: accelerating internal model formation during locomotion by transient dynamic amplification.
      • Patton J.
      • Mussa-Ivaldi F.A.
      Robot-assisted adaptive training: custom force fields for teaching movement reaching patterns.
      The present interactive exercise approach uses video games that are typically random tasks with a wide range of adjustable movement amplitudes, speeds, and precision levels. In addition a diverse range of objects can be used, depending on the client's needs and requirements. With this individualized approach, one could maximize the rehabilitative effectiveness of training.
      There is a need for goal-directed therapeutic programs that integrate sensory feedback and motor actions with visuospatial and other cognitive functions. In this regard, video games offer important values that include: facilitation of attention and focus, graded requirements of gaze control, and multi-tasking.
      • Green C.S.
      • Bavelier D.
      Action video game modifies visual selective attention.
      • Akhutina T.
      • Foreman N.
      • Krichevets A.
      • et al.
      Improving spatial functioning in children with cerebral palsy using computerized and traditional game tasks.
      After returning home or to other community sites, many people who have had a stroke, traumatic brain injury, or spinal cord injury will still require and benefit from rehabilitation. Home rehabilitation programs can be effective and comparable to institutional programs.
      Early Supported Discharge Trialists
      Services for reducing duration of hospital care for acute stroke patients.
      Although not the focus of this study, the interactive exercise gaming system was designed to work as a monitored home program with the future goal of establishing a self-sustained independent rehabilitation program, and thus providing extended practice in the client's home or in rural and remote settings.

      Conclusions

      An interactive exercise gaming system has been developed for hand function rehabilitation that involved the manipulation of real objects. The proposed intervention offers a wide range of graded and challenging training tasks that take into account the functional properties of commonly used objects. The present findings show that our therapy approach produced positive effects on the recovery of finger motion and hand function.
      Suppliers
      aMiniBIRD/microBIRD; Ascension Technology, 107 Catamount Dr, Milton, VT 05468.
      bMatlab; The MathWorks, 3 Apple Hill Dr, Natick, MA 01760-2098.

      References

        • Nakayama H.
        • Jørgensen H.S.
        • Raaschou H.O.
        • Olsen T.S.
        Recovery of upper extremity function in stroke patients: the Copenhagen Stroke Study.
        Arch Phys Med Rehabil. 1994; 75: 394-398
        • O'Mahony P.G.
        • Thomson R.G.
        • Dobson R.
        • Rodgers H.
        • James O.F.
        The prevalence of stroke and associated disability.
        J Public Health Med. 1999; 21: 166-171
        • Tarkka I.M.
        • Pitkanen K.
        • Sivenius J.
        Paretic hand rehabilitation with constraint-induced movement therapy after stroke.
        Am J Phys Med Rehabil. 2005; 84: 501-505
        • Wolf S.L.
        • Winstein C.J.
        • Miller J.P.
        • et al.
        Effect of constraint-induced movement therapy on upper extremity function 3 to 9 months after stroke: the EXCITE randomized clinical trial.
        JAMA. 2006; 296: 2095-2104
        • Liepert J.
        Motor cortex excitability in stroke before and after constraint-induced movement therapy.
        Cogn Behav Neurol. 2006; 19: 41-47
        • Johnson M.J.
        • Feng X.
        • Johnson L.M.
        • Winters J.M.
        Potential of a suite of robot/computer-assisted motivating systems for personalized, home-based, stroke rehabilitation.
        J NeuroEng Rehabil. 2007; 4: 6
        • Colombo R.
        • Pisano F.
        • Mazzone A.
        • et al.
        Design strategies to improve patient motivation during robot-aided rehabilitation.
        J NeuroEng Rehabil. 2007; 4: 3
        • Merians A.S.
        • Poizner H.
        • Boian R.
        • Burdea G.
        • Adamovich S.
        Sensorimotor training in a virtual reality environment: does it improve functional recovery poststroke?.
        Neurorehabil Neural Repair. 2006; 20: 252-256
        • Fasoli S.E.
        • Krebs H.I.
        • Hogan N.
        Robotic technology and stroke rehabilitation: translating research into practice.
        Top Stroke Rehab. 2004; 11: 11-19
        • Lambercy O.
        • Dovat L.
        • Gassert R.
        • Burdet E.
        • Teo C.L.
        • Milner T.
        A haptic knob for rehabilitation of hand function.
        IEEE Trans Neural Syst Rehabil Eng. 2007; 15: 356-366
        • Broeren J.
        • Rydmark M.
        • Bjorkdahl A.
        • Sunnerhagen K.S.
        Assessment and training in a 3-dimensional virtual environment with haptics: a report on 5 cases of motor rehabilitation in the chronic stage after stroke.
        Neurorehabil Neural Repair. 2007; 21: 180-189
        • Gelfond H.S.
        • Salonius-Pasternak D.E.
        The play's the thing: a clinical-developmental perspective on video games.
        Child Adolesc Psychiatr Clin N Am. 2005; 14 (ix): 491-508
        • Piron L.
        • Tonin P.
        • Piccione F.
        • Iaia V.
        • Trivello E.
        • Dam M.
        Virtual environment training therapy for arm motor rehabilitation.
        Presence. 2005; 14: 732-740
        • Otto C.
        Magnetic motion tracking system for interactive gaming [dissertation].
        Univ of Manitoba, Winnipeg2007
        • Green C.S.
        • Bavelier D.
        Action video game modifies visual selective attention.
        Nature. 2003; 423: 534-537
        • Green C.S.
        • Bavelier D.
        Enumeration versus multiple object tracking: the case of action video game players.
        Cognition. 2006; 101: 217-245
        • Akhutina T.
        • Foreman N.
        • Krichevets A.
        • et al.
        Improving spatial functioning in children with cerebral palsy using computerized and traditional game tasks.
        Disabil Rehabil. 2003; 25: 1361-1371
        • Barreca S.R.
        • Stratford P.W.
        • Lambert C.L.
        • Masters L.M.
        • Streiner D.L.
        Test-retest reliability, validity, and sensitivity of the Chedoke arm and hand activity inventory: a new measure of upper-limb function for survivors of stroke.
        Arch Phys Med Rehabil. 2005; 86: 1616-1622
      1. Interactive exercise gaming.
        (Accessed August 14, 2008)
        • Jebsen R.H.
        • Taylor N.
        • Trieschmann R.B.
        • Trotter M.J.
        • Howard L.A.
        An objective and standardized test of hand function.
        Arch Phys Med Rehabil. 1969; 50: 311-319
        • Carey J.R.
        • Greer K.R.
        • Grunewald T.K.
        • et al.
        Primary motor area activation during precision-demanding versus simple finger movement.
        Neurorehabil Neural Repair. 2006; 20: 361-370
        • Yamanaka H.
        • Kawahira K.
        • Arima M.
        • et al.
        Evaluation of skilled arm movements in patients with stroke using a computerized motor-skill analyser for the arm.
        Int J Rehabil Res. 2005; 28: 277-283
        • Zatsiorsky V.M.
        • Gao F.
        • Latash M.L.
        Motor control goes beyond physics: deferential effects of gravity and inertia on finger forces during manipulation of hand-held objects.
        Exp Brain Res. 2005; 162: 300-308
        • Winges S.A.
        • Soechting J.F.
        • Flanders M.
        Multidigit control of contact forces during transport of handheld objects.
        Neurophysiol. 2007; 98: 851-860
        • Flanagan J.R.
        • Bowman M.C.
        • Johansson R.S.
        Control strategies in object manipulation tasks.
        Curr Opin Neurobiol. 2006; 16: 650-659
        • Quaney B.M.
        • Cole K.J.
        Distributing vertical forces between the digits during gripping and lifting: the effects of rotating the hand versus rotating the object.
        Exp Brain Res. 2004; 155: 145-155
        • Scheidt R.A.
        • Dingwell J.B.
        • Mussa-Ivaldi F.A.
        Learning to move amid uncertainty.
        J Neurophysiol. 2001; 86: 971-985
        • Emken J.L.
        • Reinkensmeyer D.J.
        Robot-enhanced motor learning: accelerating internal model formation during locomotion by transient dynamic amplification.
        IEEE Trans Neural Syst Rehabil Eng. 2005; 13: 33-39
        • Patton J.
        • Mussa-Ivaldi F.A.
        Robot-assisted adaptive training: custom force fields for teaching movement reaching patterns.
        IEEE Trans Biomed Eng. 2004; 51: 636-646
        • Early Supported Discharge Trialists
        Services for reducing duration of hospital care for acute stroke patients.
        Cochrane Database Syst Rev. 2002; (CD000443)